© Universidad Politécnica de Madrid
Escuela Técnica Superior de Ingenieros Industriales
Subdirección de Investigación, Doctorado y Relaciones con Empresas
C/José Gutiérrez Abascal 2
28006 Madrid









ISBN: 978-84-16397-58-7


CONTENS


Foreword of the Director 4
Organizing Committee 5
Technical program of IRM17
Conference at a glance 6
Research works and patents 8
Research Groups, Centers and Institutes 18
Companies at IRM17 20
Closing ceremony & awards 22
Posters of research works
Acoustics Engineering 23
Aerospace 27
Automation 31
Bioengineering & Health 42
Chemical Engineering 50
Electronics 56
Electrical Engineering & Machines 69
Energy 72
Environmental Engineering 80
Laser Technology 84
Materials 90
Management and Industrial Organization 100
Mechanics, Manufacturing & Structures 119
Mathematics & Statistics 131
Nuclear Technology 134
Seismic Engineering 140
Vehicles & Railways 144
Posters of patents 154
UPM Institutional posters 178
Posters of Research Groups, Centers and Institutes 183





FOREWORD OF THE DIRECTOR

Dear participants,

Industriales Research Meeting (IRM17) in its second edition, is a reference for our doctoral
students, professors, non-academic staff, and the rest of students of the ETS Ingenieros
Industriales of the Universidad Politécnica de Madrid, in which the research activity in the School
is shown to all the community.

This year the meeting has been organized in two days. One for the doctoral students of the
Institutes, Research Centres, Research Groups and Departments of this School, and the second day
is for networking between our researches and the large number of engineering companies in the
industrial sector, that will present their research work.

This meeting has the full support of our University and the General Direction of the Research and
Universities of the Madrid Community, and the SAE (Sociedad Amigos de la Escuela). It is also a
privilege the presence in a Plenary Session of Prof. Avelino Corma, Príncipe de Asturias Award in
the Scientifical and Technical Research 2014, to whom I appreciate the acceptance of participating
in the meeting.

Finally, I thank to the Organizing Committee for the fantastic enthusiasm to organizing this second
edition.

Our best wishes to all participants


Emilio Mínguez Torres
Director

ORGANIZING COMMITTEE

Óscar García Suárez
Nuria Martínez Carpintero
Lucía Alvaredo Olmos
Eduardo Caro Huertas
Ismael Díaz Moreno
Sara Lauzurica Santiago
Ignacio Pavón García
Antonio Rivera de Mena

Subdirección de Investigación, Doctorado y Relaciones con Empresas
Escuela Técnica Superior de Ingenieros Industriales
Universidad Politécnica de Madrid

Last year we launched “Industrials Research Meeting” (IRM) as a new experience in our School.
We really did not know the real impact it might have. Considering the number of participants and
attendees we considered it as a success. After it, we received a lot of positive feedback and new
ideas to consider. Therefore, we are back again with a new edition IRM17 trying to cover new
things and with a more extended program.

In the first day of this edition, we will have research works and projects in both oral and poster
sessions. We have included an area of patents to show the industry our capability of innovation.
During the second day, our research groups, centres and institutes will share the Sala de la
Máquina with a nice representation of companies that will show us their main research lines and
projects. We believe this is a great opportunity to know each other and stablish good relations that
could make easier future collaborations.

I want to highlight the conference of our main speaker, D. Avelino Corma, awarded with the
Príncipe de Asturias de Investigación 2014 award. Thank you very much for accepting our
invitation. Also we have selected the research works of Sara Lauzurica and Alberto Abánades to be
presented in an oral format.

I would like to thank many people for their contribution to IRM17: all the authors and speakers for
their contributions. This is extended to the Research Groups, Centers and Institutes of the School
for their support. Also, the staff of ETSII that is always willing to help.

I am very grateful to our Director Emilio Mínguez and the rest of the management team of ETSII-
UPM. I cannot imagine a better environment to work.

And finally, special thanks to my IRM17 team: Nuria, Lucía, Sara, Eduardo, Ismael, Ignacio and
Antonio. You have made an excellent work and I enjoyed a lot during our meetings. It is very easy
to run this initiative with people like you.

Óscar García Suárez
Subdirector de Investigación, Doctorado y Relaciones con Empresas

CONFERENCE AT A GLANCE

April 4th, 2017
RESEARCH WORKS AND PATENTS

Registration 8:30

Opening Session
Asunción de María Gómez, Vice - Rector (UPM)
Emilio Mínguez, Director (ETSII - UPM) 9:00 - 9:20
Óscar García, Vice - Director (ETSII - UPM)
(Salón de Actos)

Keynote Presentations
by Professors of the ETSII - UPM 9:20 - 10:00
(Salón de Actos)

Oral Session
by Doctoral Students of the ETSII - UPM 10:00 - 12:00
(Aula C / Aula D)

Poster Session:
Research Projects and Patents 12:00 - 14:30
(Sala de la Máquina)

The April 4th day will be dedicated to the presentation of research works carried
out by PhD students and Professors/Researchers of our School. Additionally, the
existing patents will be presented in this session.

Cover designed by: Jesús Tordesillas Torres
CONFERENCE AT A GLANCE

April 5th, 2017
RESEARCH GROUPS AND COMPANIES
Welcome
Guillermo Cisneros
Rector of the Universidad Politécnica de Madrid
José Manuel Torralba 9:00 - 9:30
Director Gral. Universidades e Investigación de la Comunidad de Madrid
Emilio Mínguez
Director (ETSII - UPM)
(Salón de Actos)
Plenary Session
Avelino Corma 9:30 - 10:30
Premio Príncipe de Asturias de Investigación Científica y Tecnológica 2014
(Salón de Actos)
Keynote Presentations
by Professors of the ETSII - UPM 10:30 - 11:30
(Salón de Actos)
Poster Session:
Research Groups and Companies 11:30 - 14:00
(Sala de la Máquina)
Awards Ceremony & Closing 14:00 - 14:30
(Salón de Actos)

The April 5th day will be dedicated to the Research Groups, Centres and Institutes
of the ETSII that will show their main lines. In the same way, partner companies
will show their research works in which they are involved.
SCIENTIFIC AGENDA

TUESDAY, April 4th 2017
8:30 REGISTRATION
9:00 - 9:20 OPENING SESSION (Salón de Actos)
Asunción de María Gómez Pérez, Vice - Rector for Research, Innovation
and Doctorate
Emilio Mínguez Torres, Director of the ET SII - UPM
Óscar García Suárez, V ice - Director for Research, Doctorate and
Relations with Companies of the ETSII - UPM

9:20 - 10:00 KEYNOTE PRESENTATIONS (Salón de Actos)
Gonzalo Jiménez Varas, A ssistant Professor ET SII (UPM )
La belleza de descubrir: un viaje desde Mileto hasta el mundo post-Fukushima

Francisco Javier Cara Cañas, A ssistant Professor ET SII (UPM )
Aportaciones estadísticas al análisis modal de edificios y puentes

10:00 - 12:00 ORAL SESSION
SESSION 1 (Aula C)
Chairmen Ignacio Pavón, Joaquín Martínez
10:00 Respuesta de los R-NPs en seco o inmersos en agua. Ana López Hernández
10:15 Elongación de nanopartículas metálicas embebidas en un material dieléctrico
producidas por irradiación con iones pesados altamente energéticos. Alejandro
Prada Valverde
10:30 Análisis de los efectos de las rupturas estructurales y de la volatilidad
concentrada en los precios del gas de importación en España. Pablo Ángel
Cansado Bravo
10:45 Diseño y estudio del comportamiento de estructuras basadas en Metamateriales
Geomiméticos Inspirados en las Redes De Bravais. Graciela Fernández Méjica
11:00 Visualización de datos en acústica ambiental como forma de disminuir las
molestias por ruido. Luis Gascó Sánchez
11:15 Evaluación del reciclado mecánico de PLA y sus nanocompuestos dedicados a
aplicaciones de envasado. Freddys Beltrán González
11:30 Definición de procesos de metalurgia secundaria basados en tratamientos al
calcio como herramienta para la mejora del mecanizado a alta velocidad de los
aceros de colada continua calmados con aluminio . Izaskun Alonso Oña
It will be presented an award to the best presentaƟon of this session
SCIENTIFIC AGENDA

SESSION 2 (Aula D)
Chairmen Sara Lauzurica, Claudio Rossi
10:00 Seguimiento e interceptación de peatones: Una aplicación de la robótica móvil a la
vigilancia de las infraestructuras críticas. Mario Andrei Garzón Oviedo
10:15 Una arquitectura general para la navegación autónoma de sistemas aéreos no
tripulados. José Luis Sánchez López
10:30 Inversor monofásico para la competición “"the little box challenge" . Regina Ramos
10:45 Interfaz de control y coordinación multi-robot. Juan Jesús Roldán Gómez
11:00 Análisis de accidentes en contenciones nucleares bajo una nueva perspectiva.
Kevin Fernández-Cosials
11:15 Robot actuado por cables tirantes para simulación de condiciones subacuáticas
con aplicación en robots humanoides submarinos. Alejandro Rodríguez Barroso
11:30 Remote Drive: Un UGV militar para misiones de vigilancia. Miguel Clavijo
Jiménez
It will be presented an award to the best presentation of this session

12:00 - 14:30 POSTER SESSION (Sala de la Máquina)
The layout of the Sala de la Máquina with the poster distribution will be shown in the final version
of the program.

ACOUSTICS ENGINEERING
[ACO-01] Noise data communication as a non-acoustic factor for mitigating aviation noise. Gascó, Luis;
Asensio, César; de Arcas, Guillermo
[ACO-02] Validation of therapies for Tinnitus treatament. Enriched Acoustic Enviroment (EAE).
Cobo, Pedro; Garzón, Christiam
[ACO-03] Acoustic stimulation for Parkinson's Disease: a longitudinal pilot study. Gálvez, Gerardo;
González, David; Recuero, Manuel; de Arcas, Guillermo

AEROSPACE
[AER-01] Multi-UAV coordination and control interface. Roldán, Juan Jesús
[AER-02] Low Power Distribution Module for Space Application: Analysis and Comparison of Different
Architectures and DC/DC Topologies. Stevanović, Branislav; Salinas, Guillermo; Alou, Pedro;
Oliver, Jesús Angel; Vasić, Miroslav
[AER-03] Real time operating system with reconfiguration support for adaptable satellite operations. Pérez,
Arturo; Suriano, Leonardo; de la Torre, Eduardo

AUTOMATION
[AUT-01] A flight altitude estimator for multirotor UAVs in dynamic and unstructured indoor environments.
Bavle, Hriday; Sánchez-López, José Luis; Rodríguez-Ramo, Alejandro; Sampedro, Carlos;
Campoy, Pascual
[AUT-02] A general architecture for autonomous navigation of unmanned aerial systems. Sánchez-López,
José Luis; Molina, Martín; Campoy, Pascual
SCIENTIFIC AGENDA
[AUT-03] Computer Vision for Sense & Avoid Onboard Unmanned Aerial Vehicles. Carrio, Adrian;
Campoy, Pascual
[AUT-04] CHEMOCONTROL. Rossi, Claudio; Barrientos, Antonio; Del Cerro, Jaime; Marrazza, GIovan
na; Ravalli, Andrea
[AUT-05] UNEXMIN - underwater explorer for flooded mines. Claudio Rossi; Sergio Domínguez; Ramón
Suárez; Zorana Milosevic
[AUT-06] Tracking and intercepting pedestrians: a robotic approach to surveillance of critical
infrastructures. Garzón, Mario
[AUT-07] Cable - driven robot to simulate underwater environments. Rodríguez, Alejandro; Saltarén,
Roque; Aracil, Rafael; Portilla, Gerardo A; Cely, Juan S; García, Cecilia; Carpio, Marco
[AUT-08] Force muscle stimation based on mechanomyography sensors. Samper, José L.; Sánchez-Urán,
Miguel A.; Ferre, Manuel; Cordoba, Daniel; García, Ángel; Ruiz, Ricardo; Garnacho,
Fernando
[AUT-09] Multimodal Vest for Virtual Reality Applications. Ferre, Manuel; García, Gonzalo; Breñosa, José
M.; Vargas, David.
[AUT-10] Advanced Telerobotic System for Scientific Nuclear Facilities. Ferre, Manuel; Breñosa, José M.;
Coloma, Sofia; Rubio, Luis E.

BIOENGINEERING AND HEALTH
[BIO-01] Comparison of the biosensing performance of Resonant Nanopillars arrays under dry conditions
or immersed in fluid. Hernández, Ana L. ; Casquel, Rafael; Holgado, Miguel; Cornago, Iñaki;
Fernández, Fátima ; Ciaurriz, Paula; Sanza, Francisco J; Santamaría, Beatriz ; Maigler,
María V.; Laguna, María Fe
[BIO-02] Fabrication and characterization of a label-free optical biosensor based on SiO2 resonant
microstructures. Quintero, Sergio A.; Sanza, Francisco J.; Holgado, Miguel; López-Romero,
David; Laguna, Maria F.; López, Ana; Casquel, Rafael; López, Rocio; Lavin, Alvaro;
Santamaría, Beatriz.
[BIO-03] Developing optical label-free Point-of-Care biosensing devices. Maigler, M. Victoria; Holgado,
Miguel; Lavín, Álvaro; Vicente, Manuel; Domingo, Martín; Sanza, F. Javier; Laguna, M. Fe;
Santamaría, Beatriz; López, Rocío; López-Romero, David
[BIO-04] Detection of CST4, MMP9 and S100A6 biomarkers for dry eye disease on an integrated optical
label-free biosensor. Santamaria, Beatriz; Hernéndez, Ana López; Maigler, María Victoria,
Ramirez, Yolanda; Sanza, Francisco Javier; Casquel, Rafael; Lavín, Álvaro; Laguna; María
Fe; Holgado, Miguel
[BIO-05] Biochemical assay development for optimization of immunological system inhibitors. López ,
Rocío; Ramírez, Yolanda; Maigler, María Victoria; López Hernández, Ana; Sanza,
Francisco Javier; López-Romero, David; Casquel, Rafael; Lavín, Álvaro; Laguna, Mª Fé;
Holgado, Miguel
[BIO-06] Local blood pressure estimations: The dependence of the subharmonic resonance on ambient
pressure. Jiménez Fernández, Javier
[BIO-07] Design of an Aspiration Thrombectomy Device for cerebral vessels. Talayero, Carlos; Romero,
Gregorio; Pearce, Gillian

CHEMICAL ENGINEERING
[CHE-01] Maceration process of fruits in liquor: Spectroscopic studies of the kinetics. Paz, Isabel; Matías,
Carmen; Fernández, Ascensión; Pinto, Gabriel
[CHE-02] Science, Technology and Art. Pinto, Gabriel; Alonso, José Vicente; Díaz, Francisco; Díaz,
Víctor Manuel
[CHE-03] Developing STEM educative and outreach tools. Pinto, Gabriel; Prolongo, Marisa
[CHE-04] Degradation of poly(lactic acid) in different aqueous solutions. Beltrán, Freddys; Alarcón, Ana;
Gordillo, Rodrigo; Cañete, Elena; Perona, Elia; de la Orden, Mª Ulagares; Lorenzo,
Vicente; Martínez-Urreaga, Joaquín
[CHE-05] DecarGas. Low-carbon clean fossils for the next generation. Abánades, Alberto; Wetzel,
Thomas; Stoppel, Leonid
SCIENTIFIC AGENDA
ELECTRONICS
[ELE-01] Control of a single phase inverter with multiple modulation strategies. Ramos, Regina; Serrano,
Diego; Oliver, Jesús A.; Alou, Pedro; Cobos, José A.
[ELE-02] Advanced topology solutions for high power DC-DC converters for aircraft applications.
Bouvier, Yann E.; Alou, Pedro
[ELE-03] Isolated Single-Stage Three-Phase Full-Bridge with Current Injection Path PFC Rectifier for
Aircraft Application. Zhao, Sisi; Oliver, Jesús A.; Alou, Pedro
[ELE-04] Three-phase Buck Rectifier For Aircraft Application With Variable Line Frequency. Borović,
Uroš; Alou, Pedro;Oliver, Jesus
[ELE-05] A 99.5% efficient 50kW/dm3 hybrid DC-DC step-up converter for PV applications. Serrano,
Diego; Vasić, Miroslav; Oliver, J.A.; Alou, Pedro; Cobos, J.A.
[ELE-06] IFEC challenge: High efficiency and high power density isolated DC-DC converter for server
applications. Vasic, Miroslav; Čohadžić, Dusan; Millán, José Luis; Murga, Ignacio; Gil, Diego;
Zubitur, Iñigo.
[ELE-07] Linear Assisted Versus Purely PWM Switched Mode Envelope Amplifier for EER Transmitter.
Lazarević, Vladan; Vasić, Miroslav; García, Óscar; Patiño, Moisés; Ortega, Francisco
[ELE-08] On the use of blackbox polytopic models to simulate dc microgrids. Francés, Airán; Asensi,
Rafael; García, Óscar; Prieto, Roberto; Uceda, Javier
[ELE-09] An Analysis of the Influence of Binarization Techniques Applied to Swarm Optimization when
Solving the Set Covering Problem. Lanza-Gutierrez, Jose M.; Riesgo, Teresa
[ELE-10] Automatic Integration of HLS-Generated Hardware Accelerators in an Embedded Parallel
Computing Architecture. Rodríguez, Alfonso; de la Torre, Eduardo
[ELE-11] Resilient communications for a multi-FPGA, hardware-accelerated cluster. Ortiz, Alberto;
Rodríguez, Medina; de la Torre, Eduardo
[ELE-12] Distributed Artificial Neural Network over a Wireless Sensor Network. Aledo, David; Moreno,
Félix
ELECTRICAL ENGINEERING AND MACHINES
[ELM-01] Contribution of DFIG wind turbines to Fast-Frequency Response and Power Smoothing. Ochoa,
Danny; Martinez, Sergio
[ELM-02] Improved Predictive Direct Power Control of Doubly Fed Induction Generator during
unbalanced grid voltage based on Four Vectors. Zarei, Mohammad Ebrahihm; Veganzones
Nicolás, Carlos; Rodríguez Arribas, Jaime.

ENERGY
[ENE-01] Supercritical CO2 in a concentrating solar power plant with thermal energy storage. González,
Luis F.; Muñoz, Javier; Martínez-Val José M.
[ENE-02] Reloj de Sol. Cano, Javier; Martínez-Val, Jose María; Abbas, Rubén; San Millán, Julio;
Rodríguez, Andrés
[ENE-03] Solar heater: solar tower power at different enthalpies. Muñoz - Antón, Javier; González-
Portillo, Luis F.;Martínez-Val José M.
[ENE-04] A comparison of optical features between parabolic trough collectors and linear Fresnel
collectors. Abbas, Rubén; Martínez-Val, José María
[ENE-05] Regulatory proposals for the development of renewable energy self-consumption in Spain.
Mármol-Acitores, Gloria; Rodríguez-Monroy, Carlos
[ENE-06] A review on the effects of structural breaks and volatility clustering characteristics of Spanish
gas and Brent prices. Cansado, Pablo Angel; Rodriguez, Carlos
[ENE-07] Time-dependent atomic physics in the interaction of ultraintense X-ray laser with matter.
Fernandez Tello, Elisa V; Cotelo, Manuel; Oliva, Eduardo; De la Varga, Alberto G.;
Velarde, Pedro
SCIENTIFIC AGENDA

ENVIRONMENTAL ENGINEERING
[ENV-01] Environmental sustainability assessment of municipal solid waste management through carbon
footprint. Pérez, Javier; Lumbreras, Julio; Rodríguez, Encarnación
[ENV-02] Improvements of air quality simulations of particulate matter in the Iberian peninsula with road
dust resuspension and dust emission modules. de la Paz, David; Borge, Rafael; Vedrenne, Michel
[ENV-03] Modelling the effects of local emission abatement measures in hot-spots of Madrid. Quaassdorff,
Christina; Borge, Rafael
LASER TECHNOLOGY
[LAS-01] Assessment of 2D geometries in immune systems responses using LIFT (Laser Induced Forward
Transfer) as a high accuracy bioprinting technique. Lauzurica, Sara; Gómez-Fontela, Miguel;
Márquez, Andrés; Notario, Laura; Lauzurica, Pilar; Molpeceres, Carlos
[LAS-02] Laser direct writing techniques for metallization of solar cells. Munoz-Martin, David; Chen Yu;
Morales, Miguel; Canteli, David; Molpeceres, Carlos
[LAS-03] Laser functionalization of surfaces. Canteli, David; Márquez, Andrés; Munoz-Martin, David;
Faleiro, Gabriel; Santamaría, Samuel; Lauzurica, Sara; Morales, Miguel; Molpeceres,
Carlos
[LAS-04] Laser crystallization of silicon and study by finite element modelling. Canteli, David; Munoz-
Martin, David; Morales, Miguel; Molpeceres, Carlos
[LAS-05] Numerical study on the effects of dimensionless parameters on Laser Induced Forward Transfer
printability range. Morales, Miguel; López, Iván; Munoz-Martin, David; Molpeceres, Carlos

MATERIALS
[MAT-01] Crystallization Behaviour of PHBV-INT-WS2 Inorganic Nanotubes Nanocomposites. Silverman,
Tyler; Naffakh, Mohammed; Marco, Carlos; Ellis, Gary
[MAT-02] Electrode modified with Thiolated DAB dendrimers to form self-assembled monolayers with AuNPs
and HRP for biosensing. Ospina, Evelyn; Garcia M. Pilar; Alonso, Beatriz, Casado,
Carmen
[MAT-03] Poly(propyleneimine) Dendrimers: Combined Experimental and Simulation Study. León,
Salvador; García, Pilar; Fernández, Andrés
[MAT-04] Elongation of metallic nanoparticles embedded in a dielectric matrix produced by irradiation with
swift heavy ions. Prada, Alejandro; Peña-Rodríguez, Ovidio; Olivares, José; Oliver,
Alicia; Rodríguez-Fernández, Luis; Bringa, Eduardo; Perlado, José Manuel; Rivera, Antonio
[MAT-05] In situ monitoring and modelling the optical properties of silica irradiated with swift heavy ions.
Díaz-Núñez, Pablo; Peña-Rodríguez, Ovidio; R. Páramo, Angel; Prada, Alejandro; Rivera,
Antonio; Sordo, Fernando; Perlado, Jose Manuel; Crespillo, Miguel; Olivares, José
[MAT-06] Influence of grain boundaries on the radiation-induced defects and H retention in nanostructured
and coarse-grained tungsten. Panizo, Miguel; Valles, Gonzalo; González, César; Martín, Ignacio;
González, Raquel; Gordillo, Nuria; Iglesias, Roberto; Guerrero, Carlo; Perlado, José
Manuel; Rivera, Antonio.
[MAT-07] Water absorption in polypropylene-cellulose ecocomposites. González-Defix, Anaïs; de la Orden,
Mª Ulagares; González-Sánchez, Carlos; Beltrán, Freddys; Lorenzo, Vicente; Martínez
Urreaga, Joaquín
[MAT-08] Definition of a secondary metallurgy process for the modification of oxide inclusions by Ca
treatment in Al-killed high speed machining steels. Alonso, Izaskun; Berg, Martin
[MAT-09] Determination of the interfacial behavior in graphene oxide-epoxy nanocomposites by multiscale
modeling. Fernández Zapico, Guillermo; Muñoz Guijosa, Juan Manuel
SCIENTIFIC AGENDA

MANAGEMENT AND INDUSTRIAL ORGANIZATION
[MIO-01] Project Management Coaching Model: First insights. Ballesteros, Luis; Ortiz, Isabel
[MIO-02] Cultural risks in projects developed in South and East countries. An approach to identify risks in
international cooperation projects. Rodríguez, Rocío; Ortiz, Isabel.
[MIO-03] Open data assessment of spanish cities. Sisto, Raffaele; García, Javier
[MIO-04] Mapping city metrics methodologies. García, Javier; Sisto, Raffaele
[MIO-05] Entrepreneurship policy agenda in the EU explained: a contribution to understand the European
entrepreneurial path and its way forward. Arenal, Alberto; Armuña, Cristina
[MIO-06] Exploring the factors that contribute to transform entrepreneurial intention into action.
Reflections on potential entrepreneurs' needs and entrepreneurial education programs. Armuña,
Cristina; Arenal, Alberto
[MIO-07] Big Data for Sustainability in the Digital Era: Methodological Framework towards a Conscious
and Smart Governance. Del Río, Gema; González, Camino; Uruburu, Ángel
[MIO-08] Unwarranted Variations in Healthcare and the Role of Business Intelligence. Lessanibahri,Sina;
Gastaldi,Luca;González Fernández,Camino
[MIO-09] Increasing the trading prediction by mining aggregated human texting messages. Yang Liu;
Joaquín Ordieres Meré
[MIO-10] Measuring general elections in Spain via Twitter, based on geolocated user comments. Franco
Riquelme, José N.; Ordieres Meré, Joaquín B.
[MIO-11] Revisiting the past and addressing the future of public infrastructure project. Hetemi, Ermal;
Ordieres M., Joaquin; Dominguez, María L.
[MIO-12] International Manufacturing Network: Operational issues and reshoring. Sayem, Ahmed; Ortega
Mier, Miguel, Feldmann, Andreas
[MIO-13] Application of Optimization and Simulation in the Aeronautic Industry. Pulido, Raúl; Borreguero-
Sanchidrián, Tamara; García-Sánchez, Álvaro; Ortega-Mier, Miguel.
[MIO-14] Mandatory Convertibles: An Efficient Form of Capital Issuance. Huerga, Angel; Rodriguez-
Monroy, Carlos
[MIO-15] ICTs and their impact on the profitability of SMEs in the main tourist areas of Guatemala.
Rivera, Sigfrido; Rodríguez-Monroy, Carlos
[MIO-16] Innovation Management in Knowledge Intensive Service Firms. Lemus-Aguilar, Isaac; Hidalgo,
Antonio; Cagliano, Raffaella
[MIO-17] Determinants of governance structure in multipartner alliances: a multiple case study. Barbic,
Frano; Hidalgo, Antonio.
[MIO-18] Network Cooperation - an emerging model of connected collaboration for the Sustainable
Development Goals 2030 agenda. Acevedo, Manuel; Moreno, Ana; Mataix, Carlos.

MECHANICS, MANUFACTURING AND STRUCTURES
[MMS-01] Process analysis of a 4 axis single point micro cutting machine. Lopez-Estrada, Luis;
Fajardo,Marcelo;Vizan,Antonio
[MMS-02] Cutting parameters identification in end milling operations. Leal-Muñoz, Erardo; Vizan,
Antonio
[MMS-03] Development of a 3PRS-XY Machine Tool to Single Point Cutting Process in Microcutting.
Fajardo, Marcelo; López-Estrada, Luis; Vizán, Antonio
[MMS-04] Characterization physical and chemical properties of epoxy resins for additive manufacturing by
photopolymerization. V. Niño, Ely D.; Endrino, José L.; Diaz, Andrés
[MMS-05] Cell scaffolds design using pentamode structures based on Bravais Lattices. Fernández Méjica,
Graciela; Díaz Lantada, Andrés
[MMS-06] Modelling of the mechanical behaviour of the photopolymerization processes of resins intended for
additive manufacturing (AMT) using laser-stereolithography: The influence on part quality. de
Blas, Adrián; Fernández, Guillermo; Díaz, Andrés
SCIENTIFIC AGENDA

[MMS-07] Similarity index as a tool to compare impact history records. Pacios Álvarez, Antonia; Huerta
Gómez de Merodio, M. Consuelo; Alonso Álvarez, Jesús; Parra-Hidalgo, José Antonio;
López Lázaro, Luis
[MMS-08] Human Impact on glass: study of the excitation level measured on a plate versus energy applied
by the pendulum. Huerta, Mª Consuelo; Pacios-Álvarez, Antonia; Parra-Hidalgo, José A.;
Alonso, Jesús; Pardo, Sergio
[MMS-09] Smartwatch’s usefulness for hand - arm vibration exposure assessment. Sigcha, Luis; Pavón,
Ignacio; López, Juan Manuel; De Arcas, Guillermo
[MMS-10] Bending strength study of repaired coaches structures. Rincón, Dávila; Alcalá, Enrique
[MMS-11] Design, fabrication and test of a short mechanical model of the orbit corrector magnets for the
High-Luminosity LHC. García-Matos, Jesús A.; Abramian, Pablo; Calero, Jesús; Gómez,
Pablo; Gutiérrez, Jose Luis; García-Tabarés, Luis; Lopez, Daniel; Munilla, Javier; Toral,
Fernando; Bourcey, Nicolas; Fessia, Paolo; Izquierdo-Bermudez, Susana; Pérez, Juan
Carlos; Todesco, Ez

MATHEMATICS AND STATISTICS
[MST-01] Extreme Value Theory: Applications for adverse weather conditions. Sáez, Cristina; García-
Martos, Carolina; Sánchez, María Jesús
[MST-02] Local forecasting of electric demand: Madrid. Carretero, Miguel; Caro, Eduardo; Juan, Jesús;
Cara, Javier

NUCLEAR TECHNOLOGY
[NUC-01] Advanced variance reduction methods in Monte Carlo calculation. Magán, Miguel;Sordo,
Fernando;Perlado, Jose M.
[NUC-02] Monte Carlo study of detection systems of explosives and illicit substances using a D-D neutron
generator. Cevallos, Lenin; Gallego, Eduardo; Guzmán, Karen; García Gonzalo
[NUC-03] Fukushima DAiichi Unit 2 Accident Analysis with MELCOR 2.1. Bocanegra, Rafael; Di Marcello,
Valentino; Sánchez-Espinoza, Victor H.; Jiménez, Gonzalo
[NUC-04] Towards the definition of a local damage criteria for containment safety analysis. Fernández-
Cosials, Kevin; Bocanegra, Rafael; Estévez-Albuja, Samanta; Goñi, Zuriñe; Jiménez,
Gonzalo
[NUC-05] Nuclear Data for the Safety of Lead-based Cooled Facilities. Romojaro, Pablo; García-Herranz,
Nuria; Álvarez-Velarde, Francisco

SEISMIC ENGINEERING
[SEI-01] Numerical simulation of shaking table tests on a reinforced concrete waffle-flat plate structure.
Galé, David; Benavent, Amadeo
[SEI-02] Seismic upgrading of frame structures with a soft first story by using hysteretic dampers. Mota
Paez, Santiago Felix; Benavent Climent, Amadeo
[SEI-03] Seismic design of structures with seismic isolation based on rocking columns. Rios, Germán;
Benavent, Amadeo

VEHICLES AND RAILWAYS
[VEH-01] Algorithms to aid operational management in urban transport systems. Herrero, Daniel; García,
Álvaro; Ortega, Miguel
[VEH-02] Development of an energy management embedded controller for a range-extended hybrid vehicle
using model-based design and HILs. Resino, Daniel; Marcos, David; López, José María;
Fernández, David
[VEH-03] A comparison of local path-planning interpolation methods for autonomous driving in urban
environments. Artuñedo, Antonio; Godoy, Jorge; Villagra, Jorge
SCIENTIFIC AGENDA

[VEH-04] Method for knowledge based maintenance of current collection systems in railway installations.
Calvo, Álvaro; Sanz, Juan de Dios
[VEH-05] Accuracy improvement of finite element models of electric vehicles components manufactured with
thermoplastic materials in crash environments. Rivera, Javier; Martínez, Luis; García Antonio
[VEH-06] Strategies for grid integration of electric vehicles. Alvaro-Hermana, Roberto; Fraile-Ardanuy,
Jesús; Martinez, Sergio
[VEH-07] Large-signal modeling of bidirectional battery charger for electric vehicles. System-level approach
based on agent-based models in v2g environment. Naziris, Antreas; Asensi, Rafael; Uceda, Javier
[VEH-08] Hybridization Percentage Influence on Hybrid Energy Storage Systems Efficiency for Electric
Buses. Nájera, Jorge; Rodríguez, Jaime; de Castro, Rosa María
[VEH-09] Remote Drive: A military UGV for surveillance missions. Clavijo, Miguel; Jiménez, Felipe;
Naranjo, José E.; Gómez, Óscar

PATENT
[PAT-01] Horizontal rotary system for concentrating solar radiation. Martínez-Val, José M
[PAT-02] Brayton cycle with heat sink temperature close to the critical isothermal line. Martínez-Val,
José M
[PAT-03] Intermediate heat exchanger of free-dilatation concentric tubes. Martínez-Val, José M
[PAT-04] Convection inhibitted open-air solar thermal receiver. Martínez-Val, José M
[PAT-05] A concentric tube penetration device. Martínez-Val, José M
[PAT-06] Gasket for assembling rotatory pipes. Martínez-Val, José M
[PAT-07] Procedure and system for the automatic detecion of thrust reverse activation during aircraft
landing. Asensio, César;Ruiz, Mariano; de Arcas, Guillermo;Recuero, Manuel; López, Juan
Manuel
[PAT-08] On board car system and method for road status surface identification. López, Juan Manuel;
Alonso, Jesús; Ruiz, Mariano; de Arcas, Guillermo; Pavón, Ignacio; Asensio, César
[PAT-09] Electronic system for enviromental acidity (pH) measurment based on optical sensors. Mª Á.
Villegas, M. García, J. Peña, G. de Arcas, J.M. López, E. Barrera, A. Llorente
[PAT-10] Sistema de Calibración de Instrumentos Acústicos. G. de Arcas, J.M. López, M. Ruiz, M. Recuero,
R. Fraile
[PAT-11] System and procedure for the real time detection and identification of sound produced by specific
sound sources. C. Asensio, M. Ruiz, M. Recuero
[PAT-12] System and method for monitoring a rapid de-excitation system for synchronous machines. Platero,
Carlos;Granizo,Ricardo;Rebollo,Emilio;Blanquez,Francisco
[PAT-13] Rapid de-excitation system for synchronous machines with indirect excitation.
Platero,Carlos;Blázquez,Francisco;Redondo,Marta;Granizo,Ricardo
[PAT-14] Process of Fungal Biodegradation on Ground Tyres. Ruibal, Constantino; García, Ana M.;
Moreno, Diego A.; Benavides, Enrique
[PAT-15] Bioremediation method which is used to concentrate and eliminate radionuclides in radioactive
water. Moreno, Diego A.; Montero, Felipe
[PAT-16] Portable and autonomous device for airborne particles collection. Núñez, Andrés; García, Ana M.;
Moreno, Diego A.
[PAT-17] Procedure for marking, encryption, labelling and optical coding. M. Holgado, A. Barranco, R.
Casquel. A.R. González-Elipe, F.J. Sanza aM.F. Laguna, bSanchez valen cia Juan Ramon; bA
Rebollo aF.J. Sanza
[PAT-18] Interferometric Detection Method. Holgado, Miguel; Sanza, Francisco Javier; Laguna, María Fe;
Lavín, Álvaro; Casquel, Rafael
[PAT-19] Optical detection system for labelling-free high-sensitivity bioassays. Holgado, Miguel; Casquel,
Rafael; Laguna, María Fe, Molpeceres, Carlos, Morales, Miguel, Ocaña, José Luis
[PAT-20] Method and system for estimating the forces and torques generated by the contact of a tire with the
road in an instrumented wheel. Javier, García De Jalón De La Fuente; María Dolores, Gutiérrez
López
SCIENTIFIC AGENDA

[PAT-21] Medical devices based on intelligent materials and complex geometries patented and developed
during last decade within UPM’s Product Development Laboratory. Diaz Lantada, Andrés; Ortego
García, Pedro; Lafont Morgado, Pilar
[PAT-22] Feeding carriage for compressor machine and use of same. Ordieres, J.; Alba Elias, F.; González
Marcos, A.; Martínez Martínez, L.
[PAT-23] Method for increasing the net electric power of solar thermal power plants. Platero, Carlos;
Blázquez, Francisco; Redondo, Marta; Rodríguez, Jaime
[PAT-24] Método de guiado de dispositivos robóticos teleoperados. Ferre, Manuel; Aracil, Rafael; Barrio,
Jorge; Suárez, Francisco A.

UPM Institutional participation

Oficina de Proyectos Europeos (OPE)

Oficina de Transferencia de Resultados de Investigación (OTRI):

 Área de Propiedad Intelectual e Industrial
 Área de Cultura Científica
 Área de Colaboración público-privada

Centro de Apoyo a la Innovación Tecnológica (CAIT)
SCIENTIFIC AGENDA

LAYOUT OF THE POSTER SESSION
SCIENTIFIC AGENDA

WEDNESDAY, April 5th 2017
9:00 - 09:30 WELCOME (Salón de Actos)
Guillermo Cisneros Pérez, R ector of the Universidad Politécnica de M adrid
José Manuel Torralba Castelló, Director General Universidades e Investigación
de la Comunidad de Madrid
Emilio Mínguez Torres, Director of the ET SII - UPM

9:30 - 10:30 PLENARY SESSION (Salón de Actos)
Avelino Corma Canós, Premio Príncipe de A sturias de Investigación
Científica y Tecnológica 2014
Diseño molecular de catalizadores, desde la investigación fundamental hasta la
aplicación industrial

10:30 - 11:30 KEYNOTE PRESENTATIONS (Salón de Actos)
Chairman Antonio Rivera
Alberto Abánades Velasco, Professor ETSII (UPM)
DecarGas: El camino hacia el aprovechamiento de recursos fósiles sin emisiones.

Sara Lauzurica Santiago, Assistant Professor ETSII (UPM)
Bioimpresión asistida por láser para el estudio de los mecanismos de respuesta del
Sistema inmune

11:30 - 14:00 POSTER SESSION (Sala de la Máquina)

This session is dedicated to both Research Groups of the ETSII and Companies.
The layout of the Sala de la Máquina with the poster distribution will be shown in the final version
of the program.

RESEARCH GROUPS, CENTERS AND INSTITUTES
Administración de Empresas
Bioingeniería y Materiales (BIO-MAT)
Ciencia y Tecnología de Sistemas Avanzados de Fisión Nuclear
Control Inteligente
Estadística Computacional y Modelado Estocástico
Fusión Nuclear Inercial y Tecnología de Fusión
Generación Eléctrica con Energía Eólica
Ingeniería de Calidad Alimentaria
SCIENTIFIC AGENDA

Ingeniería de Fabricación
Ingeniería de Máquinas (GI - IM)
Ingeniería de Organización y Logística
Ingeniería Sísmica
Ingeniería y Aplicaciones del Láser
Instrumentación y Acústica Aplicada (I2A2)
Laboratorio de Sistemas Autónomos
Manufactura Avanzada del Láser
Mecánica de fluidos aplicada a la Ingeniería Industrial
Modelado y Simulación en Ingeniería Mecánica
Modelización de Sistemas Termoenergéticos
Óptica, Fotónica y Biofotónica
Polímeros: Caracterización y Aplicaciones
Proyectos y Calidad
Robótica y Cibernética
Robots y Máquinas Inteligentes
Seguridad e Impacto Medioambiental de Vehículos y Transportes (GIVET)
Tecnologías Ambientales y Recursos Industriales
Teoría de Aproximación Constructiva y Aplicaciones
Termodinámica Aplicada a la Ingeniería Industrial (TAII)
Visión por Computador y Robótica Aérea

Centro de Electrónica Industrial
Centro de Investigación en Automática y Robótica
Centro Láser

Centro de Investigación en Tecnologías Ferroviarias (F2I2)
Centro de Modelado de Ingeniería Mecánica (F2I2)

Instituto de Fusión Nuclear
Instituto Universitario de Investigación del Automóvil
SCIENTIFIC AGENDA

COMPANIES

Chairmen Antonio Rivera, Nuria Martínez
The following companies have confirmed its participation in IRM. The speakers will show the main
working lines and projects to the attendees in the electronic poster session
SCIENTIFIC AGENDA

LAYOUT OF THE POSTER SESSION
SCIENTIFIC AGENDA

14:00 - 14:30 AWARDS CEREMONY & CLOSING (Salón de Actos)

In the closing session, several awards will be given:

- Award to the Design of the cover page of IRM17. This award has been won by
Jesús Tordesillas Torres.
- Two awards to the best oral presentation of sessions 1 & 2. A Committee will select the
best presentation of these sessions.
- Award to the best poster of poster session. Another Committee will select this poster
among all posters presented during April 4th poster session.
- Raffle of a submar ine camer a (the registered attendees will receive the ballots that
should be deposited during the posters sessions of April 4 & 5). The winner should be at
this session to receive the prize.
Noise data communication as a non-acoustic
factor for mitigating aviation noise.
L.Gascó , C.Asensio, G. de Arcas

Instrumentation and Applied Acoustics Research Group (I2A2)
Universidad Politécnica de Madrid – luis.gasco@i2a2.upm.es

Despite the efforts of aeronautical sector in reducing noise in the vicinity of airports, the number of complaints has
not decreased as much as expected. For that reason, airports have started new approaches to increase the
acceptance of these infrastructures. One of these strategies is to improve the communication to the public. Noise
information is a difficult technical topic for a non-expert, and there are many ways to implement it. This poster
summarizes how 70 airports worldwide communicate noise matters. In the future, the effectiveness of these data
communications will be evaluated.
Transparency and communication to the public are gaining importance over the years. For that reason, it has
become a major challenge for airport authorities. One of the keys of the communication process is the data
visualization that is used. This is why airports use maps to show noise data, aircraft pathways and noise complaints.
Noise maps Flight-path maps Noise complaints maps

Noise maps are used for assessing the Flight-path maps give information about the The number and distribution of complaints can
environmental noise exposure in a specific area runway usage. Airports usually display this be good indicators of the acoustic effects on the
prior conducting action plans to reduce noise data in two ways: areas surrounding airports. There are some
emission: • Individual lines that represent each path methods to give this information:
• Use of Leq based indicators followed by aircrafts during the landing • Areas near airports filled in different colours
• They are simulated, not measured and the take off. according to the number of complaints.
• Use of isophonic lines, or coloured areas • Density of flights that overfly the areas • Flight density areas with individual
using a colour gradient layer. complaints represented by points in the map

Public access to noise web systems
Some airports provide access to real-time information using web
applications. The key aspect of these tools is the data they provide and
how it is reported:
• Noise levels: Citizens can see noise level in two ways. Using a colour
tag based on the noise level measured and through a time-
evolution graphic that is updated every second.
• Routes: This type of platform shows the pathways taken by
airplanes in real-time, using an airplane icon and with a line in
different colours according to the operation (Land or Take off) or
with a colour gradient that represents the aircraft height during its
path
• Complaints: These applications offer the possibility of identifying an
annoying airplane and make a complaint referring directly to that
plane.

Gaps and needs • User feedback: A citizen can complain online in the application framework, but
The generalized use of internet in recent years has improved data why should citizens use this channel? New life-trends have changed people,
transparency and accessibility for the general public. Finding noise and the use new channels, as social networks, to express their discomfort and
communication is easier now that it was a few years ago. However, to access information. Airports should actively use this channels.
the increasing transparency has highlighted some important • Overly technical reported information: It is difficult for non-experts to
drawbacks: distinguish between different indicators that use the same units (dB). Noise
Gasco, L., et al., Communicating airport noise emission data to the general public,
information metrics should be customized to different groups of users, experts
Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2017.02.063 and non-experts.
Valida-on of therapies for Tinnitus treatment.
Enriched Acous-c Enviroment (EAE).
a P. Cobo, b C. Garzón
a Ins-tuto de Tecnologías Físicas y de la Información (ITEFI), Consejo Superior de Inves-gaciones

CienIficas (CSIC) (pedro.cobo@csic.es)
b Facultad de Ingeniería. Universidad de Las Américas, Quito-Ecuador, Programa de Doctorado de Ingeniería

Acús-ca. Universidad Politécnica de Madrid (chris-am.garzon@udla.edu.ec)

Abstract
Tinnitus is defined as the percep0on of sound in the absence of an external acous0c s0mula0on. However, Jastreboff and Jastreboff [1] prefer to define it
as the percep0on of a sound that results exclusively from the ac0vity within the nervous system, without any corresponding vibratory mechanical ac0vity
within the cochlea; this pathology has no cure, but according to studies carried out in the last years, the hypothesis is that it can be reversed at an early
stage. The objec0ve of this work is the experimental valida0on of Enriched Acous0c Enviroment (EAE) in a controlled number of pa0ents by the
s0mula0on of the auditory pathway with acous0c pulses that include the inverse curve of HL, which is the cause of its appearance in a very high
percentage of pa0ents.
Burst tones
Framework Burst tone is a pure tone limited to a reduced number of cycles. This
s0mulus is the one that best mediates the need for frequency specificity
Tinnitus and short dura0on. The enriched acous.c environment (EAE) consists of a
Most people have ever experienced a ringing in the ears that has not been sequence of this type of tones of 100 ms dura0on, with an interval
caused by any external sound source. Typically, this sensa0on is associated between s0muli of 100 ms (± 10 ms). An integer sequence contains this
with some reversible cause [2], such as exposure to very loud music, fever, type of tonal pulses, of random variable frequency between the lowest
ototoxicity, or transient disturbance of the middle ear. However, in 5-15% frequency of the loss curve and the frequency corresponding to the
of the popula0on this internal noise does not return. This sensa0on is more Level_max, with intervals of 1/64 octave and weighted amplitude
prevalent in elderly popula0on (12% in the over 60s) than in younger according to the loss curve. [8]
popula0on (5% between 20 and 30 years). In 1-3% of the general The most interes0ng thing about this therapy is that it is based on the loss
popula0on, it is high enough to affect quality of life, involving sleep curve, and therefore is specific to each pa0ent. It would therefore be a
disturbance, impairment of work capacity, and psychiatric distress [2]. personalized therapy.
According to data from Jastreboff and Jastreboff [1], between 25 and 52
million Americans suffer from this symptom, being more prevalent in the
popula0on over 65 years (30%). In 5% of the general popula0on (about 13
million of pleple from the USA), this noise is significant enough to qualify as
a clinician. Weisz et al. [3] state that 5 million Europeans experience
moderate to severe inner noise, resul0ng in sleep problems, difficul0es in
concentra0on, increased anxiety or depression. In Spain, 17% of pa0ents in
a ENT clinic reported these types of noises [4].

Neural Plas-city
Fig.5 Audiometric threshold values and
Fig.4 Audiogram of person with Tinnitus
Neural plas0city refers to the ability of the nervous system to change its frequencies
structure and func0oning throughout its life. As a reac0on to the
Environment Neuroplas0city, it allows neurons to regenerate both
anatomically and func0onally forming new synap0c connec0ons. [5]
Neural plas0city represents the brain's ability to recover and restructure.

Fig.6 Enriched Acous0c Enviroment Program

References
1. Jastreboff P.J., Jastreboff M. M. (2000). Tinnitus RetrainingbTheraphy (TRT) as a method
for tratmet of Tinnitus and hyperacusis pa0ents. J am Acad Audio, 11: 162-177.
Fig.1 Neural Networks Fig.2 Neural Networks Fig.3 Neural Networks 2. Eggermont JJ (2006). “Proper0es of correlated neural ac0vity clusters in cat auditory
before training 2 w e e k s o f c o g n i 0 v e 2 months of cogni0ve cortex resemble those of neural assemblies.” J. Neurophysiol. 96:746-764.
s0mula0on s0mula0on 3. Weisz N., Voss S., Berg P., Elbert T. (2004). Abnormal auditory mismatch response in
0nnitus sufferers with high-frequency hearing loss is associated with subjec0ve distress
The explana0on of 0nnitus as a plas0c phenomenon predicts that it is level. BMC Neuroscience, 5: 1-9.
possible to reverse it by properly s0mula0ng the auditory pathway. These 4. Herráiz C, Hernández Calvín F.J. (2002). Acúfenos. Actualización. Ars Medica
plas0c changes can result in habitua0on and/or residual inhibi0on, this
5. Kolb, B., Mohamed, A., & Gibb, R., La búsqueda de los factores que subyacen a la
being the diminu0on or disappearance of the sensa0on of 0nnitus when the plas0cidad cerebral en los cerebro normal y en el dañado, Revista de Trastornos de la
s0mulus ceases. [6] Comunicación (2010), doi: 10.1016/j.jcomdis.2011.04 0.007
Therapies 6. Cobo P (2006). “Acúfenos: revisión del estado del arte.” Informe Técnico, Ins0tuto de
Acús0ca, CSIC.
Different sound therapies have been proposed, such as masking, 0nnitus 7. Cobo P, Bañuls V, Delgado C, Pons J (1994). “Potenciales evocados: ventanas al sistema
retraining (TRT), auditory discrimina0on training (TAD), phase change audi0vo”. Informe Técnico, Ins0tuto de Acús0ca, Madrid.
treatment (PST), and Neuromonics. The use of EAE as the best s0mulus for 8. Cobo P (2008). “Revisión de las terapias sonoras del acúfeno.” Informe Técnico, Ins0tuto
0nnitus therapy [7]. de Acús0ca, CSIC.
ACOUSTIC STIMULATION FOR PARKINSON'S
DISEASE: A LONGITUDINAL PILOT STUDY
aG. Gálvez, aD. González , aM. Recuero, aG. de Arcas
aInstrumentation and Applied Acoustics Research Group (I2A2-UPM)
Technical University of Madrid (UPM). (Email: gerardo.galvez@i2a2.upm.es)

Abstract
Binaural auditory beats could be a potential method to assist pharmacological treatment in neurological diseases, such as Parkinson's disease (PD). Despite
pharmacology improves several symptoms, it has important side effects and loss efficiency (long-term). For that reason, complementary methods such as
acoustic therapy could play an important role. Reviewing the existing literature in this context, there is no consistent methodology to apply sound to
patients and we propose a methodology for a longitudinal pilot study with 12 Parkinson Disease’s patients who participated in a 24 weeks listening
program. We collected data from patients in three sessions: before, in the middle and in the end of the listening program. Every session consisted in: an
interview, a set of specific questionnaires and electroencephalography recording. This methodology proposes a further step in the knowledge of the
neuronal mechanisms involved in the processing of this kind of acoustic signals.

Background Methodology
Neurological disorders, a big problem Acoustic stimulus design

Neurological disorders correspond
50% of all disabling disease
Digital own design Acoustical properties Sound recorded in
sound with Matlab evaluation with HATS mp3 player
and Pulse Analyzer
Complementary therapies are Participants profiles Ethics
sought to manage symptoms • 12 PD patients; ≤ 3 H & Y scale • Double blind and placebo-
controlled experimental design
• Good response to medication
• Normal Hearing (Audiometry in • Following “Principles of the
the 1st session) Declaration of Helsinki” is required
• DBS not implanted • Patients were in “ON” state

Longitudinal study
Schedule
JUNE SEPTEMBER DECEMBER
Parkinson Disease (PD) 2016
3 months
2016
3 months
2016
Inter-sessions Inter-sessions
1ST Session 2ND Session 3RD Session
PD is 2nd most prevalent Sessions
neurological disease PDQ-39 With & without
WOQ-19 acoustic stimulus
Interview Questionnaires Cognitive Voice EEG Recording
Test Recording
Current treatments don’t stop Inter-sessions
progression and side effects are Listen to the acoustic stimulus at home 10 min/day - 3 days/week
significant and varied.
Analysis
Symptomatology
• Statistical pre-post analysis with SPSS and Matlab
• Functional Connectivity and EEG Power

Rigidity Bradykinesia Instability Tremor Challenges & future
Challenges Future
Acoustic engineering in Parkinson Disease Understanding of neural basis of Test different stimulus strategies:
processing sound mechanisms FEATURES – DURATION – DOSE
Objetives
Acoustic
engineering Acknowledgements
We thank “Asociación de Parkinson de Madrid” and “Center for Biomedical
Technology” of UPM.
Diagnosis Treatment
Multi-UAV Coordination and Control Interface
Juan Jesús Roldán Gómez
Centre for Automation and Robotics (CAR, UPM-CSIC)
jj.roldan@upm.es

Abstract
Let’s imagine a fire control mission performed by multiple robots and commanded by a single operator. This scenario poses two challenges related to
human factors: workload - the operator has to receive data, discover information, make decisions and send commands - and situational awareness -
he/she has to know what is happening at any time of the mission. This work aims for the selection of the information that is shown by the interface to the
operator. This information should be outlined according to the mission’s state and evolution together with the operator class, state and preferences. The
expected result is an intelligent adaptive interface that provides the most relevant information for each task.

Chapter 1: A game of drones
Multi-robot missions generate a huge amount of data. This fact may cause human factor problems...

? ?
?

Workload Situational awareness

Chapter 2: Deconstructing mission
Data The data from previous missions is collected Petri nets are used to define the mission Information
1 by event logs. 3 state and decision trees predict its evolution.
Robot 1:
- Distance to base.
Case Activity Time Resource Robot 2:
M1 Surveillance 1:10 UAV 1 - Battery.
- Altitude.
M1 Reconnaissance 1:13 UAV 2 DT Robot 3:
M1 Capture 1:17 UAV 1 PN - Camera.
Robot 4:
- Temperature.
The models are automatically generated The models are applied to discover relevant
2 from the event logs with process mining. 4 information according to the situation.
- Water tank.

Chapter 3: There’s something about Operator
Data mining allows to discover … and machine learning can be
patterns and classify operators… used to find their preferences.
Now the
information Operator #1
- Reduce amount of information.
is adapted to Class Professional - Select preferred variables.
the mission, Stress High
but what
about the Operator #12
operator? - Information useful for training.
Class Training - Activate decision support.
Stress Low

Chapter 4: The lord of the UAVs
Priority Map One moment… Why not try something else?
UAV 3
Priority
UAV 3 Task: Surveillance
The result is Progress: 95%
Targets: 0
an intelligent
UAV 1
adaptive
interface UAV 2 UAV 1
that helps Task: Tracking
the operator. Target: V1
UAV 4 Distance: 2,000m

UAV 1 is finishing
Immersive interface with
Events UAV 3 battery is low.
surveillance.
UAV 2 is landing. … virtual / augmented reality.
Low Power Distribution Module for Space
Application: Analysis and Comparison of
Different Architectures and DC/DC Topologies
aB. Stevanović, G. Salinas, P. Alou, J. A. Oliver, M. Vasić
aCentro de Electrónica Industrial (branislav.stevanovic@upm.es)

SUMMARY The goal of this project is to analyze and compare different possibilities for the design of the Power Distribution Module for space application that is
supplied from the Main Bus which can operate in the two voltage ranges: [45V, 55V] and [19V, 38V] and needs to provide four different outputs: 28V, 1.75A; 5V,
7A and ±15V, 250mA galvanically isolated from the Main Bus. The outputs are not necessary to be galvanically isolated between each other. Two possible
approaches are analyzed and compared: design of the system that operates in the wide input voltage range [19V, 60V] that would cover the both operating
voltage ranges of the Main Bus and two designs of the system that operates in two different input voltage ranges that correspond to the operating voltage
ranges of the Main Bus. Two architectures and, among them, well known isolated (Half-Bridge, Push-Pull, Forward with Active Clamp (FWAC) and Flyback with
Active Clamp (FLAC)) and non-isolated (Buck and Synchronous Buck) DC/DC topologies are analyzed and compared in terms of efficiency and size, meanwhile
taking care of the cost, design time, reliability and flexibility of the whole system.
Analyzed Solutions
Power Distribution Module  Narrow ??? range for the Converters 2 and 3
 No need for galvanic isolation in the Converter 2
x Converter 1 handles full output power
 Sync. Buck
x Reduced flexibility
 Buck

 Half-Bridge
 Push-Pull
Results of the
 FWAC Architecture A Comparison
 FLAC  Two-Outputs  Semiconductor losses
Flyback comparison was
Architecture B
sufficient to decide on
 Low power converters the Architecture
 Reduced cost and design time
 Increased flexibility
x Wide operating ??? range  Design of magnetics are obtained for
x All the converters need to be Push-Pull, FWAC and FLAC.
galvanically isolated  Total losses comparison is done

Experimental Results

 Specific design for short ??? ranges
shows minor advantage in efficiency
 Experimental results are provided as comparing to the single design of the
CONCLUSIONS a verification of theoretical analysis system that covers both ??? ranges
The goal of this project was to analyze and compare different possibilities for the specific design of the Power Distribution Module for Space Application. Two
architectures and four well known DC/DC topologies (Half-Bridge, Push-Pull, FWAC and FLAC) are analyzed and compared in terms of efficiency and size,
meanwhile taking care of the cost, design time, reliability and flexibiility of the whole system. Semiconductor losses comparison was sufficient to choose
Architecture B as more efficient. Magnetics designs are obtained for FWAC, FLAC and Push-Pull topologies for the purpose of the further examination. FWAC
topology shows the best efficiency for the design of both, Converter 1 and 2, for wide ??? range. Design of the FWAC topology for short ??? range shows minor
improvement in the efficiency (up to 0.68%), comparing to the possibility of having the single design of the system that can cover both operating voltage ranges
of the Main Bus. Additional analysis of the system’s dynamics is considered as a subject of the future work in order to have more clear view on the selection of
the topology and operating input voltage range. Laboratory prototype for FWAC topology for the Converter 1 of the Architecture B is built and measured in order
to validate the presented theoretical results. The prototype’s minimal 92.3% and maximal 93.2% efficiency match pretty good with expected 92.9% and 94%.
Real Time Operating System
With Reconfiguration Support
For Adaptable Satellite Operations
Arturo Pérez García, Leonardo Suriano, Eduardo de la T.

arturo.perez@upm.es, leonardo.suriano@upm.es

Abstract
This work presents an approach to make powerful Commercial-Off-The-Shelf (COTS) devices reliable enough to be used in space applications. New devices
open up a new range of applications never seen in space embedded systems. Integrating Reconfigurable Computing techniques with standard software
can ensure a longer lifetime of satellites working in harsh environments.
The most important characteristic of systems used for space applications is reliability, so the features of RTEMS (a certifiable operating system used in
many space missions) have been expanded to support dynamic partial reconfiguration (DPR) and fault mitigation techniques, such as scrubbing. This OS
can ensure deterministic behaviour in the application, a fundamental requirement for satellites, which are not allowed to work under unforeseen
conditions. Different image processing algorithms are executed by hardware accelerators, which via DPR can be exchanged at run-time.

SRAM-based FPGAs in space Workflow and tools used
New technology for a traditional sector
Rad-hard. devices COTS devices

Traditionally in the space domain, radiation hardening by construction has been used.
New applications, which require very high computing capabilities, cannot be Results
implemented in those circuits, based on old chip generations. Newer devices must be
The starting point
used and different functions and strategies have to be implemented in the design stage to
make them robust enough to face the space conditions. Time (us) Throughput Data size
(MB/s) (MB)
Proof of concept DPR
Read
9999.2 130.189 1.302

20.22 199.98 4.043
Validation application configuration
data
Checking 20.44

These results expose the viability of the proposed
solution where several functionalities are going to be
implemented. The time taken to check and correct
RTEMS is a reliable operating
the configuration memory are the reference to
system which has been used in
develop new strategies with higher performance. The
several space missions. In this
high configurability of RTEMS applications allow
work, it has been utilized to
creating small executables in memory, but the
manage different functions related
different files used to verify the system integrity have
to either specific application
a big impact on the system memory footprint. Those
objectives (mission tasks) or to
files need to be stored in expensive radiation
ensure the integrity of the system
hardened memories, hence fostering research about
(support tasks). An application
having them compressed.
developed with the goal of
checking the feasibility of File Memory Total memory
integrating it in a reconfigurable utilization by one utilization (MB)
device was done. This application is file (MB)
the starting point to develop a Executable 4.3 4.3
more complex system which will be Partial
used in two different space 1.3 3.9
bitstreams
applications: autonomous satellite Golden copies 4.45 26.7
navigation and Earth observation
using hyperspectral cameras. These
applications will take advantage of Conclusions
the intrinsic adaptability of • First implementation of RTEMS in Zynq devices
reconfigurable devices by using with multiple reconfiguration support functions.
DPR strategies, which have been
• RTEMS can be used in reconfigurable platforms to
already implemented in RTEMS.
manage configuration functions.
A flight altitude estimator for multirotor UAVs
in dynamic and unstructured indoor
environments
aHriday Bavle bJose Luis Sanchez-Lopez cAlejandro Rodriguez-Ramos
dCarlos Sampedro ePascual Campoy
a PhD Student at Computer Vision and Aerial Robotics Group, UPM (hriday.bave@upm.es)

Abstract
A reliable estimation of the flight altitude in dynamic and unstructured indoor environments is an unsolved problem. Standalone available sensors, such as
distance sensors, barometers and accelerometers, have multiple limitations in presence of non-flat ground surfaces, or in cluttered areas. To overcome
these sensor limitations, maximizing their individual performance, this paper presents a modular EKF-based multi-sensor fusion approach for accurate
vertical localization of multirotor UAVs in dynamic and unstructured indoor environments. The state estimator allows to combine the information provided
by a variable number and type of sensors, including IMU, barometer and distance sensors, with the capabilities of sensor auto calibration and bias
estimation, as well as a flexible configuration of the prediction and update. Several autonomous indoors real flights in unstructured environments have
been conducted in order to validate our proposed state estimator, enabling the UAV to maintain the desired flight altitude when navigating over wide
range of obstacles. Furthermore, it has been successfully used in IMAV 2016 competition. The presented work has been made publicly available to the
scientific community as an open source software within the Aerostack (www.aerostack.org) framework.

Components of the Proposed State Estimator
Proposed Multi-Sensor Fustion • Aerial Robot: The aerial robot consists of the following states:
State Estimator  Vertical linear position, velocity and accleration.
The proposed state estimator has the following features:  Attitude of the robot (euler angles).
 Modularity in the number of sensors used, that allows  Angular velocity.
for usage of a variable number and kind of sensors. • Ground Object: The ground object state includes:
 Auto calibration of sensors, allowing to estimate the  Altitude of the ground object.
internal state of some sensors. • IMU Sensor: The IMU sensor state includes:
 Synchronous prediction of the estimated state, whenever  Bias on the vertical acceleration measurements.
an estimation of the state is required, the prediction The IMU sensor measurements used include:
stage is executed.  Acceleration measured in vertical axis.
 Asynchronous update of the estimated state based on  Euler angles about x-axis and y-axis.
the available measurements.  Angular velocity about x-axis and y-axis.
• Barometer Sensor: The barometer sensor state includes:
 Bias on the vertical position measurements.
The barometer sensor measurements used are:
 Linear veritcal position.
• Distance Sensor: The distance sensor does not incorporate any state.
Performance of the Proposed The measurements provided by the distance sensor are:
 Distance measured to the ground object.
State Estimator
The proposed state estimator was tested through several
autonomous navigation missions in unknown and
unstructured indoor scenarios.
 The scenarios conisted of several unknown obstacles in
terms of size, height, 3D shape and surface.
 The state estimator provided a robust estimate of the flight
altitude even in presence of several anamolies in the sensor
measurements.
 Additionaly, state estimator was successfully used in IMAV
2016 indoors competition.
A general architecture for autonomous
navigation of unmanned aerial systems
www.vision4uav.com
a,bJose Luis Sanchez-Lopez, a,cMartin Molina, a,bPascual Campoy
aComputer Vision and Aerial Robotics (CVAR), (jl.sanchez@upm.es)
bCentre for Automation and Robotics (CAR), UPM-CSIC
cDepartment of Artificial Intelligence, Technical University of Madrid (UPM)
www.aerostack.org

Abstract
To simplify the use of the micro aerial vehicles and to extend their utilization to a great number of applications, achieving the fully autonomous navigation
of a fleet of aerial robots performing complex dynamic missions in challenging unstructured environments is required.
This thesis presents a system architecture for aerial robotics that defines the required components, their functionalities and their relationships in a generic
and abstract level. This abstract definition enables to use the proposed system architecture in a wide range of applications, ensuring the perfect fully
autonomous behavior.
The system architecture is implemented by means of an open source software framework, Aerostack, to ease the coding, testing, documentation, and
maintenance task to components developers.
The validation of the proposed work is done thanks to the participation in four international aerial robotics competitions, the development of several
challenging experiments, and the exhibition of the complete system in multiple public demonstrations.

System architecture for
Aerial Robotics
- Fully autonomous operation
- Multi-robot
- Layered architecture
- Multi-system architecture
- Ontology
- Hardware agnostic

Software framework:
Aerostack
- Open-source
- Modular
- Multi-process
- Library of Aerial Robotics Components

Search and rescue mission
Fully autonomous fleet of aerial robots to respond to an emulated emergency in
an indoor environment.
Two stages: search and rescue:
- Search: Two exploratory aerial robots navigate from the rescue equipment
base to the inside of the house, entering through an open window or door,
searching for a target.
- Rescue: When the subject is detected, a rescue aerial robot, navigates to the
target’s position, guaranteeing the survival of the subject, interacting with him
by means of natural interfaces.

Video: https://youtu.be/t2mJftbBHWc
References:
[1] J. L. Sanchez-Lopez, J. Pestana, P. de la Puente, P. Campoy. A Reliable Open-Source System Architecture for the Fast Designing and Prototyping of
Autonomous Multi-UAV Systems: Simulation and Experimentation. Journal of Intelligent and Robotic Systems. 2016, vol. 84, num. 1, pp. 779-797.
[2] J. L. Sanchez-Lopez, R. A. Suarez Fernandez, H. Bavle, C. Sampedro, M. Molina, J. Pestana, P. Campoy. AEROSTACK: An architecture and open-source
software framework for aerial robotics. 2016 International Conference on Unmanned Aircraft Systems (ICUAS), Jun. 7-10, 2016, pp. 332-341.
[3] Accepted. J. L. Sanchez-Lopez, M. Molina, H. Bavle, C. Sampedro, R. A. Suarez Fernandez, P. Campoy. A Multi-Layered Component-Based Approach for the
Development of Aerial Robotic Systems: The Aerostack Framework. Journal of Intelligent and Robotic Systems.
COMPUTER VISION FOR SENSE & AVOID
ONBOARD UNMANNED AERIAL VEHICLES

Adrian Carrio and Pascual Campoy
Computer Vision & Aerial Robotics Group, Center for Automation and Robotics (UPM-CSIC),
Calle José Gutiérrez Abascal 2, Madrid, Spain (adrian.carrio@upm.es)

Abstract
Unmanned Aerial Vehicles (UAVs) have recently become a versatile platform for many civilian applications including inspection, surveillance and mapping.
Current regulations require these vehicles to be capable of maintaining self separation for conflict avoidance, providing an Equivalent Level of Safety
(ELOS) to that of human pilots. Sense & Avoid (SAA) systems provide solutions to this problem with the final objective of safely integrating UAVs in our
airspace, which is currently one of the main technological challenges for using these platforms in civil applications. Electro-optical systems, in particular
RGB and thermal cameras, are suitable for onboard obstacle detection, given their low cost, reduced size and light weight. In our research, we propose
solutions for obstacle detection and tracking onboard UAVs using Computer Vision.

Synthetic image generation
Obtaining real flight imagery of potential collision scenarios is hard and dangerous, which
complicates the development of Vision-based detection and tracking algorithms. We have
developed software for synthetic imagery generation, allowing to blend user defined flight
imagery of a simulated aircraft with real flight scenario images to produce realistic images with
ground truth annotations. These are extremely useful for the development and benchmarking of
Vision-based detection and tracking algorithms at a much lower cost and risk.
Related publications
• Carrio, A.; Changhong Fu; Pestana, J.; Campoy, P., "A ground-truth video dataset for the
development and evaluation of vision-based Sense-and-Avoid systems," in International
Conference on Unmanned Aircraft Systems (ICUAS), 2014, pp.441-446, 27-30 May 2014
Fig 1. Real (above) and simulated (below)
• Carrio, A.; Changhong Fu; Collumeau, J.F.; Campoy, P., “SIGS: Synthetic Imagery Generating aircraft intruders in real footage captured
Software for the Development and Evaluation of Vision-based Sense-And-Avoid Systems," in onboard a UAV
Journal of Intelligent & Robotic Systems, 2015, Springer Netherlands

Sky segmentation and horizon line detection
Vision-based sky segmentation can be extremely useful to perform important
tasks onboard Unmanned Aerial Vehicles (UAVs), such as pose estimation and
collision avoidance. We investigate how to perform sky segmentation in RGB
images using a supervised Machine Learning approach by first splitting the image
into fixed-size patches, extracting and classifying color descriptors for each patch
and performing a final post-processing stage to improve the quality of the
segmentation. A method for automatic horizon line detection is also proposed.
The performance of our approach was evaluated on flight images captured
onboard UAVs, achieving performance accuracies above 93% at real-time frame
rates.
Fig 2. Sky segmentation (left) and horizon line detection (right) Related publications
• Carrio, A.; Sampedro, C.; Changhong Fu; Collumeau, J.F.; Campoy, P., "A Real-
time Supervised Learning Approach for Sky Segmentation Onboard Unmanned
Aerial Vehicles," in International Conference on Unmanned Aircraft Systems
(ICUAS), 2016.
Aircraft detection and tracking
We present a lightweight obstacle detection system for small Unmanned Aerial Vehicles (UAVs) combining a Thermal Infrared (TIR) camera and an
Automatic Dependent Surveillance-Broadcast (ADS-B) receiver. Additionally, synchronized visible spectrum (VIS) images have been captured in order to
compare the capabilities of TIR and VIS imaging technologies for aircraft detection in Sense-and-Avoid (SAA) under different lighting conditions. TIR
imaging is proved to be an effective technology for vision-based SAA systems, allowing for operations under extreme illumination conditions such as direct
sun exposure or during the night, outperfoming current vision-based SAA systems.
Related publications
• Carrio, A.; Saripalli, S.; Lin, Y; Campoy, P., "Obstacle Detection System for Small UAVs using ADS-B and Thermal Imaging," in Journal of Intelligent &
Robotic Systems, 2017, Springer Netherlands.

Fig 3. Aircraft detection in a RGB image (left), thermal image (center) and UAV equipped with our obstacle detection system (right)
CHEMOCONTROL

aC. Rossi, aA. Barrientos, aJ. Del Cerro, bG. Marrazza, bA. Ravalli
With the help of to the students Raquel Sanchez and Miguel Gomez

aCentrefor Automation and Robotics (CAR) UPM-CSIC
bLaboratorio Biosensori, University of Florence, Italy

Abstract
In CHEMOCONTROL we investigate the use of chemical messengers for transmitting control signals for artificial muscles made of a high number of
contractile units. We aim at reproducing the activation of motor fibers by chemical signals coming from motor neurons in biological muscles.

Background
• The future generations of service and personal robots will share working and living space with humans.
• They will need to be provided with advanced motor properties, such as the ones exhibited by biological muscles.
• Biological muscles are composed of a high number of contractile units activated in specific patterns. Our working
hypothesis is that such feature provides biological muscles with some of their superior properties.
• In order to reproduce biological muscles’ internal structure, we need to reduce the complexity of the control logic
required to drive a high number of contractile units (fibres). Driving each individual units with classical control
electronics would be prohibitively complex.
• We investigate the use of chemical signals instead. Chemical messengers provide a series of interesting features.
Image credits: wikimedia commons

The messengers
• Chemical signals allow exploiting the physics of the system for control purposes.
• Thanks to spatial summation and temporal persistence, signals are integrated both is space and time without
the need of any computation electronics.
• Spatial diffusion allows addressing a group of fibres (fascicle), reducing the need of wiring each individual fibre.
• Chemical signals are suitable to be used at small scales (millimetre, sub-millimetre) - in fact, the smaller the scale,
the more effective they are.
Proprioceptor
0
Chemical
Pump 2: (stretch sensor)
-0,02
bits
Buffer pH 4 Pump: -0,04
H 2O
-0,06

-0,08

-0,1

Messenger -0,12

Sensor cell 1000

900

800

700

600
Voltage
reading (to 500

analog input) Shape Memory 400

Alloy artificial 300

contractile fibre 200

100
Chemical bits trigger
fibre contraction

CHEMOCONTROL is funded by the Spanish Ministerio
de Economía y Competitividad under the EXPLORA
programme, contract no. DPI2013-49564-EXP
UNEXMIN - UNDERWATER EXPLORER FOR
FLOODED MINES
Claudio Rossi, Sergio Domínguez, Ramón Suárez, Zorana Milosevic

Centre for Automation and Robotics (CAR) UPM-CSIC

Abstract
Many of Europe’s closed underground mines are now flooded and the last piece of information on their status is decades or over a hundred years old. The
complex underground mine layout (mineral extraction was often performed on many interconnected levels) and the complex topology and geometry of
these mines makes any kind of surveying impossible using conventional tethered or remotely controlled equipment. For safety reasons it is almost always
impractical to carry out such work using human divers, as the stability and accessibility of such mine openings cannot be assessed in advance and the
water depth in most mines is considerably greater than the safe limit for scuba diving.

A novel robotic mine surveying system to be used for the autonomous mapping of flooded underground mines

OBJECTIVE S CHALLENGES IMPA C T
• Design and build a Robotic • Localization, Navigation and 3D • Pushing the EU to the forefront
Explorer (UX-1) for autonomous 3D Mapping: autonomous operation in in sustainable minerals surveying
mapping of flooded mines; diffcult heterogeneous three- and exploration technologies;
• Develop and calibrate scientific dimensional tunnel structures; • Increasing Europe’s capacity to re-
instruments for collecting • Scientific instrument design and evaluate its abandoned mines for
mineralogical information; adaptation: optimising their mineral potential, with reduced
• Develop a multi-robot platform that miniaturisation in terms of price, exploration costs and increased
allows the collaboration and task weight and power consumption, investment security for any future
distribution between UX-1robots; whilst providing valuable geological mining operations;
• Demonstrate the operation of the data; • Help to document and
prototype at representative pilot • Explorer structural design: physical safeguard Europe’s unique
sites. robustness, resilience and self- mining heritage.
diagnosis capabilities.

UNEXMIN’s pioneering solution will generate valuable information on underground mine geometry as well as geological data for the
development or update of 3 D regional geol ogical models, replacing much of the nee d for expensive drilling from surface. The multi-robot
platform will link several surveyors together into a distribute d payload system, w here each of the vehicles can carry a different set of
sensors, reducing the size, the weight and the power de mands of the individual robots. This approach will provide security of operations
for the submersibles that will be able to share data in real time, and will also provide scalability for future operations, where larger mines
could be re-explored by a swarm of collaborative robots.

Extensive pilots and demonstration will confirm the capabilities of the Platform. During the pilots UX-1 will be iteratively improve d after
each trial session, which will be increasingly demanding. This iterative piloting activity will last for approximately 16 mont hs, after which
the final, most ambitious demonstration takes place in the U K with the resurveyi ng of the s ubmerge d parts of Ecton mine that nobody has
seen for over 150 years. This final mission will be t he most challenging and most complex endeavour. At this pilot site the entire mine will
be resurveyed, using multiple robots, and all available scientific instruments in order to demonstrate the Platform’s ability to adapt to
the size and complexity of flooded underground environments.
The UNEXMIN Consortium: More information:
· University of Miskolc, Hungary
· Geological Survey of Slovenia, Slovenia
· Tampere University of Technology, Department of Mechanical, Project starting date: 1 February 2016
Engineering and Industrial Systems, Finland Duration: 45 months
Budget: 4,862,865 EUR
· Universidad Politécnica de Madrid, Centre for Automation and Robotics, Spain
· La Palma Research S.L., Spain
· INESC TEC – Institute for Systems and Computer Engineering,
Technology and Science, Portugal
· Resources Computing International Ltd, UK
· Geoplano, Portugal
· Ecton Mine Educational Trust, UK
· European Federation of Geologists, France
· Geo-montan, Hungary WWW.UNEXMIN.EU
· Empresa de Desenvolvimento Mineiro, Portugal This project has received funding from the European Union’s Horizon2020
· Idrija Mercury Heritage Management Centre, Slovenia research and innovation programme under grant agreement No 690008.
Tracking and intercepting pedestrians:
a robotic approach to surveillance of critical infrastructures.
Mario Andrei Garzón Oviedo
Grupo de Investigación Robótica y Cibernética - RobCib.
Centro de Automática y Robótica UPM - CSIC.
ma.garzon@upm.es

Abstract
This dissertation presents an approach for the inclusion of technologies from the field of robotics into security and surveillance of outdoor critical infra-
structures. The system developed is capable of performing a high level task in the context of security systems: the detection, tracking and interception
of an intruder moving around, in an outdoors critical infrastructure.
Different modules have been developed, namely: pedestrian detection, tracking, trajectory prediction, planning for interception and autonomous navi-
gation. All the components are integrated using software and communications architectures also developed in the thesis. This allowed us to use them
in combination with external algorithms or to apply them to different applications. Furthermore, all the algorithms independently and the full system
were intensively tested both in simulations and with real robots in relevant scenarios.

Global objective
Design and implement a robotic system for the
detection, tracking and interception of
pedestrians, that may be considered as
intruders, in an outdoor critical infrastructure.

Lines of work
Single and multiple Detection, localization Predict the future Planning and execution
robot system operation and tracking of trajectory of a of trajectories for
in semi-structured pedestrians pedestrian interception
scenarios

System highlights
On-line processing and execution Strong practical approach Modular design
Any change in the movements of the All the developments have been deployed and Facilitated development and testing of
pedestrian result in a modification of the tested in real robots and relevant scenarios. each one of the algorithms and allows
trajectory of the pursuer robot. their use in different applications.
Cable - Driven Robot to Simulate
Underwater Environments
aAlejandroRodríguez, aRoque Saltarén, aRafael Aracil,
aGerardo A. Portilla, aJuan S. Cely, aCecilia García, aMarco Carpio

aCenter for Automation and Robotics (CSIC-UPM) (aroba2005@gmail.com)

Abstract
The development of motion control algorithms for underwater robots in a real environment is a complex task due to
the limited workspace in pools and the low friction in its floor. Experiments in lakes or oceans have the uncertainty
of external perturbations, turbidity of water and difficult access to the robot. This research line focusses in the
design, construction and control of a cable-driven parallel robot. This is able to simulate, out of the water, the
dynamic present in underwater environment like buoyancy, underwater currents or collisions. The Group has an
underwater humanoid robot (DiverBot) able to walk on the seabed and swim. Those control algorithms are going to
be upgraded using the present cable driven robot able to set controlled underwater conditions in end effector.

Introduction
The end effector of cable-driven robot has 6
controlled degrees of freedom using 8 cables of
flexible steel in a suspended configuration. The
system use hydraulic actuators and it can manage
loads of 150Kg with speeds until 0.9 m/s.
Radio telescopes, accurate cranes, haptic and
medical devices are examples of use of cable robots. Characteristics
Geometry
Many parameters affect the decision of
the geometry: Boundaries in cable
tensions, workspace, condition index,
wrenches required in effector, power or
stiffness.
Taking consideration of all of these
parameters, the ideal dimension for this
application is 10 x 7 x 6 meters.

An
A1
un
l1 u1

ln
u2 A2
Kinematics and dynamics ?1ሶ Bn B1
?1 ⋯ ?1 × ?1 ?0 ?
Obtaining the Jacobian and Wrench ⋮ = ?? ⋯ =? 0
?? × ?? ? ? p1
??ሶ pn
matrix is direct by using Screw Theory. l2
?1 p2
τ: Tension of cable. F and M are applied −? ?1
= ? ×?
⋯ ??
⋮ = ??
−?0 ⋯ ?? × ?? O
forces and torques in end effector. 1 1 ?? B2

Tension configuration Future works
The configuration of all cable tension varies with end effector Other application of cable-driven robotics in
displacement and wrench applied so it is needed to ensure the Group is the connection of ships to the
continuity, bounded tension and avoidance of resonance. The electrical power supply of Spanish ports with
redundant configuration of the robot, provides some freedom the European Project “Master Plan for OPS
where a control in the null space of Wrench matrix is used. ports” (2015-EU-TM-0417 S).
The authors would like to thank the financial support of Spanish Government CICYT Project Ref. DPI2014-57220-C2-1-P, DPI2013-49527-EXP Universidad
Politécnica de Madrid Project Ref. AL14-PID-15 and also to Comunidad de Madrid who supports the project ROBOCITY2030-III P2013/MIT-2748
Force muscle estimation based on
mechanomyography (mmg) sensors
José L. Samper, Miguel A. Sánchez-Urán, Manuel Ferre, Daniel Cordova, Ángel García, Ricardo Ruiz, Fernando
Garnacho
CAR UPM-CSIC (joseluis.samper@upm.es)
Universidad Politécnica de Madrid

Abstract
The purpose of this investigation is determining the suitability and possibilities of applying mechanomyography to exoskeletons. Consequently, a new soft
microphone-based sensor is being designed and built up. This poster introduces some of the results and conclusions that have been achieved during the
initial design stage as well as briefly outlines the main features and keypoints of the mmg sensor. As a consequence of the development stage where the
process is at the moment, it is mandatory to highlight that this mmg sensor does only focus its attention on measuring the MVC (Muscle Voluntary
Contractions) that occur in the bíceps brachii.

Introduction Results & discussion
The term “mechanomyography” refers to the mechanical nature of the The sensor introduced along this poster has been tested in different
muscle sounds that can be measured by different technologies such as subjects whom were told to perform unload MVC as well as bicep curls. The
phonomyography (air-coupled microphones), vibromyography results extracted from those experiments have reveal some important
(accelerometers) or acousticmyography (piezoelectric contact information about the characteristics of the sensor:
sensors)[1][2]. Those sounds emitted by the muscles are inaudible for the
human ear and are normally located at frequencies below 60hz. Some
1. Importance of the material
authors like [2] consider that mmg’s studies were started in 1980s by The structure exposed in the previous section has been produced in three
Oster and Jaffe. different materials: polish aluminium, ABS and resin. The experiments
performed have shown that the crucial feature is the surface finish (coating)
Late in the same decade, it was demonstrated that muscles oscillate of the air chamber’s inner side rather than the employed material.
during twitches, being the individual response of each motor unit added 2. Importance of the positioning
to a vibration signal that propagates through the muscular membranes to
The correct positioning of the sensor shows an intrinsic relation with the
the superficial skin’s layer. [1][2][3]. Furthermore, [1] refers to studies like
accuracy and measurement capabilities of the system. Furthermore, the
the ones from Stokes and Cooper [4] or Kimura[5] that suggest a relation
displacements of the fabric and movements of the sensor during measuring
between the MMG frequency and the motor unit firing rate. On the other
induce artifacts that highly affect the signal.
hand, other researches like [6] support the possibility of extracting
information about the fiber type composition of a muscle from the MMG 3. Sensitivity to external perturbations
Power Density Espectrum and its bands configurations. During experiments it has also been noticed that some external events like
tappings on the skin or air flows over the sensor’s surface could affect the
Design measurements. Consequently, a new design proposal for guaranteeing the
device immunity is under development.
1. Sensor architecture outline Future development
The first prototype of the sensor is based on a commercial condenser Despite of the objectives pursued with this first prototype, the experiments
microphone with an acceptable signal to noise ratio (SNR), whose have shown that our sensor could have enough potential for provoking a
frequency bandwidth moves from 20Hz to 20Khz. Aiming to improve the high impact in exoskeleton’s sensoring as well as in some medical fields.
response of the sensor, It has been designed a case that does not only Consequently, the following design iterations are focused in two main
insulate the microphone from the external noise but does it also amplify branches:
the longitudinal vibrations propagated through the skin result of the
muscle activity. As regards to the materials; several materials has been a) Correction of the mechanical issues that have been discussed above
probed for the structure: polish aluminium, ABS and resin; meanwhile the b) Enhancement of the features of the sensor by improving the signal
insulation has been achieved by a LD-PP Foam covering. processing stages and testing new transducers.

References
[1] T. W. Beck, T. J. Housh, G. O. Johnson, J. T. Cramer, J. P. Weir, J. W. Coburn,
M. H. Malek. “Does the frequency content of the surface
mechanomyographic signal reflect motor unit firing rates? A brief review”.
Journal of Electromyography and Kinesiology 17 (2007)
[2] M. Watakabe, K. Mita, K. Akataki, Y. Itoh. “Mechanical behaviour of
condenser microphone in mechanomyography”. Med. Biol. Eng. Comput.,
2001, 39, 195-201

2. Positioning of the sensor [3] Orizio C. “Muscle sound: bases for the introduction of a
mechanomyographic signal in muscle studies”. Crit Rev Biomed Eng
The placement of the sensor could be considered one crucial matter of 1993;21:201–43.
myography. Hence, matrixes of microphones have been placed at the [4] Stokes MJ, Cooper RG. “Muscle sounds during voluntary and stimulated
positions where the changes in the muscle volume reach its maximum contractions of the human adductor pollicis muscle”. J Appl Physiol 1992.
values. To achieve and secure this placing, the microphones have been [5] Kimura T, Hamada T, Massako Ueno L, Moritani T. “Changes in contractile
sewed to an acoustically insulated fabric which becomes the sensing soft properties and neuromuscular propagation evaluated by simultaneous
fabric that can be seen in the picture below. mechanomyogram and electromyogram during experimentally induced
hypothermia”. J Electromyogr Kinesiol. 2003;13:433–40.
[6] C. Orizio, M. Gobbo, B. Diemont, F. Esposito, A. Veicsteinas. “The surface
mechanomyogram as a tool to describe the influence of fatigue on biceps
brachii motor unit activation strategy. Historical basis and novel evidence”.
Springer-Verlag 2003
Multimodal Vest for Virtual Reality
Applications
aManuel Ferre, aGonzalo García, aJose Breñosa, aDavid Vargas
aCentre for Automation and Robotics (UPM – CSIC)
m.ferre@upm.es, gonzalo.gvalle@upm.es, jose.brenosa@upm.es

Abstract
The development of a multimodal vest for virtual reality applications is focused in increasing the immersion and realism in virtual environments, using
vibrotactile and thermal feedback. Several actuators have already implemented in order to deliver touch-based communication and thermal effects.
Touch-based communication is based on vibration motors, while thermal effects are based on Peltier cells. Other actuation methods will be developed in
near future, achieving a high variety of haptic effects. In this manner, several patterns have been validated for helping users to move and receive stimuli
from virtual reality. Moreover, research is being developed for analysing human torso resolution and perception of vibration and thermal patterns,
evaluating different kinds of actuators and their location on the vest. This research allows to create sensations with great similarity to real sensations.

Background
The development of the multimodal vest is framed on the European Project “AUGGMED” (Automated Serious
Game Scenario Generator for Mixed Reality Training). ‘AUGGMED’ objective is the development of a serious game
platform for training of security forces assisted with virtual or mixed reality interfaces. The haptic vest aim is to
develop a vibrotactile interface in order to convey impact effects (e.g., bullets or explosions), thermal effects (e.g.,
heat or cold) as well as touch-based communication between members of the same training team (e.g., touching a
partner shoulder for confirming presence or for guiding when an explosion has affected lighting and caused
temporary deafness).

Stimuli Types Figure 1: Current vest prototype
Vibrotactile Stimuli
Vibration motors are used to convey vibrotactile effects over different body areas, allowing users
can feel several sensations through touch-based communication. Motors are distributed over
upper torso due to most of touch contacts take place on those areas.
The vest can generate tactile sensations as follows:
Figure 2: Vibrotactile actuators over the vest • Contacts with virtual environment elements as walls, shrapnel, rain, etc.
• Sensations coming from other trainees or virtual characters in order to help users to do a
particular action.
Thermal Stimuli
Thermal effects are simulated through Peltier cells. These actuators are distributed by lower back
and abdomen, achieving a complete temperature sensation throughout the body. Peltier cells can
generate heat and cold sensations with a single actuator, depending on the electronic control:
• Heat produced by a nearby explosion or by hot conditions inside virtual environments.
• Cold produced by the environment or low temperature of some virtual elements.
Other Stimuli
The main objective is the creation of an actuator impact in order to perceive effects like bullet
impacts or hits generated by a shock wave when a explosion is produced near to the user.
Moreover, there are a lot of sensations that can be implemented on the vest, as pressure or soft
sensations. However, there are not any development to convey that information, so it is required
create completely new actuators.
Figure 3: User during tests

Vest for Virtual Reality
Once actuators have been distributed over the selected areas of the vest, next step is
performing several tests in order to determine if the vest is a valid tool for increasing realism
and immersion inside virtual environments. To do that, users wear the haptic vest and a
commercial head-mounted display (HMD) through is reproduced a virtual reality
environment.
first test consist on reproducing a video recorded with a 360º video camera where several
touch contacts were simulated with the help of the vest, as sand impacts, taps or hugs. The
second test consists on a virtual environment created by ‘AUGGMED’ partners, the vest
simulate touch contacts with virtual reality elements (characters or walls) or thermal effects
when the virtual character approaches to a fire. Those tests are used for validating the
system as a virtual reality tool for increasing immersion and realism, allowing users feel the
environment in a completely real way. Figure 4: Virtual environment used during experiments
Advanced Telemanipulation Method for
Scientific Nuclear Facilities
Manuel Ferre Pérez; José Manuel Breñosa Martínez; Sofía Coloma Chacón; Luis E. Rubio Rodríguez.
CAR UPM-CSIC (joseluis.samper@upm.es)
Universidad Politécnica de Madrid

Abstract
This project is focused on teleoperation of robots specialized in manipulation of heavy loads inside environments with radiation. The main challenges of
this type of task are, the limitations that arise in force reflection, constraints in sensor use and the convenience of using hydraulic actuators. Environments
with radiation are mainly seen in big installations, such as Nuclear Fusion Plants (JET or ITER), or Research Centres with particle accelerators like CERN or
GSI.
The tasks described in this project have the objective of improving methods about information displayed to the operator, in a way so as to enhance the
most adequate information for the execution of the task. Considering force reflection, an initial hypothesis is that manipulation of heavy load can be
controlled based on the events that occur during manipulation, and that it is not necessary to reflect a force proportional to the one exerted by the robot
to the operator. Moreover, it is considered that the correct combination of haptic feedback or force feedback with visual information is complementary in
order to manipulate heavy loads. Furthermore, works related to the control of binocular cameras that improve visualization of stereoscopic images are
also included in this proposal. Another challenge that arises in the application of manipulation of heavy loads is the use of hydraulic actuators, which have
better weight–power ratio compared to electrical actuators, but with more complex control due to non-linear effects in its behaviour.

Rate-Position Haptic Controller for
Telemanipulation in Large
Workspaces
This work is focused on bilateral telemanipulation. It presents a novel
haptic rate-position controller. This controller has been design to
overcome the problems related to the control of a slave with large
workspace using a small haptic device during bilateral telemanipulation.
The rate-position controller can be used to couple kinematic dissimilar
master-slave systems. The effectiveness of the proposed method
increases when the workspace ratio (slave to master) is greater than the
unit.
It is able to swap between control modes automatically. It uses haptic aids
to inform the user about changes in the control mode. This automatic
approach presents advantages against pressing a button or a pedal to
switch between modes, reducing the cognitive load of the operator and
making teleoperation more natural and intuitive for the user.
It allows performing tasks in a large remote environments using a haptic
device with a reduced workspace. The rate-position controller is
evaluated through two case studies. First, a welding and cutting task is
emulated. The user has to follow the curve line with the maximum
accuracy possible. Second, the telemanipulation of fragile objects in
unstructured environments is studied. The user has to grasp a light bulb,
transport it to a designated point and place it back in the initial position.
These two case studies require accurate and precise movements, which is
Results & discussion
an excellent manner to judge the performance of the proposed algorithm. A welding task has been emulated following a bend track with the slave
The results obtained show that the accuracy is improve and the time to robot. The same experiment has been carried out several times using both
task completion is similar when comparing the rate-position controller the proposed rate-position algorithm and classic position controller.
against alternative methods such as position only or indexing control. Analyzing the result obtained, the rate-position algorithm has proven to be
65:36% more accurate than the classic control algorithm. It is due to the fact
that two modes are defined (position and rate control) to control the entire
workspace. However, in terms of time, the classic position control seems to
be slightly faster since it does not need to change between operation
modes.
Experimental results performing dexterous manipulation of a fragile object,
a light bulb, with different control methods have shown that by using the
rate-position control method, the time to task completion is reduced and
the accuracy of the system allows for full dexterity and correct completion
of the task. This has been compared with state-of the-art methods such as
indexing control and the use of controllers supplied by the manufacturer of
the robotic hand and the slave robot.
Patent
M. Ferre, R. Aracil, J. Barrio, and F. Suárez-Ruiz. Método de guiado de
dispositivos robóticos teleoperados para alternar modo de control en
posición y modo de control en velocidad. 2011.
How affect the biosensing performance of Resonant
Nanopillars arrays under dry condi;ons or immersed in fluid
Hernandez, Ana.L1 ; Rafael, Casquel1; Lavin, Alvaro1; Quintero, Sergio1; Sanza, FJ2; Canalejas,
Victor1; Holgado, Miguel1,2; Lopez-Romero, David2; Laguna, MariFe1,2; Maigler, Mavi1,2
1 Centre for Biomedical Technology. UPM. Campus de Montegancedo. Pozuelo de Alarcón, 28223,Madrid

Spain
2Bio Op3cal Detec3on. Centro de empresas de la Universidad Politécnica de Madrid.

Pozuelo de Alarcón, 28223Madrid Spain

Abstract
In this work we demonstrate, for the first 3me, the advantage of performing the biosensing process of a refrac3ve index op3cal biosensor under dry
condi3ons, in comparison with the biosensing structure immersed in fluid. We developed a biosensing experiment over a specific transducer based on
resonant nanopillars (R-NPs) arrays. The op3cal interroga3on to monitor the recogni3on events was firstly performed with the R-NPs in dry, only in
contact with the air, and secondly with the R-NPs immersed in water. We observed a significant enhance in the sensi3vity of the biosensing curve
response for the R-NPs in dry condi3ons, leading an improvement of the Limit of Detec3on (LoD) in more than one order of magnitude. According to this
result, any interferometric op3cal bio-transducer for in-situ diagnosis will improve the sensi3vity in case it can operate in dry condi3ons.

1 Introduc;on 2 Biofunc;onaliza;on of R-NPs
A

Biosensor layout and scheme response: R-NPs are based on Si3N4/SiO2 RN-Ps were biofunc3onalized with IgG through a silaniza3on
Bragg Reflectors with a SiO2 cavity and they are distributed in 8 different arrays called BICELLs protocol
along substrate

Piranha activation of R-NP
Op3cal response of R-NP R-NP Cleaning of R-NP surfaces
Chip with BICELLs A surfaces
nm
resonance mode
wavelength (nm)

14
B Antibody A P T MS
immobilization
wavelength (nm)

B Silanization 13 Ig G
A 12
a Ig G
TOPSIDE
OH OH Drop of1 150 !g/mL IgG

OH
in buffer solution
Reflectors
NH2
10

Δ wa ve numbe r (nm)
OH OH OH OH
9
Bicell +
OH

CH3O Si OCH3 8
OH

OCH3
7
14
OH

B 5mm
13
A P T MS
Ig G
6
OH

TOPSIDE 1 μm 12
a Ig G
5
Incubation of antibody
R

over
4 R-NP
B

11
10 3
D
Δ wa ve numbe r (nm)

reflectivity (au)
9

3 Characteriza;on of R-NPs
2
8
1

4 Results
7
B IC E L L 11 B IC E L L 12 B IC E L L 13 B IC E L L 14 B IC E L L 21 B IC E L L 22 B I
6
5
4

A_er biofunc3onaliza3on, R-NPs were ver3cally interrogated firstly
3

C 2
Recogni3on of increasing
immersed in water (3A) and secondly in dry condi3ons (3B) with a 1
concentra3ons of an3-IgG
bundle of fibers connected to a spectrometer (3C) B IC E L L 11 B IC E L L 12 B IC E L L 13 B IC E L L 14 B IC E L L 21 B IC E L L 22 B IC E L L 23 B IC E L L 24

were monitored as a
func3on of the wavelength
posi3on of the resonance,
showing a higher shi_ of the
resonance mode when the
pillars are interrogated in
dry condi3ons

Higher shi_ implies higher
sensi3vity (m) and thus
lower LoD= U/m

Bulk signal show the
resonance posi3on is
shi_ed to a greater 4 Conclusion
wavelength when
pillars are immersed, In this work we measured one same sample, with the
also the intensity of
the signal is
same measuring system but in two different media:
decreased and water and air. It is demonstrated a reduc;on of the
therefor the
uncertainty (U) is
limit of detec;on in more than one order of magnitude
higher for R-NP when R-NP are interrogated in dry condi3ons; therefore,
immersed in water
this is a first approach of R-NPs to Point of Care devices
for in-situ diagnosis.
Fabrication and characterization of a label-free optical
biosensor based on SiO2 resonant microstructures
aS.
A. Quintero, F. J. Sanza, M. Holgado. D. López-Romero, M.F. Laguna,
Ana López, R. Casquel, R. López, A. Lavin. B. Santamaría
aCenter of Biological Technology, UPM (andres.quintero@ctb.upm.es)
bEscuela Ténica Superior de Ingenieros Industriales, UPM

Abstract
In this work we present the fabrication of a label-free optical biosensor based on SiO2 microstructures lithographed on a silicon (Si) substrate.
The fabrication of the microstructures was performed by optical contact lithography, using the commercial photoresist AZ5214E and a selective chemical
attack with Buffered Hydrofluoric Acid (BHF). It was studied several parameters of the fabrication process.
Topology characterization of the structures was fulfilled by both optical and confocal microscopy and SEM.
Finally, an Immunoglobulin G (IgG) and its antibody (aIgG) immunoassay was performed to evaluate the use of the microstructures as an affinity biosensor,
using vertical interrogation techniques in order to measure their optical response in the visible range.

Fabrication process

Topology characterization at CLUPM
Fabry-Perot interferometers Arrays of micropillars Arrays of microholes
Chip Mask

Confocal 10x Confocal 100x Confocal 100x
Chip fabricated

Confocal 100x SEM SEM SEM

Inmunoassay IgG-aIgG Conclusions:
Biofuncionalizatio • It was studied a first aproximation into the
n Sequence: microfabrication of structures in SiO2.
• Cleaning • SiO2 microstructures has potencial in the
• Activation of development of biosensors.
the surface
with piranha
Future lines:
• Silanization
with APTMS & • Improvement in the aspect ratio of the
glutaraldehide Vertical interrogation structures.
detects a shift in the • Fabrication of tinier structures.
• IgG
resonance peak in the • Deep study of the biofuncionlazation process at
inmovilization
recognition process the SiO2 structures
• aIgG
recognition • Biosensing of virus and bacteria.
Developing optical label-free Point-of-Care
biosensing devices
a,bM. Victoria Maigler, aM. Holgado, aA. Lavín, bM. Vicente, bM. Domingo,
bF.J. Sanza, aM.F. Laguna, aB. Santamaría, aR. Lopez, bD. López-Romero

a Center forBiomedical Technology, Optics, Photonics and Biophotonics Lab
bBio Optical Detection. Centro de Empresas de la UPM

(m.maigler@biod.es)

Abstract
The aim of this work is to develop a compact point-of-care device as a proof of concept of a novel read-out methodology than enhances
the biosensing response by improving significantly the LoD and makes the interrogation process simple and more reliable. The measured
signal is called Increase Relative Optical Power (IROP) and is a function of the optical power or irradiance for a certain optical interrogation
band of two interferometers (reference and signal). The functionality of this device is validated by performing standard immunoassays and
comparing these results with high resolution spectrometry.
Point-of-care device description
A is the area by the reference chip. C is
Theoretical simulation the area defined by the signal chip
when the biofilm thickness is 0 nm. B
IROP signal quantifies the accumulation or recognition
represents the shift in the signal
of biomolecules onto the sensing surface and is given interferometer for a biofilm thickness
by the optical power of two different interferometers of 40 nm.
(IF) - reference and signal, for a given spectral range.

Read-out device

Bio-Kits
The signal IF is based on Fabry-Perot
Interferometers with thin layers of
SU8 and nitrocellulose. The chip
surface is bio-functionalized to
detect the target biomolecule in an
specific way. The chip is set in a glass
substrate and the sample is drop in a
well defined with a vinyl sticker.

Comparative results: Anti-IgG_IgG immunoassay results
In order to compare the device sensitivity, a standard immunoassay was
performed, based on the conjugated Anti-IgG_IgG. The antibody is
immobilized onto the sensing surface by direct absorption, incubating the
sample (50 μg/mL) during 12 hours at 37 degrees avoiding evaporation. The
blocking step is next (BSA- 3%) to avoid unspecific bounding of biomolecules.
Lately, several concentration of IgG were incubated onto the chip surface for
20 minutes at 37 degrees to obtain the calibration curve and the correlation
shown in the Figure.
Results are compared with a FT-VIS-NIR high-resolution spectrometer
Vertex 70 which experimental sensitivity is estimated in 29.2 cm-1/mg mL-1
and a LoD of 277.7 ng mL-1 . However, for the read out device described in
this work, the resolution is estimated in 1.92 %ΔIROP/mg mL-1 and the
LoD for the same transducers and for the same immunoassay resulted in a
value of 25.5 ng mL-1. Therefore, the IROP methodology significantly
enhances the LoD in more than one order of magnitude.
Conclusion
The novel read-out methodology (IROP) presented in this work will allow to develop cost-effective, compact and
easy-to-use Point-of-care devices with better performance in terms of LoD and time in comparison with high
resolution spectrometry. The PoC based on IROP provides a quantitative result in only 2 minutes with 50 μL of sample
for the Anti-IgG_IgG immunoassay and reaches a LoD of 25,5 ng·mL-1. This biosensing system is therefore feasible to
be applied for ideal and reliable PoC devices.
Detection of CST4, MMP9 and S100A6 biomarkers for Optics
Photonics
dry eye disease on an integrated optical label-free Biofotonics

biosensor
a*B. Santamaría, aA. L. Hernández, aR.l. López, bM.V. Maigler, bY. Ramirez,
bF. J. Sanza, aR. Casquel, aA. Lavin, aM. F. Laguna, aM. Holgado.
a Center for Biomedical Technology UPM (*beatriz.santamaria@ctb.upm.es)
b Bio Optical Detection S.L.

Abstract
Dry eye disease (DES) affects in tears and changes its chemical composition. Due to this variation, a number of biomarkers have been
detected related to this pathology. In the market there are some diagnosis tests to detect this biomarkers but they use an expensive
technology and they do not include a multiplex diagnostic. In this work an optical label –free biosensor is used to detect CST4, MMP9 and
S100A6 biomarkers and their quantitative concentration, all into one integrated kit. The read-out method used for the detection reduces
the cost of this test and the sizes of the devices, becoming a cost-effective technology.

Integration of 3 biomarkers on a label-free biosensor
Scheme of the sensor and read-out technique
The sensor is fabricated with an specific layers related with the measurement requires.

Biofunctionalization of the sensor integrating 3 bio-receptors
Every cell is biofuntionalizated with a concrete antibody specific to the protein that is wanted to detect.

Detection of the proteins:
biomarkesrs for DES
Interaction antibody-antigen.

 MMP9
 S100A6
 CST4
Results

Conclusions
• Biofuntionalization is optimized
• There is cross-reactivity
• Best affinity is between S100A6
antibody-antigen.

Future research lines
• Avoid cross-reactivity
• Measure real samples
• Obtain stable measurements

For more information
Biochemical assay development for optimization Óptica
Fotónica
of immunological system inhibitors Biofotónica
aR.López*, bY.
Ramírez, bM. V. Maigler, aA. L. Hernández, bF. J. Sanza,
bD. López-Romero, aR. Casquel, aA. Lavin, aM. F. Laguna, aM. Holgado.

a Center for Biomedical Technology UPM (*rocio.lopez@ctb.upm.es)
b Bio Optical Detection S.L.

Abstract
The aim of this work was to adapt the BiOD Point of Care technology in order to evaluate immunologic system inhibitors through a chemical assay that is
able to identify the inhibition ability of new drugs for treatment of autoimmune disease. For that purpose, the GST protein, which behaves as an abnormal
T-cell, was immobilized onto Biokits based on label-free sensors. The GST protein has been optimized varying incubation times and incubation
concentrations to obtain a homogenous surface.
The read-out signal for GST, not being attacked for the drug, serve as a positive blank and comparing the response between GST and GST blocked by the
inhibitors, it is possible to obtain a dose-response curve for each inhibitor. The proposed assay allows evaluating different types of inhibitors with good
results.

Label-free optical biosensor: Assay Model:

Inhibitor Behaviour:

Optical Read-out based on Increase Relative
Optical Power (IROP)

Results

GST Protein Optimization (Positive Blank): Dose-Response Curve:

Conclusions:
• The assay methodology to obtain a dose-response curve has been developed
• The accuracy and reliability of the device and Biokits have been demonstrated in order to test the
inhibition ability of the new drugs.
For more information
Local blood pressure estimations: The dependence
of the subharmonic resonance on ambient pressure
aJ. Jiménez-Fernández
aGrupo de Mecánica de Fluidos aplicada a la Ingeniería Industrial.
Dpto. Ingeniería Energética, (jajimenez@etsii.upm.es)

Abstract
aCorresponding author affiliation (contact e-mail)
bAffiliation2
The knowledge of the local blood pressure provides important information regarding the diagnosis of pulmonary hypertension, cardiac abnormalities
and other severes diseases. In recent past, research effort has been devoted to the development of the clinical techniques for noninvasive local blood
pressure measures by means of encapsulated microbubbles driven into oscillatory motions by the action of ultrasound fields. Analysis of the
subharmonic component in the scattered signal, has been proposed as a feasible method to local pressure estimations. In this work the dependence of
the subharmonic response on ambient pressure has been theoretically investigated

The results obtained show
that as the ambient
overpressure is increased,
all harmonic components
are displaced to higher
frequencies. This
displacement is significant
for the subharmonic of the
order 1/2

The influence of the ambient
pressure on resonance
frequencies has been analyzed. It
has been found that the
subharmonic resonance
frequency is very sensitive to
ambient pressure variations and,
in the physiological range,
increases linearly as the ambient
overpressure is increased. This
linear relation suggests further
experimental work
to establish an alternative
procedure for ambient pressure
estimations.
Design of an Aspiration Thrombectomy Device
for cerebral vessels

Carlos Talayeroa, Gregorio Romerob*, Gillian Pearcec
a IndustrialEngineering Department - Universidad Europea de Madrid (PhD. student)
b Mechanical Engineering Department - Universidad Politécnica de Madrid [*contact person: gregorio.romero@upm.es]
c School of Engineering and Applied Science, Aston University - United Kingdom

Abstract
Aspiration Thrombectomy method is being used for recanalization of occluded arteries, especially the basilar (intracranial, in the Circle of Willis) and in
the internal carotid (extracranial). During suction, the distal section of the catheter is located close to the thrombus and a negative pressure difference is
generated resulting in an aspiration inside the catheter. In case of exit, the suction leads to an immediate recanalization. By using simulation models,
new geometries have been designed with a better behavior than existing solutions.

Stage 1: problem definition
• Occluded arteries in cerebral areas due to blood clots
• Aspiration Thrombectomy  no rigid parts and minimize
the risk of embolism upstream (suction of the rests)
• Thrombus extraction can be improved by changes in the
catheter geometry [1]

Stage 2: 3D geometry and models
• Cylindrical tube, Ø 3mm, different
interference conditions. Isotropic material.
(vessel).
Stage 3: CFD + Non-linear analysis • Cylindrical bar 10 mm and 30 mm length.
• Internal flow analysis Hyperelastic material (thrombus).
• 30 KPa aspiration • “Balloon effect” when suctioning desired.
• Initial velocity = 0 m/s
• Variable resistive force (vessel - thrombus contact).
• Pressure and shear loads imported from de CFD results.
• Different catheter materials: PTFE, PLA, silicone, Ninjaflex™

Stage 4: testing
• Scale 3D printed prototypes to validate the analysis Pressure field for full and partial occlusion
• Colored jelly. Different sections tested
• Pressure measured when suction
• Qualitative analysis of the thrombus
• Real scale model accuracy achieved (Ø 3mm)
Conclusions
• Better results for perimetric holes solution: Displacements plot for elastic catheter when suctioning
1. Better movement initiation  loads applied on the vessel – thrombus interface
2. Better disaggregation when the catheter – thrombus contact happens
3. Better guidance through small diameter arteries
• Balloon effect:
1. Problems for 3D printing prototypes. “Two pieces” catheter joined with
polyethylene film used in testing
2. It gives no significant improvement in simulation and in experiments if the
catheter – thrombus distance is lower than 1 mm
3. Low stiffness on the distal end for a correct guidance due to the required thickness
(0.005 mm)
References Experiment sequence and 3D printed prototypes
1. Talayero, C; Romero, G et al (2016). Optimization of an Aspiration Thrombectomy Device (TAD) for using in cerebral vessels. ETSII, UPM. Industriales Research Meeting
2016. p 59
2. Romero, G. et al. (2010). Computational Modeling of a New Thrombectomy Device for the Extraction of Blood Clots. AEMB Vol. 680. Ed. Springer. pp. 627-633
3. Pearce, G. et al. (2010). In Vitro Testing of a New Aspiration Thrombus Device. J Stroke Cerebrovasc Diseases 19(2), pp. 121-129
4. Romero, G. et al. (2013). An investigation into the performance of a new Mechanical Thrombectomy Device using Bond Graph modelling: application to the extraction
of blood clots in the middle cerebral artery. Simulation: Transactions of the Society for Modelling and Simulation International, 89(3), pp. 381-391
Maceration process of fruits in liquor: 
Maceration process of fruits in liquor:
Spectroscopic studies of the kinetics
Isabel Paz, Carmen Matías, Ascensión Fernández, Gabriel Pinto

Departamento de Ingeniería Química Industrial y del Medio Ambiente,
Universidad Politécnica de Madrid (isabel.paz@upm.es)

Abstract
The effect of temperature on the kinetics of pigment extraction during the maceration of different fruits (cherry,
(cherry raspberry,
raspberry blackberry…)
blackberry ) into hard spirits
was evaluated. The analytical method used was UV‐Vis spectrophotometry. Initial extraction rates showed an Arrhenius‐type dependence. Studies about
the evolution of the colour ((from colourless to reddish)) duringg the soakingg p process was done byy calculatingg the CIE tristimulus values ((X,, Y,, Z)) for
illuminant C, until reaching the apparent stabilisation of colour, which occurs after about two to four weeks for the studied temperatures (5 to 40°C).

Introduction
 Fruits liquors are popular hard spirits in several countries, such as Spain (e.g. licor de
cerezas and pacharán). There is an increasing industrial production of these drinks made
by soaking different fruits (cherries,
(cherries strawberries,
strawberries dry figs,
figs and others) in several spirits,
spirits
where the ethanol softens the fruits and promotes the extraction of pigments and
substances
bt th t give
that i tot the
th liquor
li a characteristic
h t i ti colour,
l fl
flavour and
d smell.
ll
 The colour evolution duringg the soakingg pprocess is of major
j interest since it is related
to organoleptic properties evolution. Knowledge of the changes that pigments of fruits
undergo with processing is important with respect to their role in colour quality because
the appearance of a food product can greatly influence a consumer’s purchasing
decision
decision. 1
Abssorbaance

Soaking time (h)
M i bj i
Main objectives 0,8 2664

 To study experimentally the influence of temperature 1320
on the kinetics of pigment extraction during the 0,6 384
maceration
ti off fruits
f it into
i t hard
h d spirit.
i it 216
 To ggain insight
g into the colour evolution of the p
process. 0,4 96
48
0,2 24
Note: Dark red colour in fruits studied arises from pigments called
anthocyanins (Greek: anthos, flower; kyanos, blue), a family of phenolic
phytochemicals that give flowers, fruits and leaves of some plants their 0
red, blue and purple colours. They consists of sugar molecules bound to a 400 500 600 700
benzopyrylium salt (canthocyanidin), and it is part of the flavonoid family.
Wavelength (nm)
Wavelength (nm)

Conclusions 120
ulus vvalue Z

 UV‐Vis spectrophotometry is a good way for studying
90
the kinetics of anthocyanin pigments extraction during
the maceration of fruits in liquor.
q For the range
g of 5 ºC
Triistimu

studied temperatures this process has apparent energy 60
activation (in kJ/mol) of: 28.8 28 8 (raspberry),
(raspberry) 55.6
55 6 23 ºC
C
(cranberry), 69.8 (blackberry), and 78.2 (cherry). 30
 All these values of activation energy were calculated 50 ºC
through the Arrhenius equation with an excellent 0
0 50 100
precision.
Time (days)
 The apparent stabilization of colour for liquors takes Related publications
place in practice in a few weeks, as it was assessed by ‐ I. Paz, G. Pinto, “Spectroscopic Study about the Kinetics of the Anthocyanin
means off the
th determination
d t i ti off theth tristimulus
t i ti l values
l X
X, Pigments Extraction
E traction During
D ring the Maceration of Cherries in Liquor”,Liq or”
Spectroscopy Letters, 35, 357‐368 (2002).
Y and Z.
‐ G.
G Pinto,
Pinto I.
I Paz,
Paz “Study
Study of the Evolution of Colour During the Maceration
 Studies about the evolution of colour in the soaking Process of Cherries in Liquor: an Application of Visible Spectroscopy”,
process of this kind of fruit liquors can lead to a better Spectroscopy
p py Europe,
p , 16 ((1),
), 12‐16 ((2004).
)
understanding of this process, and thus to a better ‐ I. Paz, A. Fernández, M. C. Matías, G. Pinto, “Effect of Temperature on the
Evolution of Colour Duringg the Maceration of Fruits in Liquor”,
q Czech Journal
control over the mechanisms underlying it. it of Food Sciences, 32, 90‐95 (2014).
Science Technology and Art
Science, Technology and Art
Gabriel Pinto, José Vicente Alonso,
G bi l i é i l
Francisco Díaz, Víctor Manuel Díaz
Grupo de Innovación Educativa de Didáctica de la Química,
U i
Universidad Politécnica de Madrid (gabriel.pinto@upm.es)
id d P li é i d M d id ( b i l i @ )

Abstract
In the
h last
l years, we have
h d l
developedd different
diff artistic
i i activities
i i i in
i order
d to promote STEAM (science,
( i technology,
h l
engineering and mathematics relations, together with art) approaches, with increasing importance in educative programs.

Art to learn STEM…
• We
W presentt ideas
id to
t incorporate
i t discussion
di i about
b t
paintings, sculptures, cinematography, postage
stamps, and other works of art in STEM classrooms
and outreach events.
• Through suggested activities, students and general
public are encouraged to discuss a wide variety of
topics (nomenclature, science phenomena, applied
physics
h and
d chemistry,
h h
historical
l contexts, materials,
l
stoichiometry, ethics of science and technology…).

… and STEM to
and STEM to make art
• We have developed different works of art based on aspects of science.
• Examples: paintings, sculptures (evoking the Newton´s experimentum
crucis, crystallization of salts…), 
i t lli ti f lt )
cyanotypes (for discussion of 
physics and chemical principles 
of photography), graphic design,  
p g p y), g p g ,
and scientific photography.

Acknowledgments
We recognize supports provided by Universidad 
Politécnica de Madrid (“Promoting Experiential 
Learning of Chemistry” project) and by 
Fundación LaCaixa (“Science and Technology for 
Everything and for Everyone” project).
Everything and for Everyone project)
Related publications
‐ G. Pinto, “A Postage Stamp About the Periodic Table”, J. Chem. Educ., 84, 1919 (2007).
‐ G. Pinto, “A Postage Stamp Honoring Marie Curie: an Opportunity to Connect Chemistry and History”, J. 
Chem. Educ., 88, 687‐689 (2011).
( )
‐ G. Pinto, “An Example of Body‐Centered Cubic Crystal Structure: The Atomium in Brussels as an Educative 
Tool for Introductory Materials Chemistry”,  J. Chem. Educ.,
Tool for Introductory Materials Chemistry J Chem Educ 89, 921‐924 (2012).
89 921‐924 (2012)
‐ G. Pinto, A. Garrido, “Ciencia y Arte: las Pinturas de los Calendarios MAXAM (antes Unión Española de 
Explosivos) como Recursos para la Difusión y la Enseñanza de la Química”,
p p y An. Quím., 111, 104‐108 (2015).
‐ F. Díaz, J. V. Alonso, J. Ramírez, F. Sotres, G. Pinto, “Dispersión de la Luz a través de un Prisma: Una
Experiencia Artística y Divulgativa, Rev.  Esp. Fís., 29(2), 41‐44 (2015).
‐ G. Pinto, A. Garrido‐Escudero, “Chemistry
G Pi A G id E d “Ch i and Explosives: an
dE l i A
Approachh to the
h Topic
T i through
h h an Artistic
A i i
and Historical Contribution made by a Spanish Global Supplier”, J. Chem. Educ., 93, 103‐110 (2016).
‐ G. Pinto, M. L. Prolongo, J. V. Alonso, 
G Pinto M L Prolongo J V Alonso “Química Química y Física de Algunos Efectos Especiales en Cinematografía: 
y Física de Algunos Efectos Especiales en Cinematografía:
Una Propuesta Educativa y para la Divulgación”, Rev. Eureka  Enseñ. Divulg. Ciencias, in press.
‐ G. Pinto, “Antonio de Ulloa and the Discovery of platinum: An Opportunity to Connect Science and History 
through a Postage Stamp, J. Chem. Educ., in press.
Developing STEM educative 
p g
and outreach tools
a Gabriel Pinto, b Marisa Prolongo

a, b Grupo de Innovación Educativa de Didáctica de la Química,

Universidad Politécnica de Madrid (gabriel.pinto@upm.es)
(g p @ p )
b Centro de Ciencia Principia (Málaga) 

Abstract
During the past three decades we have developed contextualized experiments and cases related to real‐life
real life applications to investigate a good variety of
STEM (Science, Technology, Engineering and Mathematics) topics in the context of inquiry‐based learning. Recently this project was selected to be
presented at the Science on Stage
p g European
p Festival ((Debrecen,, Hungary,
g y, June 2017).
)

Main objectives:
Main objectives:
• to encourage active and critical thinking skills of STEM subjects
in students
d off allll educative
d l l and
levels d generall public;
bl
• usingg contemporary
p y teachingg tools ((PBL,, inquiry‐based
q y
education, interdisciplinary approaches, active learning...);
• to
t promotet effectiveness
ff ti i the
in th learning
l i process;
• to improve
p students’ and teachers’ motivation;
• and to promote science literacy and social responsibility.

Examples:
p
• cooling of water in porous ceramic pitchers (pot‐in‐
pott refrigerator,
fi t interpretation
i t t ti off related
l t d phenomena
h
such as the “drinking bird”, mathematics...);
• stoichiometry and every day life (medicaments,
dentifrices fertilizers,
dentifrices, fertilizers mineral waters,
waters solar power…);
power );
• thermo‐chemistry and condensing boilers;
• ice melting rate in various environments;
• cars fuel consumption vs.
m)

vs CO2 emission;
on (g/km

300
• spattering of hot cooking oil with water;
CO2 eemissio

• self‐heating / self‐cooling beverages; 200
• osmotic rehydration of legumes;
• chemistry of moth repellents; 100
4 9 14
• information in labels of commercial products. Gasoline consumption (L/100 km)

Acknowledgments
We recognize
g support
pp provided byy Universidad Politécnica de Madrid ((“Promotingg Experiential
p p Learningg of
Chemistry” project) and by Fundación LaCaixa (“Science and Technology for Everything and for Everyone” project).

Related publications (more papers and full‐texts: https://www.researchgate.net/profile/Gabriel_Pinto3)
Related publications (more papers and full‐texts: https://www researchgate net/profile/Gabriel Pinto3)
‐ J. I. Zubizarreta, G. Pinto, “An Ancient Method for Cooling Water Explained by Means of Mass and Heat Transfer”, Chem. Eng. Educ., 29, 96‐99 (1995).
‐ G. Pinto, “Using Balls from Different Sports to Model the Variation of Atomic Sizes”, J. Chem. Educ., 75, 725‐726 (1998).
‐ G. Pinto, A. Esín, “Kinetics of the Osmotic Hydration of Chickpeas”, J. Chem. Educ., 81, 532‐536 (2004).
‐ G. Pinto, “Chemistry of Moth Repellents”, J. Chem. Educ., 82, 1321‐1324 (2005).
G Pi “Ch i fM hR ll ” J Ch Ed 82 1321 1324 (2005)
‐ G. Pinto, “Stoichiometry of Calcium Medicines”, J. Chem. Educ., 82, 1509‐1512 (2005).
‐ G. Pinto, M. T. Oliver‐Hoyo, “What is in your Bottled Water?, Look at the Label!”, Chem. Educator, 13, 341‐343 (2008).
G Pi t M T Oli H “Wh t i i B ttl d W t ? L k t th L b l!” Ch Ed t 13 341 343 (2008)
‐ G. Pinto, C. V. Gauthier, “Spattering and Crackle of Hot Cooking Oil with Water”, J. Chem. Educ., 86, 1281‐1285 (2009).
‐ G. Pinto, 
G Pinto “Fluorine
Fluorine Compounds and Dental Health: Applications of General Chemistry Topics
Compounds and Dental Health: Applications of General Chemistry Topics”, J. Chem. Educ., 86, 185‐187 (2009).
J Chem Educ 86 185 187 (2009)
‐ M. L. Prolongo, G. Pinto, “Las Bebidas Autocalentables y Autoenfriables como Recursos para un Aprendizaje Activo”, Educ. Quím., 7, 4‐14 (2010).
‐ G. Pinto,
G Pinto “An An Example of Body‐Centered Cubic Crystal Structure: The Atomium in Brussels as an Educative Tool…
Example of Body Centered Cubic Crystal Structure: The Atomium in Brussels as an Educative Tool ”, J. Chem. Educ.,  89, 921‐924 (2012).
J Chem Educ 89 921 924 (2012)
‐ G. Pinto, M. L. Prolongo, “Una Propuesta de Divulgación Científica: Actividad Festiva para Niños… “, An. Quím.,  108, 344‐351 (2012).
‐ G. Pinto, 
G Pinto “Termoquímica
Termoquímica de las Calderas Domésticas de Condensación: Un Caso de Aprendizaje Contextualizado…
de las Calderas Domésticas de Condensación: Un Caso de Aprendizaje Contextualizado ”, Educ. Quím.,
Educ Quím 14, 29‐38 (2013).
14 29‐38 (2013)
‐ G. Pinto, M. L. Prolongo, “Stoichiometry in Context: Inquiry‐Guided Problems for Encouraging Critical Thinking…”, Int. J. Eng. Pedagogy, 3, 24‐28 (2013).
‐ M. L. Prolongo, J. Corominas, G. Pinto, 
M L Prolongo J Corominas G Pinto “Química Química de Medicamentos de Hierro: Propuestas Educativas Contextualizadas
de Medicamentos de Hierro: Propuestas Educativas Contextualizadas”, An. Quím., 110, 218
An Quím 110 218‐224 224 (2014).
(2014)
‐ G. Pinto, P. Lahuerta, “Velocidad de Fusión del Hielo en Distintas Disoluciones: Un Ejemplo de Aprendizaje Activo…”, Educ. Quím., 21, 54‐62 (2015).
Degradation of poly(lactic acid) in different
aqueous solutions
aF.R.
Beltrán, aA. Alarcón, aR. Gordillo, aE. Cañete, aE. Perona, bM.U. de la
Orden, aV. Lorenzo, aJ. Martínez Urreaga.
a Dpto. Ingeniería Química Industrial y Medio Ambiente, Grupo “Polímeros, Caracterización y
Aplicaciones”, E.T.S.I. Industriales, UPM, Madrid (joaquin.martinez@upm.es)
b Dpto. Química Orgánica I, Facultad de Óptica y Optometría, UCM, Madrid.

Abstract
Poly(lactic acid) (PLA) is one of the main bioplastics. More than 2⋅105 t were produced in 2014 (1), most of which were used in food packaging due to the
biocompatibility, good processability and good mechanical, thermal and optical properties of PLA. Although PLA is considered a biodegradable plastic,
several studies have revealed that the PLA grades used in packaging do not compost easily. Moreover, studies based on Life Cycle Analysis have concluded
that mechanical recycling is the best alternative for PLA wastes coming from packaging.
In order to investigate the viability of the mechanical recycling of PLA, it is necessary to compare the properties of the recycled and the virgin plastic. One
of the key properties in food packaging is the degradation of the plastic and the migration of the degradation products into the food. In this work the
degradation and migration of virgin PLA in different liquid food simulants have been studied.

Materials and methods
PLA samples (Ingeo 2003D, Nature Works) were melt compounded in a Rondol twin-screw microextruder. Films of 230 ± 20 µm were immersed in
deionized water, NaOH 10-3 M and 3 wt. % acetic acid (simulant used for acid food) for different times at 40 and 60 ºC. Lactic acid and lactide were used as
model compounds. The polymer degradation was characterized using intrinsic viscosity (IV, measured in trichloromethane at 25 ± 0.5 ºC). Structural
changes in the polymer were followed by IR spectroscopy. The migration of the degradation products to the media was studied by UV spectroscopy.

Results PLA degradation 24 h at 60 ºC in
920
PLA suffers hydrolytic degradation, which leads to a acid simulant
decrease of the average molecular weight , observed as a
24 h at 60
decrease of the intrinsic viscosity. The degradation is
ºC in water
faster in the acid food simulant.
140 30 d at 40
ºC in water
120
PLA
IV (mL/g)

100

80
H2O Structural changes studied by IR spectroscopy
NaOH
The absorption centered at 920 reveals the crystallization of PLA during the immersion
CH3COOH (2). The change is very important at 60 ºC. Is more important in the acid simulant
60 because the crystallization is favored by the decrease of the average molecular weight
0 10 20 30
during the hydrolytic degradation.
Time (d)

Migration studied by UV spectroscopy
Some degradation products migrate to the food PLA Oligomers
simulant. While at 40 ºC only low molecular weight
Lactic acid
oligomers were observed, at 60 ºC the monomer
lactic acid was also observed.
Absorbance
Absorbance

PLA Oligomers

40 ºC

60 ºC
Wavelength (nm)

Wavelength (nm) Conclusions
 PLA suffers hydrolytic degradation in contact with water or wet food.
References The degradation is faster in acid.
1. F. Aeschelmann, M. Carus. Biobased Building Blocks and Polymers in  The decrease of the average molecular weight during the hydrolytic
the World: Capacities, Production, and Applications–Status Quo and degradation favors the crystallization of PLA.
Trends Towards 2020. Industrial Biotechnology 11, 154-159, 2015.  At 40 ºC, UV spectroscopy shows the migration of low molecular
2. F.R. Beltrán, M.U. de la Orden, V. Lorenzo, E. Pérez, M.L. Cerrada, J. weight PLA oligomers to the simulant. At 60 ºC, the monomer lactic
Martínez Urreaga. Water-induced structural changes in poly(lactic acid is also detected.
acid) and PLLA-clay nanocomposites. Polymer 107, 211-222, 2016.
DecarGas. Low-carbon clean fossils for the
next generation
aA. Abánades, bT. Wetzel, bL. Stoppel.
aETSII-Technical University of Madrid (abanades@etsii.upm.es)
bKarlsruhe Institute for Technology (KIT)

Abstract
Decarbonization of fossil resources, as natural gas, produces clean hydrogen and pure carbon particles. DecarGas achieves the
development of an industrial scalable reactor for decarbonization that enables the utilization of fossil raw materials for
hydrogen and pure carbon production in the energy and chemical sector. It will constitutes a real tool to avoid greenhouse
gases emissions (as recognised by New Scientist), producing competitive marketable products.
DecarGas propose a technology to produce both high purity industrial hydrogen (up to several ton/h) without CO2 emissions
and huge amounts of pure graphitic carbon in a sole process, reducing drastically their environmental production impact and
combined cost.

Technology solution
We have verified the concept of an innovative methane decomposition reactor
based on methane injection into liquid metal that may be scalable to industrial
scale and economically competitive, providing an alternative pathway to use
natural gas while safeguarding the climate and facilitating the integration of a
clean energy carrier like hydrogen in the energy sector.
A simple bubbling of natural gas into a liquid metal bath allows the large scale
implementation of fossil decarbonization. The process is an effective carbon
capture technique with two worthy products:
- Graphitic carbon showing potential applications as structural, light and high
conductivity material.
- Pure hydrogen produced from fossil without carbon dioxide generation or CO
traces.
100

A Suitable glass reactor material and bubbling device have been identified for
Hydrogen mole fraction in the product gas in %

90 Tref = 1175 °C

the operation with liquid tin at 1200 ºC achieved, producing 100 W of hydrogen 80 Tref = 1100 °C

power with 78% yield for several days.
70

60 Tref = 1050 °C

Life cycle assessment (LCA) shows lower environmental impact than steam
50
Tref = 1000 °C
40

reformer and similar to electrolysis driven by wind. 30
Tref = 950 °C
20

10

Market potential
Tref = 930 °C
0
50 75 100 125 150 175 200
Methane volume flow rate in mln/min

Energy
• Fossil resources supply 80% of the energy demand worlwide. A technology to avoid CO2 emissions from these sources is a must until a sustainable
system based on renewables might supply energy needs. Chemical industry, as refineries and ammonia production requires process H2.
• Hydrogen economy is expected to need of the order of 400 Mtoe (toe=ton of oil equivalent) in 2050 [World Energy Technology Outlook 2050]
• Hydrogen economy deployment in the transport sector foresees a scenario with up to 3000 new Hydrogen fuelling stations in 2030 only in Germany
[E4tech. The Economics of a European Hydrogen Automotive Infrastructure].
Environment
• Reduction of greenhouse gases emissions is a must to achieve environmental targets claimed in COP21 conference in Paris
• Environmental regulations may penalty CO2 emission technologies, and short-term transitional solutions are needed.
• Availability of low-carbon technologies using fossil resources are required to avoid the collapse of sectors as energy and chemical industry.
• With hydrogen production with decarbonization, the mass amount of carbon captures will be 2,5 times lower that the CO2 produced bt steam
reforming and 6 times lower that coal gasification. Alternative/complementary technology to carbon capture and sequestration (CCS)
• Low CO2 emission technologies will be needed to comply with environmental agreements keeping industry competitively running.
Materials
• New technologies based on carbon-based materials as carbon fibres and graphene may be boost their development with the availability of cheap,
CO2-free pure carbon.

Competitive advantages
• Hydrogen costs are estimated between 1.5 and 3 €/kg, depending on the cost of natural gas.
• Environmental Life Cycle impact similar to wind and electrolysis (4 kgCO2/kgH2)
• Real solid carbon capture technique from fossils.

Next steps
Development from TRL4 to TRL5 by the integration of all the components of the industrial process at laboratory scale: thermal heating, CH4/H2
separation, heat recovery. Visit:
Achievement of TRL6: Design, construction and operation of a mid-scale prototype (20 kW-H2) https://www.youtube.com/watch?v=caxwx7o2v10
Control of a single phase inverter with
multiple modulation strategies
R. Ramos, D. Serrano, J. A. Oliver, P. Alou, J. A. Cobos

CEI-UPM (regina.ramos@upm.es)

Abstract
This work proposes a control linearization technique for a single phase inverter with multiple modulation strategies along the line period based on the
plant inversion. Multiple modulation strategies are used along the line cycle to optimize the size and efficiency of the converter, but this benefit from the
point of view of the energy processing generates a big problem for the control. This technique allows decoupling the control of the input current and the
output current which simplifies the control of the output voltage and the storage capacitor voltage in two ways: first, the control of both variables is
independent and second, the controller parameters will be independent on the modulation strategy due to the plant inversion. These advantages are
critical for the successful control of this highly non-linear time varying system.

Specifications PV panel application
• 2 kVA loads Our converter consists on:
• power density ≥50 W/in3 • An input DC voltage source, ??
Inverter
• PF= 0.7–1 • A bulky storage capacitor, ??
• Volume ≤ 40 in3 • A LC output filter
• Output 240 Vrms, 60 Hz
• A family of switches
• Input ripple current of < 20%
These set of switches can be commutated in different
• Input ripple voltage of < 3% patterns (sequences) that will be able to synthesize the
• η ≥ 95% desired inductor current, ?? , and the desired input
• temperature ≤ 60ºC current, ?? . The storage capacitor current will be given by
• EMC: FCC Part 15 B applying energy balance during the switching cycle.
• Does not require galvanic isolation

Simplified schematic
Problem statement
Averaged current through the inductor and the input current:
?? = ?? ?1 , ?2 , ?, ??
?? = ?? ?1 , ?2 , ?, ??
?? and ?? are a non-linear algebraic function

For a given reference input current and reference output current
?? ??? = ?? ?1 , ?2 , ?, ? ??
???
?? = ?? ?1 , ?2 , ?, ? ?? Plant inversion
Taking into account the differential equations of the output the control
problem will be reduced to:
??
?? ??? = ?? ??? − ??
???
?? · ?? = ?? · ?? ??? − ?? · ??
??
The output voltage can be controlled using ?? ??? and the storage capacitor
voltage can be controlled using ?? ??? , and both controllers are independent

Simulation results

Conclusions
This work has shown that it is possible to control highly complex power converter
topologies with multiple switching sequences by applying plant inversion
technique. The technique has been applied to a highly compact single phase
inverter in which different modulation strategies are applied along the line cycle
in order to optimize the size and losses of the converter.
The proposed technique allows decoupling the control of the output voltage and
the storage capacitor voltage and, what is more important, a very simple linear
controller can be applied to all the sequences what simplifies the control design.
Advanced topology solutions for high power
DC-DC converters for aircraft applications
aYann E. Bouvier, bPedro Alou
aCentro de electrónica industrial UPM (yann.bouvier@upm.es)
bCentro de electrónica industrial UPM

Abstract
Modern aircraft design have a strong tendency towards a More Electrical Aircraft (MEA). Airborne converters need to have
high efficiency and power density. Pareto front analysis of Volume and Losses is required to get the optimal design. Dual
active bridge topologies are the candidates for high power applications and multiphase interleaved converters can be
used to increase the efficiency at a cost of higher volume.

Comparison of Dual Active Bridge Topologies
S S 1S S 2
DCM 5 6
CCM LLC
Tr
Vin I L I L C L
Cout R I

S3 S4 S7 S8
t t t

o High RMS Current o Low RMS Current o Low RMS Current
Dual active bridge topologies have CCM and DCM modulations with o 2 Hard turn off switchings o 4 Hard turn off switchings o ZCS and q-ZCS
inductor in series. With a resonant tank in series and the o ZVS and q-ZCS o Only ZVS o Quality factor issues
transformer’s magnetizing the topology becomes the LLC converter. o Stability control issues

Pareto Front - CCM Multiphase Interleaved Converters
 Multiphase interleaved converter can reduce output capacitance with
output ripple reduction
 Each converter share the total power load, so the number of transformers
increases but the current rating decreases, diminishing proximity effects
 Pareto front shows the trade-off between Volume and Power Losses taking
multiphase interleaving into account

Magnetic Integration
Magnetic integration can be achieved in the different DAB topologies.
Leakage inductance can be used as the series inductor.
The estimation of the increase of
losses in the magnetic component
have to be correctly estimated
accurately in order to have a
successful integration. The value of
the leakage inductance of the
Power Losses
transformer is also critical to
estimate.
Finite element software is a useful
tool for both the estimation of
losses and the accurate calculation
of parasitics.

Series-Parallel Topology Prototype – Experimental Results
 Two 7:1 transformers instead Input Voltage Output Voltage Output Power Frequency Efficiency
S1 S2 Sa1 Sa2 of one 14:1. Magnetic design 400 V 28 V 10 kW 100 kHz 94%
Cres
Lres Ta is better
Vin Ca RLOAD  Current sharing ensured with
VIN primary series connection
S3 S4 Sa3 Sa4  Magnetizing current dc bias
VIN IRESO
elimination with synchronous
IRESO operation in secondary
10  Independent driving of the
Sb1 Sb2
0
two output bridges.
Admitance (dB)

-10
Tb  Power loss spreading,
-20 Cb VGATE
decreases heat dissipation
-30

-40
problems in output bridges
-50
Sb3 Sb4 and transformers.
 Quality factor can be
1 10 100 1000 10000
Frequency (kHz)
Q = 10 Q=1 Q = 0,1 improved
Isolated Single-Stage Three-Phase Full-Bridge
with Current Injection Path PFC Rectifier for
Aircraft Application
Sisi Zhao, Jesús A. Oliver, Pedro Alou
Centro de Electrónica Industrial ( sisi.zhao@upm.es )

Abstract
Nowadays there has been an important increase of electric equipment in modern aircraft which leads to a rising demand of electrical power. This
highlights the necessity of efficient, reliable and high energy density power converters which rectify three-phase 115 Vac to 270 Vdc for power distribution.
This work presents an isolated single-stage PWM rectifier with the integration of a current injection path and a full-bridge topology (IS2FBCIP PFC
Rectifier). The operating principle and modulation method are presented. Simulation results are shown to validate the functionality of the topology. Lab
prototype is being constructed to further validate the circuit concept. The advantages of the proposed rectifier are the relatively low number of switching
components, low implementation effort and its ZVS feature which leads to a high overall efficiency.

Proposed Topology Design Specifications

Buck-Type
Rectifier
Operating Principle & Space Vector Modulation (SVM)

Space Vector Modulation

Operating Principles
 Transformer volt-second balance
 SVM method on rectifier current

Conclusions
• Isolated three-phase single-stage PFC Rectifier:
Compact, without bulky DC link capacitor Future work
• Relatively low number of semiconductors, ZVS in all
• Lab prototype construction upon finishing
switches
• Experimental verification
• Modulation method verified in simulation
• Comparison with other single-stage or two-stage
• Detailed design guidelines
topologies in terms of efficiency and power density
Three-phase Buck Rectifier For Aircraft
Application With Variable Line Frequency
Uroš Borović, Pedro Alou, Jesús Oliver

uros.borovic@upm.es
Centro de Electronica Industrial, Universidad Politécnica de Madrid

Abstract
Clean Sky is the most ambitious aeronautical research programme ever launched in Europe. Its mission is to develop breakthrough technologies to
significantly increase the environmental performances of airplanes and air transport. The goal of this project is to develop a 10kW three-phase buck type
rectifier with variable line frequency for applications in future More Electric Aircrafts.

Architecture
Three-phase Buck Rectifier
Experimental Results
Specification
 Voltage: 230V RMS phase to neutral point
 Main frequency: 400Hz
 Nominal Bus Voltage (Vout): 400V
 Rated output power: 10kW
 Military derating 30%
 Frequency range 360-800Hz
 PF: 0,95 as minimum
 THD: Each harmonic has an individual limit
 DO 160 F avionic standard

Prototype

Conclusions
A three-phase buck-type PFC rectifier was designed with 97.4% efficiency suitable for supplying 400V dc distribution systems. A modified damping network
has been used to reduce the damping resistor’s losses. Selected modulation strategy is current Space Vector Modulation (SVM). The experimental results
show very good agreement with the theoretical ones.
A 99.5% efficient 50kW/dm3 hybrid DC-DC
step-up converter for PV applications
Diego Serrano, Miroslav Vasić, J.A. Oliver, Pedro Alou, J.A. Cobos

CEI (diego.serrano.lopez@alumnos.upm.es)

Abstract
A novel hybrid DC-DC converter for PV applications has been developed in collaboration with Huawei. The main objectives of the project is to obtain an
ultra-efficient (nominal efficiency 99.5%), ultra-compact (50kW/dm3) DC-DC converter where the input voltage is changed in a very wide range (from
300V to 1000V) with a total power of 15kW.
Due to the very wide range of the input voltage, the proposed converter is implemented as a combination of a fixed-gain converter and a regulated step-
up converter. The fixed gain converter is designed to process power in a very efficient way, while the regulation stage provides the desired control of the
output voltage.
Depending on the value of the input voltage the amount of the load power processed by the each stage differs significantly and it has a great impact on
the design of the system.
P (kW) Input Power
Specifications 20
15
Parameter Value 10
Rated Power 15kW 5
Input Voltage 300V to 1000V 0
Output Voltage 600V to 1000V 300 500 700 900 Vin (V)
DC Bus Voltage Control Yes, for Vout=700V Vout (V) Regulation Curve
Peak Efficiency 99.5% @ Vin=580V 1000
Losses 75W 900
Power Density 50kW/dm3 800
700
Volume 0.3dm3 This DC-DC converter is used to control the input
600
voltage of the following inverter
300 500 700 900 Vin (V)

Project Stages Converter
Successfully tested and Final prototype
delivered to Germany Digital Control
implemented in
Prototypes FPGA to achieve
Control regulation and ZVS;
and Tests
and PCBs modular PCB design
Open loop
prototypes of
both stages
Components and
Optimization
Topology Selection An optimization
process and detailed
Different topologies were component testing
analyzed for different were performed
regulation curves

Results Two 7.5kW converters in parallel
10000
Fixed-Gain Stage Regulation Stage
5000
Power (W)
0
300 500 700
Maximum Maximum
40 Losses (W) power power

20 Without auxiliary
circuitry losses Conclusions
0
300 500 700 This challenging project has been successful thanks to the hard integrated work of a broad and
100 diverse team of undergraduate and graduate students and professors, who covered all the stages
Efficiency of the project, from its initial conception and design methodology to the final implementation
99,5 and tests in Germany, including visits to China and Sweden. The co-authors of this poster wish to
Target specifically acknowledge the work of Víctor, Marcelo, Helena, Regina, Rafael, Sergio and Pablo.
We also acknowledge the strong support and direction of Dr. Petar Grbović and
99 Roland Huempfner, at Huawei.
300 500 700
IFEC challenge: High efficiency and high power
density isolated DC-DC converter for server
applications
aM. Vasič, I.Zubitur, D. Čohadžić, I. Murga, D. Gil, & J.L. Millán
aCEI-UPM (miroslav.vasic@upm.es)

Abstract
Within a data center, all major devices are powered from a high current 48V or 12V distribution bus. Conduction losses for those buses are excessive. Due
to its lower conduction loss, high-voltage DC distribution power architectures have drawn attention recently.
In this architeccture, 380V is stepped down to 12V in a single stage, thus, the high output current isolated DC-DC converter is the key stage. In order to
place the high voltage DC-DC converter on the CPU board (motherboard), this converter has to be very efficient and high-density to be compatible with
core processors, memories, et al. available space. In this project, an LLC converter has been built. The integrated planar matrix-transformer design,
optimization of the output termination and high frequency operation losses are studied.

HV-DC System Architecture
 Converter Specifications
Input Voltage 360-400Vdc
Output Voltage 12Vdc ±0.1V
Output Power 750W (peak), 60A
>91%@10%load
Efficiency
>97%@50%load
@380Vdc :
>96%@full load
>15kW/L
Power Density:
(Volume=Length*Width*Height)
Temperature ΔT <40°C with 500W cont. op.
rising 200LFM airflow
20A to 40A step
<300mV over/under-shoot Comparison of data center power architectures. (a) Current architecture
@di/dt=2A/µS
with 12V bus. (b) Alternative architecture with 48V bus.
(c) Future architecture with 400V bus, targeted in this work.
 Topology
 High frequency operation >600kHz-1MHz
PCB-Magnetics Integration Allows dramatic reduction of the transformer
size.
 Series – Parallel Matrix transformer
400V 4 modules of 4:1 transformers, instead of one
12V 16:1 transformer.
Reduction of each winding turns number in
190W / 15A primary side, allows integration in a 4 layers PCB
12V module and improves power handling.
 Output termination
Cap SR1A Core SR1B Power devices
Sec.
integrated in
Careful design of the terminals of the
Pri.
Pri.
Sec.
transformer transformer, integrating the SR devices and
windings.
SR2A Core SR2B capacitors over the windings, confines the AC
currents to a smaller area, thus reducing AC
related losses.

 Experimental Results CONCLUSIONS
Generation 1 Measurements Our first design iteration has been qualified for the Semi-final of the
96%
400V bus 2017 IFEC Challenge, and the project progress will be presented
95%
during a Workshop in APEC 2017 Conference. Our project will be
Efficiency

94%
evaluated to qualify for the IFEC Final (July 2017).
93%
92%
Magnetics design and SR driving have shown to be critical to achieve
91%
higher efficiency. Further improvement of magnetics will be addressed
90%
in our final prototype, together with digital control.
0% 20% 40% 60% 80% 100%
Thermal Full load test We are actively looking for Sponsors for the final prototype.
Output Power %

Thanks for your interest.
UPM-IFEC2017 Team
Linear Assisted Versus Purely PWM Switched
Mode Envelope Amplifier for EER Transmitters
aVladan Lazarević, Miroslav Vasića, Oscar Garcíaa, Moisés Patiñob,
Francisco Javier Ortegab
a Centro de Electrónica Industrial, UPM Madrid (vladan.lazarevic@upm.es)
b Grupo de Ingeniería de Radio, UPM Madrid

Abstract
Telecommunications transmitters based on the conventional linear amplifiers (classes A, B, AB) assure prominent linearity, but exhibit poor efficiency. The
efficiency of the transmitters can be significantly improved using Kahn’s EER technique. In this poster, two different approaches in supplying a power
amplifier based on EER technique are presented. The same part of both envelope amplifiers (EA) is switched-capacitor based voltage divider combined
with switching structure, analog multiplexer. The analog multiplexer is assisted with a series linear regulator, in the first system, and with a fourth order LC
filter, in the second system. Preliminary results after the system integration with a class E power amplifier are presented.

Envelope amplifier circuits Linear assisted EA:
PWM switched filter assisted EA:

Envelope amplifier (EA): a DC-DC converter Control modes:
with very fast dynamics
❖ Linear assisted EA:
the switching frequency is not higher than
the RF signal bandwidth
the linear regulator voltage drop, in order
to avoid additional distortion
❖ PWM switched EA:
Significant conduction loss:
the output filter introduces low losses (especially
for the air core design) Increased switching losses:
arised problems caused by the high sw.
frequency (pwm signal resolution, layout issues)
➢ efficiency up to 75% with 5MHz RF bw.
➢ Efficiency up to 84% with 10MHz sw. freq.
Distortion analysis
Intermodulation distortion (the transmitter noise)
expressed via: The software tool
 EVM (Error vector magnitude), flow-chart diagram
 ACPR (Adjacent channel power ratio)
❖ Challenging to describe mathematically
modulated signal in-channel Out-of-channel emission the influence of the EA imperfections
distortion on the transmitter noise
❖ A software tool designed in order to predict the noise contribution of the EA stage to the transmitter

System integration and the experimental results

EER transmitter 2 still in the design process!
➢ PWM switched EA (10 W prototype) 1.6 MHz 64QAM signal reference, 10 MHz sw. freq.
➢ Two sw. frequencies: 5 MHz, 10 MHz Estimated distortion:
➢ EA efficiency up to 87% and 84%, respectively
EER transmitter 1:
➢ Linear assisted EA
400 kHz 64QAM signal reference, 5 MHz sw. freq.
➢ Suboptimal class E PA
Estimated distortion:
➢ 2.5 GHz carrier frequency
➢ Overall efficiency: 33.7 %
(small output power, < 2 W)

Conclusions
In this poster two different approaches in EER-based tracking power supplies are compared. The linear regulator conduction loss is a bottleneck of the linear
assisted EA, while in the PWM switched EA the efficiency is damaged by the analog multiplexer switching losses due to the high switching frequency.
The PWM switched envelope amplifier performs better efficiency in wide range of the 64QAM reference signals, up to 2.5 MHz. Linear assisted EA is a
promising solution for the wide–band EER applications and EER transmitter and exhibits an acceptable distortion up to 4 MHz RF signal bandwidth.
On the use of blackbox polytopic models to
simulate dc microgrids
A. Francés, R. Asensi, O. Garcia, R. Prieto and J. Uceda

CEI-UPM (airan.frances@upm.es)

Abstract
The huge progress of power electronics technology along last decades opens extraordinary new possibilities for the electric grid. However, there are still
some challenges ahead before Smart Grids massive deployment. Lately, a lot of research has been carried out on converters topologies and control
strategies in order to get the most out of the microgrids. Therefore, there is a need for methodologies that allow designers to foresee the behavior of
these systems comprised of several different power converters governed by the proposed control strategies. In this context, our research focus on the
performance of the polytopic models for the analysis of commercial power converters working in dc microgrids. This is a nonlinear modeling technique
which integrates small signal models obtained in different operation points by means of suitable weighting functions. Furthermore, the linear local models
can be obtained in a black-box fashion using G-parameters models.

DC Microgrid Polytopic model
DC microgrid architecture Decentralized control strategy

• High density of electronic power
converters.
• Variability of operating conditions
• Different operation modes
• Nonlinear static and dynamic
response

Voltage restoration &
System level simulation • Weighted combination of linear
Current sharing dynamic responses.
Control
• Composed of linear two-port
models obtained around
different operating points.
• Ability to describe static and
dynamic nonlinearities.
• Performance improved applying
dynamic weighting functions.
• Large-signal analysis. • Nonlinear stability analysis by
• Dynamic interactions among means of Lyapunov method.
electronic power converters.
• Effect of protections on the
overall system. Conclusions
• Impact of system level control
• DC microgrids constitute a new paradigm in electronic power distribution
strategies on the static and
systems due to the high variability of its operating conditions.
dynamic behavior of the dc
microgrid. • Different kinds of nonlinearities are present in its normal operation.
• The capability to model the behavior of the dc microgrid composed of
commercial-off-the-shelf converters before its deployment would anticipate
malfunction or help to design high level controls and protections.
• Blackbox polytopic model is a powerful tool to predict nonlinear behaviors
and perform nonlinear stability analysis.
An Analysis of the Influence of Binarization
Techniques Applied to Swarm Optimization
when Solving the Set Covering Problem
José M. Lanza-Gutiérrez, Teresa Riesgo

Universidad Politécnica de Madrid, Centro de Electrónica Industrial (jm.lanza@upm.es)

Abstract
Although the Set Covering Problem (SCP) is a classical NP-complete problem, it is widely studied in the current literature because it fits with many non-
solved problems for the industry, such as facility location and vehicle routing. Reviewing the current literature, metaheuristics usually give the best
behavior solving the SCP. However, while the SCP is a discrete problem, most metaheuristics are defined for solving continuous ones, especially Swarm
Intelligence Algorithms (SIAs). Such metaheuristics are adapted for working on the discrete scope, but most authors did not perform any study to select a
concrete binarization approach. This fact might lead us to think that selecting a binarization approach does not influence the algorithm behavior. This
situation led us to study the influence of binarization techniques when adapting a recent SIA. As a result, we conclude that selecting a concrete
binarization approach definitively affects the solution quality.

Set Covering Problem Which is the problem? SCP and SIAs
The SCP is considered in the literature for designing expert systems, which
While the SCP is defined as a discrete optimization problem, most SIAs are
emulate the decision-making ability of human experts in a given field. For
designed for solving continuous problems. Thus, the metaheuristics should
example, we find works considering the SCP for facility location, ship
be adapted for working on the discrete search space.
scheduling, production planning, crew scheduling, and vehicle routing.
The SCP is defined as follows. Given a set M of m objects and a collection S
of n sets of these objects, each set with a non-negative cost associated.
Well, we have Why does nobody
The goal is to find a minimum cost family of subsets C ⊆ S , such that each to adapt the SIA perform any study to
element i ∈ M belongs to at least one subset of the family C. for solving the select a concrete
problem, but… binarization approach?
I need to offer
fire services
at the lowest
possible cost!

Experimental methodology
Step 3: solve an accepted
Step 1: select an SIA initially
SCP benchmark considering
designed for solving
the 40 binarization
continuous optimization
approaches
Metaheuristics: swarm intelligence
There are many complex problems, such as the SCP, for which traditional
exact solving methods are not recommended, because computational Step 4: analyze if
times rise exponentially with the problem dimension. Instead, approximate Step 2: adapt the SIA
selecting a binarization
techniques should be considered, such as heuristics and metaheuristics. considering 40 usual
technique influences the
Reviewing the current literature, we check the best historical results solving binarization approaches
behavior of the
the SCP was obtained by metaheuristics, specially SIAs. from the literature
metaheuristic: statistic tools

Conclusions
As problem size grows, It is crucial to select an adequate
discretization techniques with a binarization approach to
better exploration to exploitation guarantee that the solving
ratio outperform the others algorithm reaches its full potential

It is especially significant to
select an adequate binarization
approach for solving large
problems

Further reading: "Analyzing the Effects of Binarization Techniques when
Solving the Set Covering Problem Through Swarm Optimization". Jose M.
Lanza-Gutierrez et al. Expert Systems with Applications (IF:2.981), 70, 2017.
Automatic Integration of HLS-Generated
Hardware Accelerators in an Embedded
Parallel Computing Architecture
TEC2014-58036-C4-2-R
Alfonso Rodríguez, Eduardo de la Torre
Centro de Electrónica Industrial, Universidad Politécnica de Madrid
(alfonso.rodriguezm@upm.es) FPU Grant Program

Abstract
High-Level Synthesis techniques have become an appealing alternative to traditional design flows in digital circuit design, being able to increase
productivity without compromising execution performance. Using high-level programming languages as the entry point, developers can easily decide
which parts of a given application are to be accelerated in hardware.
Commercial tools are able to define, at design time, the level of optimization for each hardware accelerator by using certain directives. However, with the
vast majority of these design flows, only one accelerator is generated per function. The ARTICo3 architecture faces this problem from a different
perspective, using replicated hardware accelerators to exploit dynamic data-level parallelism in an embedded GPU-like approach. In this work, a
methodology to automatically integrate HLS-generated accelerators in ARTICo3 is presented, enabling the use of the architecture by embedded software
developers with no previous experience in digital circuit design.

High-Level Synthesis Code Generation IP Packaging

C, C++, OpenCL HDL HDL
Wrapper
ARTICo3
• Embedded computing architecture
• Hardware acceleration HDL
• GPU-like computing (OpenCL)
• Run-time adaptivity
• Computing performance
• Energy consumption
• Fault tolerance

Design Flow Steps
• Developer writes high-level application
code in C, C++ or OpenCL
• Commercial HLS toolchains generate
RTL code in HDL User-driven
• Script-based code generation is carried
out, inserting the HLS output in custom
wrapping logic
• Wrapped code is packaged to be used
as an ARTICo3 accelerator
• ARTICo3 is used to provide run-time
adaptive scalability in the execution of
the application Automatic
Resilient communications for a multi-FPGA,
hardware-accelerated cluster
Alberto Ortiz Cuadrado, Alfonso Rodríguez Medina , Eduardo de la
TEC2014-58036-C4-2-R
Torre Arnanz
Centro de Electrónica Industrial, Universidad Politécnica de Madrid
(Alberto.ortiz09@gmail.com) FPU Grant Program

Abstract
FPGAs have proven to have better performance/watt ratios when compared to other computing platforms. Moreover, dynamically reconfigurable FPGAs
may provide software-like flexibility while keeping some of the benefits of dedicated hardware accelerators (e.g. performance, energy efficiency). In a
single-FPGA context, the ARTICo3 architecture provides runtime adaptation capabilities, enabling trade-offs between scalable performance, reliability and
energy consumption.
However, when facing applications with high computational loads, using only one computing device might not be sufficient. The work here presented aims
at providing, at system level, the same features that are provided by ARTICo3 at device level, but in an FPGA-based computing cluster. The proposed
solution relies on the use of AMPI as part of the communication infrastructure, being the implementation of this interface on top of a network of ARTICo3-
enabled FPGA nodes the focal point.

MPI AMPI
Having to face high demanding tasks, it is nowadays common to develop Each node could be formed by any combination of microprocessors and
parallel algorithms. As such, the communication protocol should provide FPGAs, making adaptability a critical necessity in the system. Moreover,
a way to parallelize the tasks and keep up the communication between each node will be using the ARTICo3 architecture, so the communication
nodes. The protocol used (MPI) follows the shown communication system should incorporate the same functionality. As such, AMPI
diagram, permitting the use of collective and point-to-point provides MPI with a layer dedicated to adaptability in terms of load
communication: balancing as well as heterogeneity and fault tolerance.
Rank
Parallel Program
Process 0
0

Master Process
(Rank 0) Process 1
1

Process 2
2
Slave Process Slave Process Slave Process
(Rank 1) (Rank 2) (Rank 3)
3

Time

AMPI cluster with ARTICo3
Modular Redundancy Node Redundancy
Next step: Router
ARTICo3 enables each x16 FPGAs AMPI moves redundancy
node to execute various to a higher level. Process
copies of a given task to 1 is a low demanding task
provide with reliability. executed twice in
Process 3 is executed Switch Communication Network different nodes.
twice in FPGA 2.

FPGA 0 FPGA 1 FPGA 2

Core Core Core Core Core Core
0 1 2 3 4 5

Specialization Performance

A given node can have its When facing a high
ARTICo3
HW specialized in HW
computational load, AMPI
accelerating a given task. accelerators can distribute it between
FPGA 1 has its HW nodes. Process 0 is the
specialized in executing most demanding task.
Process 2.

Process 0 Process 2
Process 1 Process 3
Distributed Artificial Neural Network over a
Wireless Sensor Network
David Aledo Ortega, Félix Moreno González

Centro de Electrónica Industrial (david.aledo@upm.es)

Abstract
The aim of this thesis is, first to implement Artificial Neural Networks (ANNs) on hardware, and second to distribute an ANN over a Wireless Sensor
Network (WSN). WSN can benefit from the parallel and distributed way of computation of the ANN, although special care must to be taken when
distributing ANNs over WSNs to avoid excessive communication burden. A deep hierarchical architecture is proposed to balance processing and
communication energy costs, in which communication costs have more impact.

Artificial Neural Networks
Backpropagation ?
Deep Learning
• Neuron model: ?? = ? ??,? ?? + ?? • Gradient vanishing problem: • By increasing number of layers can be obtained easily more
?=0
• Training: ∆?? ∝ ?? ; ?? = ? ? ??+1 ??+1 abstract patterns.
error backpropagation => Input layers cannot ∈ 0,1 • From dots and strokes on the first layers, which can be
learn well if there are many layers. combined into circles, polygons, …

comp • Theorem:

… An ANN with 2 layers of enough
inputs
… … outputs desired
hidden units can approximate any
outputs multivariate continuous function
with arbitrary accuracy.
• May be needed too many neurons in the
hidden layer. • Stacking Denoising Auto-Encoders (DAEs):
Hardware implementation • Layer-wise unsupervised pre-training:
• VHDL IP published available on OpenCores.org freely : - AE: uses input data as desired outputs
https://opencores.org/project,artificial_neural_network forcing data to be encoded and decoded.
… … …
• Configurable: … … - DAE: corrupts inputs to force robust
(through generics) encoding and avoid identity mapping.
• Final fine-tuning supervised training:
- Due to gradient vanishing preserves
low-level features and focuses on
interesting high-level ones for the
task to be soved.
Noise …
… comp

Distributed ANN over WSN
Toy example application: MNIST
• Classification of handwritten digits. It is a benchmark for machine learning.
• Emulation of different sensor catching partial information of the same event: Hardware implementation
• A Deep ANN can be distributed more efficiently on a WSN.
7 • Adding hierarchy can be exploited local differences easier and save
… …
… communications.
• Future work: simulate a WSN with tree topology and around 6 hops
to coordinator.
… …
… … … …

… …



… …
… …
… … … …
- A shallow ANN need a lot of vertical cuts =>  communication cost

Conclusions
Wireless Sensor Networks can benefit from the parallel and distributed way of computation of the ANN. Deep hierarchical ANN can fit better on the WSN
topology and save communications energy costs. More simulations are needed to confirm this result. And a real implementation with ZYBO boards will be
performed, using the developed VHDL IP.
Contribution of DFIG wind turbines to Fast-
Frequency Response and Power Smoothing.
Danny Ochoa, Sergio Martinez

Department of Electrical Engineering, ETSI Industriales, UPM
(danny.ochoac@alumnos.upm.es)

Abstract
This thesis aims at developing new control strategies applied to the power electronics converter of a Doubly-Fed Induction Generator based wind turbine
(DFIG-WT) in order to provide both fast-frequency response (FFR) and active power smoothing capability. Firstly, a simplified electromechanical model of a
DFIG-WT has been proposed. In addition, a virtual inertia control strategy version of the optimized power point tracking method (OPPT) is proposed for
this kind of wind turbines. In this method, the maximum power point tracking curve (MPPT) is shifted to drive variations in the active power injection as a
function of the grid frequency deviation and the time derivative of frequency, by exploiting the available inertial resources. The proposed methodology
integrates the model in a primary frequency control scheme to assess the interaction with the rest of the plants in the power system.

I. DFIG-WT simplified model II. FFR control strategy proposed
A simplified electromechanical model of a DFIG-WT (Fig. 1) A novel control strategy for providing fast-frequency response by a DFIG-WT as a function
oriented to frequency control studies has been presented of the grid frequency deviation and the time derivative of frequency has been proposed in
in [1]. Since the Rotor Side Converter (RSC) is responsible [2]. The corresponding control scheme (Fig. 2) is implemented in the RSC.
for controlling the rotor speed in order to deliver the f nom[pu]
optimal active power, the analysis and modeling efforts are Wash-out filter Dead Band fKopt,max
 Variable
focused in this component. f s[pu]  f[pu] s f *
[pu]
MPPT-curve
Σ fKopt  *
f Kopt P
Pg*[pu]
1 Two  s Eq. (1) Σ Kopt
v Wind speed CTRL 
(input) 
r 0[pu] g [pu]
g [pu] fKopt,min
Speed Grid S/H
Wind Turbine

DFIG controller Pg g [pu]
Sample and Hold Adapted OPPT-method
(MPPT)
Active r30[rad/s]
Electric power ΔPWT[pu] Δ fKopt f Kopt  (1)
1
  2 kvir f[Hz] p 
3
Generator d
(output) Π Eq. (2) Eq. (3)
1  sTlp
Gear box r 0[rad/s]
dt
Pitch angle Low pass filter  r 0[pu]  df s[pu]
controller PWT [pu]  2  kvir H eq[s]  f s[pu] (2)
RSC GSC   dt
Electronics Converter Time derivative of frequency loop  s [pu] 
PWT [pu]
f kopt  Wvir (3)
Fig. 1. DFIG-wind turbine general scheme. Fig. 2. Control scheme for implementing the proposed method. K opt g3[pu]

III. Simulation and results
The proposed DFIG-WT model has been incorporated into a traditional load frequency control scheme for a
single area (Fig. 3). The behavior of the grid frequency dynamics following a sudden load increase is shown
in Fig. 4. By considering actual wind speed conditions (Fig. 5), the simulations are performed again in order
to represent the power system behavior in a more realistic manner (Fig. 6). As a conclusion, the continuous
application of the proposed method allows to improve the frequency response of the power system in case
of severe load changes. Also, under variable wind speed profile, the DFIG-WT output power is slightly Fig. 5. Real wind speed profile.
smoothed, which results in a reduction of the grid frequency fluctuations.

Load
frequency
DFIG-wind farm Comb. cycle power plant
control
30 % 15 %
scheme
(Single area)

Hydro power plant Nuclear power plant
5% 35 %

Energy share Fig. 4. Grid frequency during a sudden load Fig. 6. Time domain simulations under
percentages Coal power plant increase under non-variable wind conditions. actual wind speed conditions.
15 %

Fig. 3. Test power system (Load frequency control scheme).

IV. Publications
[1] D. Ochoa and S. Martinez, “A Simplified Electro-Mechanical Model of a DFIG-based Wind Turbine for Primary Frequency Control Studies,” IEEE
Lat. Am. Trans., vol. 14, no. 8, pp. 3614–3620, Aug. 2016.
[2] D. Ochoa and S. Martinez, “Fast-Frequency Response provided by DFIG-Wind Turbines and its impact on the grid,” IEEE Trans. Power Syst., vol.
PP, no. 99, pp. 1–1, 2016. (DOI: 10.1109/TPWRS.2016.2636374)
Improved Predictive Direct Power Control of
Doubly Fed Induction Generator during
unbalanced grid voltage based on Four Vectors
Mohammad Ebrahim Zarei, Carlos Veganzones Nicolás, Jaime Rodríguez Arribas
Department of Electrical Engineering, ETSI Industriales,
Universidad Politecnica de Madrid, (me.zarei@alumnos.upm.es)

Abstract
This work presents an improved strategy in the field of predictive control (PC) of the doubly fed induction generator (DFIG). The proposed strategy applies
four voltage vector in every period in order to have constant switching frequency and low current THD. The appropriate voltage vectors in each period are
recognized when the estimated duration times of selected active vectors are positive. The suggested techniques has excellent performance during
transient and steady-state conditions. The proposed predictive control can easily flow the references under normal and abnormal voltage conditions even
if the references contain ac terms. Without any additional controller, the proposed technique could obtain smooth stator active and reactive power or
smooth electromagnetic torque or could inject sinusoidal and balance current into the grid when the voltage unbalance appears in the stator winding of
DFIG. Moreover, under unbalanced voltage conditions, still four voltage vectors are applied in every switching period. The simulation studies for this
technique are carried out for 2 MW DFIG in Matlab Simulink environment under ideal and unbalanced grid voltage. Furthermore, the experimental studies
are now conducting. The results of this technique are compared to other strategies. The comparisons show that the performance of the proposed strategy
is significantly superior to the other strategies.

I. Introduction:
The DFIG consists of two converters. The rotor side
converter (RSC) controls the stator active and reactive
power. The grid side converter (GSC) regulate the dc link
voltage. Schematic of DFIG is shown in Fig. 1. The
experimental setup is shown in Fig. 2.

II. Proposed Strategy: Fig. 1. Simulated DFIG wind turbine
Fig. 2. Experimental Setup
In this novel strategy, at each switching period, four voltage vectors are selected to be
applied by the rotor side converter of doubly fed induction generator according to a very
simple algorithm. This algorithm is shown in Fig.3. Duration time of selected vectors are
estimated by two equations. Finally, the appropriate vectors will be employed according to
their duration times.

-1 -1
Ps (MW)

Ps (MW)

-1.5 -1.5
Ps Ps
Ps-ref Ps-ref
-2 -2
0.4 0.5 0.6 0.7 0.4 0.5 0.6 0.7
0.5 0.5
Qs
Qs (Mvar)

Qs (Mvar)

Qs-ref
0 0
Q
s

-0.5
Qs-ref Fig. 3. Proposed algorithm
-0.5
0.4 0.5 0.6 0.7 0.4 0.5 0.6 0.7

-5 -5 III. Simulation results:
Tem (kNm)
Tem (kNm)

-10 -10 A comparison study have been conducted
-15 -15
between the proposed PC and two previous PC
0.4 0.5 0.6 0.7
0.4 0.5 0.6 0.7
method under the same conditions. three
2 2 control power targets have been adopted
during unbalanced grid voltage which are:
Is (kA)

Is (kA)

0 0
1- smooth active and reactive power (t= 0.4 s
-2
0.4 0.5 0.6 0.7 -2
0.4 0.5 0.6 0.7 to t=0.5 s)
2 2 2- smooth electromagnetic torque and reactive
power (t= 0.5 s to t=0.6 s)
Ir (kA)

Ir (kA)

0 0

-2
3- balance stator current (t= 0.6 s to t=0.7 s)
-2
0.4 0.5 0.6 0.7 0.4 0.5 0.6 0.7
The simulation result for the proposed PC is
Fig. 4. Result of proposed PC of DFIG during Fig. 5. Result of previous PC based on three voltage shown in Fig. 4. the results for previous PC
unbalanced grid voltage with three different vector during unbalanced grid voltage with three based on three voltage vectors are shown in
different control targets Fig. 5. As can be seen, the performance of the
control targets [1]
proposed strategy is significantly superior to
the other strategy.

IV. References:
1- M. E. Zarei; C. Veganzones Nicolas; J. Rodriguez Arribas, "Improved Predictive Direct Power Control of Doubly Fed Induction Generator during Unbalanced
Grid Voltage Based on Four Vectors," in IEEE Journal of Emerging and Selected Topics in Power Electronics , vol.PP, no.99, pp.1-1, 2017.
DOI: 10.1109/JESTPE.2016.2611004
2- M. E. Zarei, and B. Asaei, "A robust direct current control of DFIG wind turbine with low current THD based predictive approach," Wind Energy, vol.
19, pp. 1181–1199, 2016. DOI: 10.1002/we.1890.
Supercritical CO2 in a concentrating solar
power plant with thermal energy storage

Luis F. González-Portillo, Javier Muñoz-Antón, José M. Martínez-Val

Energy Engineering Department ETSII-UPM (lf.gonzalez@upm.es)
GIT

Abstract
Concentrating solar power (CSP) is one of the renewable technologies that offers appealing management of electricity. Integrating thermal energy storage
(TES) into CSP systems helps to overcome the fluctuations associated to the sun nature. In the proposed CSP plant, a field of mirrors concentrates the sun
radiation into a solar tower to heat supercritical carbon dioxide (s-CO2) and then send it, either through a regenerative Brayton cycle to generate
electricity, or through a heat exchanger to store the thermal energy to a TES. The solar tower technology is an excellent option to use s-CO2 since its fixed
receiver can withstand the high pressures required by the fluid. Therefore, thermal power goes from the tower to the power block without intermediate
heat exchangers. However, the integration of TES needs a heat exchanger, which decreases the maximum temperature in the Brayton cycle. This changes
the optimal configuration of the thermodynamic cycle and damages the thermal efficiency.

Main idea
• Use supercritical CO2 (s-CO2) as heat carrier fluid in a solar tower and as working fluid in a
regenerative Brayton cycle, coupling both parts without the need of heat exchangers.
• Integrate TES with the solar tower and the Brayton cycle in order to increase the plant revenue [1].
• Analyze the performance of this CSP plant when it works with and without TES.

CO2 pressure-enthalpy diagram with a
regenerative Brayton cycle

Main parameters of the regenerative Brayton cycle

Compressor Inlet Temperature 50°C
Turbine Inlet Temperature 500°C
Max Compressor Outlet Pressure 10 MPa
Compressor Isentropic Efficiency 0.89
Turbine Isentropic Efficiency 0.93
Performance Net Mechanical Power Output 10 MW
The off-design performance is calculated for different situations depending on
the conditions used to design the CSP plant (simulations with EES and [2]): Recuperator thermal size 3.4 MW/K

Nominal conditions
• Compressor Inlet Temperature: 50°C
0.4 0.4
• Turbine Inlet Temperature: 500°C Design points Design points
Off-Design Efficiency [-]

Off-Design Efficiency [-]

TES working 0.38 0.38
• Turbine Inlet Temperature: 460°C 0.36 0.36
Night
0.34 0.34
• Compressor Inlet Temperature: 35°C
500°C Design 50°C Design
0.32 0.32
460°C Design 35°C Design
Future Work 0.3 0.3
• Study other variations of the Brayton cycle: 440 460 480 500 520 35 40 45 50 55
o With other fluids Turbine Inlet Temperature [°C] Compressor Inlet Temperature [°C]
o Recompression vs simple regeneration Off-Design efficiency as a function of the temperature for a CSP plant designed to work with
• Analyze other configurations of solar tower with TES. TES (460°C) and without TES (500°C) during the night (35°C) and during the day (50°C)

References
[1] L.F. González-Portillo, J. Muñoz-Antón, J.M. Martínez-Val, An analytical optimization of thermal energy storage for electricity cost reduction in solar
thermal electric plants, Applied Energy, Volume 185, Pages 531-546 (2017).
[2] J. Dyreby, Modeling the Supercritical Carbon Dioxide Brayton Cycle with Recompression, PhD Thesis. The University of Wisconsin, Madison (2014)
RELOJ DE SOL
aJavier Cano Nogueras, José María Martínez-Val, Rubén Abbas,
bJulio San Millán, Andrés Rodríguez
aOficinaTécnica de la FFII (jcano@ffii.es), b Jefatura de nuevas instalaciones de la FFII,
Universidad Politécnica de Madrid ETSI Industriales

Abstract
“Reloj de Sol” is a concentrator of solar radiation with a configuration of longitudinal cylindrical mirrors, with the same focal axis, fixed on a platform that
rotates horizontally as the shadow of the whore in a sundial.
All solar radiation concentrators that use moving mirrors to focus radiation involve the use of considerable weights, by rigidizing the structure.
In “Reloj de Sol” you only need about 20kg of flat and cheap mirror (that deform in the assembly), 15kg of steel and 30kg of concrete, to capture in the
heating fluid a thermal power of 1kw; In front of the 30Kg of tempered and curved mirror; 50 kg of steel and 100 kg of concrete, which requires a
conventional assembly.
With this material decrease, “Reloj de Sol” gets to reduce the cost by 65%
This configuration is related to the so-called Fresnel Reflection assemblies, but with characteristics not common to that family

GENERAL LAYOUT OF THE PLANT
•Horizontal platform rotating around vertical axis over some sets of wheels
•The sun must be in the symmetry plane of the concentrator
•Lack of transversal incident angle: neither shading nor blocking losses

• Use of standard flat mirrors
• They are mechanically bent

PRELIMINARY RESULTS
Curvature of mechanically bent mirrors Slope deviation
• Two forces to be considered: own weight and external momentum at mirror ends
• Slope variation due to own weight
• Own weight ?
? ? • Maximum slope perturbation for ? = 4
??? ? ? ??? ? ?
? ? = ? ?+ −? −? ? ? =− ?+ ?−
?? ? ? ???? ? ? ?? ??
??? ≈ =
• External moments at mirror ends ??? · ? ???
• ? = 3º and ? = 5
? ? = ?? ?? ?? ? ?
? ? = ?− ?+ • If R=20 m -> L= 2,1 m
??? ? ? • Deviation of 2 mrad
• Ratio of moments Ratio of ordinates (4 mrad at receiver), negligible
?? ?
?= ?
?? ?? ?=− ? − ?. ? ? + ?. ? < ? ??
?

Summer Winter A SMALL SCALE PROTOTYPE
•40 m2 of mirrors
Tilt: 0 deg

•Two 4-meter PTC receivers
Flux intensity maps •Receiver at 4.5 m above the mirrors
(kW/m^2) in Aswan •8 m diameter platform
•Cylindrical cage made of 8 horizontal wheels with fixed axes
Tilt: 10 deg

•Pressurized air as HTF (3 Mpa)
•Mass flow up to 0,2 kg/s
•Temperature rise up to 200 K
•Regenerative intermediate heat exchanger (patent ES 2440088)

CONCLUSION AND FUTURE WORK References
• Very simple structure • GIT  “Futuro Solar” With OHL (2015)
• Third prize of UPM 2T Innovatech Challenge 2016
• Use of flat mirrors and other elements easily found in the industry
• https://www.youtube.com/watch?v=ydLxln-R0d4
• No need of rotating elements beside the platform rotation Patent
• Potential reduction of the LCO0E • Granted: ES2537606, ES1138715U
• Small-scale prototype to be built in Tecnogetafe (latitude 40º North) • Pending: P201630832, P201630836
• Different heat transfer fluids must be considered for future applications in large-scale Development state
power plants • Investigation and prototyping-lab
• Estimated end date of prototype construction May 2017
Solar heater: solar tower power at different
enthalpies

Javier Muñoz-Antón, Luis F. González-Portillo, José M. Martínez-Val
GIT
Energy Engineering Department ETSII-UPM (jamunoz@etsii.upm.es)

Abstract
Solar tower has become one of the most attractive concentrating solar power technologies during the last years. The common solar tower consists of a
field of mirrors concentrating solar radiation into a small area in order to heat a fluid. Previous studies have shown the benefits of concentrating that
radiation at different sectors of the tower, resulting in the design of a new concept: the solar boiler, where the water boils in a similar way to the
conventional thermal power plants boiler, but using sun radiation instead of fuel or carbon. In this study, this idea evolves to the Solar Heater, where the
solar field concentrates solar radiation at different solar tower sectors in order to heat a supercritical fluid at different enthalpies. The characteristics of
some refrigerants in supercritical state can take advantage of this concept coupled with a Brayton cycle. This coupling will accelerate the response of the
plant, which implies several benefits for the future of this technology.

Description
 Previous works [1,2] justify the advantages of using high thermal heat fluxes to reach
high temperatures, and lower thermal fluxes to reach lower temperatures.
 The receiver can be split in several sections attending to the morphology of the
impinging thermal flux in order to obtain a cheaper technological solution [2].
Collector efficiency and absorbing
surface temperature vs. concentration
factor, for a value of the temperature
of the fluid Tf = 400oC [2]

“Low” Temperature
receiver sector

CO2 pressure-enthalpy diagram [4] with a
Scheme of a Brayton cycle with close-to-critical Brayton cycle [3].
regenerator [3]

Possibilities
 This solution reduces the piping length in comparison with conventional
linear receivers solar fields. “High” Temperature
 Pumping power can be highly reduced receiver sector
 This concept constraints allow more fluid possibilities: CO2, conventional refrigerants, new
fluids
 Coupling with Brayton cycles and Thermal energy storage to improve plant management
features

Main References
[1] Javier Muñoz, Alberto Abánades, José M. Martínez-Val, A conceptual design of solar boiler, Solar Energy, Volume 83,
Issue 9, September 2009, Pages 1713-1722
[2] Javier Muñoz, Jose M. Martinez-Val, Alberto Ramos, Thermal regimes in solar-thermal linear collectors, Solar Energy, Volume
85, Issue 5, May 2011, Pages 857-870
[3] Javier Muñoz-Antón, Carlo Rubbia, Antonio Rovira, José M. Martínez-Val, Performance study of solar power plants with CO2 as working fluid. A
promising design window, Energy Conversion and Management, Volume 92, 1 March 2015, Pages 36-46
[4] Span R, Wagner W. A new equation of state for carbon dioxide covering the fluid region from the triple-point temperature to 1100 K at pressures
up to 800 MPa. J Phys Chem Ref Data 1996;25(6).
A comparison of optical features between
parabolic trough collectors and linear Fresnel
collectors
aAbbas, Rubén; bMartínez-Val, José María
aGIT-UPM (rubenabbas@etsii.upm.es)
bGIT-UPM

Abstract
Parabolic trough collectors (PTCs) are still today the most mature technology in concentrating solar power. However, linear Fresnel collectors (LFCs) have
been identified by many authors as a candidate to reduce the levelized cost of electricity, although with lower efficiencies. Within Fresnel technology, two
possibilities appear for the receiver: multitube receiver and secondary reflector receiver. In the present work a developed Monte Carlo Ray Trace code is
used in order to compare the energy effectiveness and flux intensity map at the receiver for different days of the year and different orientations in
Almería, Spain, and in Aswan, Egypt. The optical annual energy and exergy efficiencies are also obtained for PTCs and LFCs, both with multitube or
secondary reflector receiver. It is concluded that LFCs are more suitable for low latitude locations, where the efficiency difference between both
technologies is lower.

Introduction Energy effectiveness and losses
Concentrating Solar Power (CSP) plants are generally divided into three In this section the energy effectiveness, which is measured as the energy
blocks: the solar field (concentrator plus receiver), the energy storage impinging onto the receiver by the energy that would impinge onto the
system and the power block. The solar field is probably the main reflecting surface if it were normal to the sun light at all time, is obtained
characteristic of the solar plant, which justifies that they are normally during the day June 21st. Also, the different sources of energy loss are
classified depending on the field technology used: quantified. This is done for Fresnel collector with secondary reflector and
• Linear concentrators. for parabolic troughs, both with NS orientation.
• Parabolic trough collector (PTC)
• Linear Fresnel collector (LFC)
• Punctual concentrators. Secondary
• Tower solar power (TSP) reflector Fresnel
• Stirling dish (SD) collector
Parabolic trough collectors are the most mature technology both in terms
of power installed and accumulated operation hours. However, linear
Fresnel collectors are seen as a technology with high potential to reduce
the LCOE thanks to its simplified structure.
In the present work the concentration features of parabolic troughs and
linear Fresnel collectors, both with multitube receiver and secondary
reflector receiver, are compared. An owned developed Monte Carlo Ray
Trace code is used for this purpose. State-of-the-art trough collector and
for a linear Fresnel with Fresdemo geometry are studied. Simulations are
carried out in Almería (Spain) and Aswan (Egypt) and with North-South
and East-West orientation.
Parabolic trough

Flux intensity comparison
The flux intensity at the receiver (for monotube techonologies) or at the
central tubes of the receiver (for multitube receivers) is obtained. The It results from this study that the energy effectiveness is higher for
following figure shows the flux intensity for a simulation carried out for parabolic troughs mainly due to shading losses and to the increase of
Almería assuming NS embodiments during June 21st. cosine losses when the impinging transversal angle is high (morning and
afternoon for NS embodiments)

Annual optical performance
The annual optical performance is estimated by means of the exergy
efficiency, which takes into account the flux intensity at the receiver..

One may observe that Fresnel collectors with secondary reflector are able
to achieve higher flux intensities, whereas multitube receivers would drive
to flux intensities slightly lower than those of parabolic troughs during the
central part of the day.
Also, it results that the flux intensity drop during the early morning and One may observe that differences between technologies are lower at
late afternoon is much higher for Fresnel collectors than for parabolic lower latitudes, i.e. linear Fresnel collectors are more likely to be
troughs. This is mainly due to the transversal impinging angle modifier, competitive in locations with high annual irradiation values.
which does not exist for parabolic troughs.
Regulatory Proposals for the Development of
Renewable Energy Self-Consumption in Spain
aGloria Mármol-Acitores, bCarlos Rodríguez-Monroy
aTechnical University of Madrid (UPM), Spain (gloria.macitores@alumnos.upm.es)
bTechnical University of Madrid (UPM), Spain (crmonroy@etsii.upm.es)

Abstract
COP22 called for political commitment to combat climate change by using low-carbon energy sources. The European
Commission recognises that industrial and domestic sectors have the possibility of consuming their own electricity,
due to the level of development and innovation of most Member States.

The rise in electricity prices, together with the decrease in the cost of renewable generation technologies, results in
estimations foreseeing up to a 75% increase of the self-consumption rate in European households.

However, the lack of regulation on this issue at European level has derived in different regulations being approved
across Member States. In Spain, Royal Decree 900/2015 has been considered too restrictive, in the sense of
precluding the financial feasibility of self-consumption systems’ deployment, whereas other European countries
with poorer renewable energy resources are experiencing a higher growth in this field.

Research Question
What regulatory approaches should be adopted in Spain in order to foster power self-consumption among domestic
customers?

Objectives
The objective is to analyse the current situation associated to power self-consumption both at European and
national level, in order to develop regulatory proposals that help fostering its development and use among domestic
customers in Spain, ensuring the economic feasibility of the power sector.

Sale of produced
energy

Energy Remuneration
Energy fed into the
power grid

Net Metering

Yes

Contribution to the system

No
'A review on the effects of structural breaks and volatility
clustering characteristics of Spanish gas and Brent prices'
aPablo A Cansado-Bravo, bCarlos Rodriguez-Monroy
aETSIIM (pablo.cansado.bravo@alumnos.upm.es)
bETSIIM

Abstract
The case of Spanish gas market is a good example, where gas supply conditions comprising a substantial
amount of long-term contracts together with limited connectivity with France, may explain the slower
pace of implementation of a liquid gas hub in the country. Moreover, the fact that globally the relationship
between natural gas and crude oil prices is not certainly transparent, not only because gas markets are
regionally segmented, raises questions about the view of steady state growth of natural gas prices.
The objective of the study is primarily to evaluate the underlying relationship between oil and natural gas
in Spain and provide quantitative information on this relationship. The selected period, 2002 – 2014 is a
highly representative time window to also assess the progress of liberalization in the Spanish gas market.
Furthermore the aim of the thesis is to assess the resilience of traditional oil-indexed structures and as a
consequence to advance the degree of penetration of liquid hub pricing into Spain over the referred
period.
In this sense the study concentrates on two key areas providing insight on the oil-gas relationship: order of integration and volatility characteristics. Within
the first area, we analyse the scope for the order of integration of a number of energy price series, including Spanish gas and Brent oil and allowing for the
possibility of structural breaks. Here our main contribution is twofold. Firstly to evaluate the unit root hypothesis of the selected variables using advanced
econometric techniques allowing for the possibility of structural breaks and secondly, to assess the impact and timing of those breaks on the changing
patterns of time series. Conventional unit root tests indicate that all variables are nonstationary. However results from our research suggest clear
nonstationarity only for coal. Moreover in the presence of breaks, the unit root null is always rejected for NBP prices at 5 percent significance whereas
tests suggest stationarity for Spanish gas prices more frequently than for oil prices. In the second area referred above, the thesis examines the changing
patterns of volatility in the oil and Spanish natural gas sectors. This property often manifests as volatility clustering and suggests that the conditional
variance of the return series may not be constant. Our investigation confirms that oil price volatility clustering effects are clearly transmitted into SG prices
allowing for the adequate lagging effects embedded in the gas contract price formula.

Structured Methodology
In order to take advantage of the existing research, the most appropriate methodology in each area has been selected. Some of the general processes
relevant for our research could be categorized as follows:
Understanding gas price setting characteristics in Spain.
Descriptive statistics for all the variables selected included others than oil prices are analysed, advancing similarities among variables. It is interesting to
realize that in spite of a well-connected LNG market worldwide, Spanish gas prices might show very limited response to other than crude oil-related
fundamentals.

Evaluation of the unit root hypothesis
Unit root testing is not only the most common method to study the persistence of time series but it is
also intimately related with advanced theoretical approaches to regression, cointegration and modelling.
In this sense methodological issues affecting unit root estimation, testing and validity of results in the
context of structural change will be thoroughly analysed. We devise a strategy for unit root testing that
guarantees a sound result. Following the choice of a strategy, the particular conditions are essential.
Issues such as the nature of the impact over time, i.e. how fast reaction is to the event, how to estimate
the precise dating of the structural break or even whether the assumption that there is only a single
break can be qualified, are at the heart of the process. A summary of different methodologies used and
compared in the thesis are: i. ADF and KPSS. ii. Unit root with one structural break: Zivot and Andrews
(1992), Vogelsang and Perron (1998), Lϋtkepohl, Muller and Saikkonen (2001) and Kim and Perron (2007)
iii. Unit root with two structural breaks: Lumsdaine and Papell (1997) and Lee and Strazicich (2006).

Measuring volatility clustering
In order to discuss the nature of volatility clustering in a quantitative manner it is needed to introduce a
structured methodology that can provide a consistent comparison of clustering effects over time.
Following recent research (Tsang and Li, 2010) and former theory on rolling analysis of financial time
series (Alexander, 2001, Dacorogna et all, 2001 and Zivot and Wang, 2006) we introduce an index as a
quantitative measure of volatility clustering that can also be used to compare the degree of volatility
clustering along Spanish gas prices and Brent time series. This index is calculated using a moving window
reference to obtain valid rolling indicators related to volatility clustering effects.

Modelling volatility in natural gas
In spite of the a wide variety of studies addressing performance of crude oil volatility in modelling there is limited investigation into the differences and
similarities with long-term gas contracts prices characteristics. Typically for Brent, model superiority is assessed with regard to both its ability to forecast
and also to identify volatility stylized facts (Kang et al., 2009). Narayan and Narayan, 2007 favour the use of the EGARCH model, whereas Kang et al.,
2009, conclude that the CGARCH and FIGARCH models are better to capture persistence in the volatility of crude oil prices.
Modelling SG returns and their volatility is performed in three steps. The first step involves the specification of the ARMA (p, q) model for mean returns
and diagnostic tests of their residuals. The second step is the specification of the GARCH (p, q) model for conditional volatility and its diagnostic tests.
Thirdly we will analyse asymmetric effects through extended GARCH models.
Time-dependent atomic physics in the interaction of
ultraintense X-ray laser with matter

a
Elisa V. Fernandez-Tello,a M. Cotelo,a E. Oliva,a A.G de la Varga,a P. Velarde

a
Instituto de Fusión Nuclear (ea.vazquez@alumnos.upm.es)

Abstract
The field of atomic physics in hot plasmas deals with the interactions of electrons and ions in an ionized medium and the radiation emitted from this
medium. The most general method to calculate plasma population kinetics and spectra is the collisional radiative model (CRM), which is based on
highly detailed atomic structure, and successfully reproduce experimental data. When a plasma is heated up very quickly with an ultrafast laser pulse,
we obtain a plasma out of equilibrium.
Our contribution to the Atomic Plasma Physics research consists in the development of a time-dependent code for laser-produced plasmas
simulations. For these calculations, the recently code BIGBART (developed by our group) is used. This code is capable of performing simulations of
time-dependent atomic populations coupled non-linearly to the free electron population.

1.FEL 2.Laser produced plasmas
The new free electron laser facilities
with very high intensity (>1017W/cm2), We study the evolution of laser produced plasmas. We simulate
tunable photon energy of several keV the interaction of X-ray radiation with matter from ultrashort and
and pulse duration below 100 fs, are ultraintense X-ray laser. We present non thermal time-
enabling to explore new techniques dependent simulations, calculated with the code BIGBART..
and physics in image, astrophysics and
material science.
2.1.Electron distribution function

With our code we obtain more consistent atomic data by
averaging with the real energy distribution.

1.1.LCLS
LCLS produces pulses of X-rays more than a billion times brighter than the
most powerful existing sources (synchrotons).

2.2. Ionic populations and rates
Not considering the evolution of the free electron distribution
leads to big differences in ionic population and collisional rates.

1

18K0L6
19K2L5 15
1x10
0.1 Coll. Excitation
20K1L6
14 Coll. Dexcitation
21K0L7 1x10
Spontaneous Emiss
22K2L6 13
1x10 Resonance Abs
23K1L7
12 Stimulated Emiss
24K0L8 1x10
Coll. Ionization
25K2L7
11 3 Boby Recom.
0.01 26K1L8 1x10
Rad. Recombination
27K2L8 10
1x10 Photoinization
9 Stimulated Rad. Recomb.
1x10
Autoinization (Auger)
8 Resonant Capture
1x10
0.001 7
1x10

1.2.Why do we need lasers? 1x10

100000
0
6

2x10 -14 4x10 -14 6x10 -14 8x10 -14 1x10 -131.2x10 -13
1.4x10 -13
1.6x10 -13
1.8x10 -13

0.0001
-14 -14 -14 -14 -13 -13 -13 -13
0 2x10 4x10 6x10 8x10 1x10 1.2x10 1.4x10 1.6x10

To get very high energy in a very short
time to investigate the extremely hot and
3.References
dense matter (inertial confinement fusion [1] A.G. de la Varga et al.High.Ener.Dens.Phys.7.163(2011)
and astrophysics) [2] A.G. de la Varga et al. High.Ener.Dens.Phys.9.542-547(2013)
[3] A.Bar-Shalom et al. Physical Review A.40.3183-3192(1989)
To get X-rays of such shortness and [4] slacportal.slac.stanford.edu
precision that we can “take a picture” of [5] M.F.Gu, FAC 1.1.1 Manual(2001)
atoms and molecules 4.Acknowledgements
This project has received funding from European Union 's
Horizon 2020 research and innovation programme under
grant agreement of number 665207.
Environmental sustainability assessment of
municipal solid waste management through
carbon footprint
aJavier Pérez, aJulio Lumbreras, aEncarnación Rodríguez
aDepartment of Chemical and Environmental Engineering.
Universidad Politécnica de Madrid, José Gutiérrez Abascal 2, 28006- Madrid.
(javier.perez@etsii.upm.es)

Introduction
One of the most challenging issues for building sustainable cities is the improvement of municipal solid waste management (MSWM), which requires a
substantial effort to reduce its production and improve its different stages: collection (or pre-collection/containerization), transport and treatment. Each
of these stages has environmental impacts stemming from the use of bags to hold the waste generated by residents in their homes, from the containers
placed in public roads for drop-off, from the use of lorries or systems to transport waste to the processing plants, and from the construction and operation
of plants to treat each waste fraction. In Europe, the Waste Framework Directive specifies Life Cycle Assessment (LCA) as a necessary policy-making tool,
ensuring that impacts are assessed from cradle to grave, and avoiding ‘hiding’ impacts by moving them to other countries or stages of
production/consumption. One of the environmental impacts evaluated is climate change (CC), in which greenhouse gas (GHG) emissions from the whole
life cycle are calculated: carbon footprint (CF). This poster shows a methodology, based on LCA standards (ISO 14040 / 14044 / 14067), to calculate CF of
MSWM. This methodology was applied to Madrid City.

Case study: Madrid City
In Madrid, 344 kilograms of MSW are produced per inhabitant. They are collected separately in
different fractions (F1=mix waste, including organic material, F2=packaging, F3=paper/cardboard and
F4=glass). F1 and F2 fractions are managed in Valdemingomez Technology Park (VTP), comprising three
sorting plants and material recovery facilities (SP/MRF), one waste-to-energy plant (WtE), one
composting complex, two anaerobic digestion plants, one landfill, and one installation to use biogas
recovered from a sealed landfill. The remaining fractions are taken to authorized handlers, regardless
of whether they are part of integrated management systems (Fig. 1) or not, as is the case with F3 and
F4 (the other two significant fractions).
Fig. 1. Basic MSWM scheme for the four main MSW fractions in the
Methodology and results city of Madrid

1- Collection (containerization) 2- Transport
Collection CF is calculated using SimaPro software (v. 8.0.5.13). System The boundaries of the system include both the fuel life cycle (FLC) and the
boundaries: raw materials extraction and processing and container vehicle life cycle (VLC). CF is calculated taking into account actual data about
production. Functional unit: one ton of MSW collected. Inventory data: the fleet and fuel consumption, and using Copert 4.11.3 and GlobalTRANS (a
Ecoinvent 3.1, container manufacturers, Madrid City Council. The collection tool developed at UPM). In the city of Madrid, MSW transport vehicles run on
impact per unit of waste is 3.6 kg CO2 eq/tMSW collected. Fig. 2 shows total GHG compressed natural gas (CNG). Fleet’s CF is 25 kg CO2 eq/tMSW collected, 92% of
emissions distribution by MSW fraction collection and type of container which stems from FLC and the remaining 8% from the VLC. In terms of FLC,
(left), and collection CF per district (right) 86% of the impact comes from the Tank-to-Wheel (TtW) stage and 14% from
the Well-to-Tank (WtT) stage (Fig. 3). The Madrid case is compared with other
Spanish cities and past scenarios in Madrid, when vehicles ran on diesel. It
CC
Impact / ton coll. has also been evaluated a possible future where natural gas is replaced by
purified biogas from the anaerobic digestion of municipal waste, in order to
kg CO2 eq / yr / t

reduce CF (Fig. 4).

Municipal mean
3,907 t CO2 eq / year
26.6 g CO2 eq / litre
3.59 kg CO2 eq / tMSW·year
1.22 kg CO2 eq / inhab·year

Fig. 2. Collection CF. Distribution of GHG emissions per fraction and type of container (left). Impact Fig. 3. FLC (left) and VLC (right) GHG
per district (right) emissions distribution Fig. 4. Comparison between transport scenarios
3- Treatment Table 1. Proposed alternative scenarios to compare with 2013 scenario
The CF of the waste treatment Scenario
Waste-to-energy
Acronym
WtE
Organic Matter in F1
Waste-to-energy
Organic matter in F2
Waste-to-energy
Packaging in F1
Waste-to-energy
Packaging in F2
Waste-to-energy
Rest from F1 and F2
Waste-to-energy
F3
Waste-to-energy
F4
Recycling
amounts 224 kg CO2 eq/tMSW collected. It Waste-to-energy + Recycling
Waste-to-energy + All Recycling
WtE+R
WtE+aR
Waste-to-energy
Waste-to-energy
Waste-to-energy
Waste-to-energy
Waste-to-energy
Recycling
Recycling
Recycling
Waste-to-energy
Waste-to-energy
Recycling
Recycling
Recycling
Recycling

has been calculated taking into account Total disposal
Landfilling
L
L+R
Landfilling
Landfilling
Landfilling
Landfilling
Landfilling
Landfilling
Landfilling
Recycling
Landfilling
Landfilling
Landfilling
Recycling
Landfilling
Recycling

direct and indirect GHG emissions, and Landfilling + All Recycling L+aR
Total disposal without biogas recovery LwBr
Landfilling
Landfilling
Landfilling
Landfilling
Recycling
Landfilling
Recycling
Landfilling
Landfilling
Landfilling
Recycling
Landfilling
Recycling
Landfilling

burden avoided. The distribution of
Composting C+aR Composting Composting Recycling Recycling Landfilling Recycling Recycling
Biomethanation/Composting B+C+aR Biomethanation/Composting Biomethanation/Composting Recycling Recycling Landfilling Recycling Recycling

GHG emissions per individual
Fig. 5. Direct, indirect & avoided GHG emissions
treatment is shown at Fig. 5 (2013 distribution per treatment in the 2013 scenario
scenario).
This situation was compared with nine alternative scenarios (Table 1), which
describe hypothetical management routes (among those already implemented in
the city) for the different MSW fractions. Results obtained show that scenarios
based on a total recovery of valuable materials and WtE or anaerobic digestion
treatments present the lowest CF (Fig. 6).
The total CF of MSWM of Madrid City is 253 kg CO2 eq/tMSW collected, from which
1.4% corresponds to the collection stage, 9.9 % to transport and 88.7% to final
treatments.

Acknowledgements Fig. 6. Comparison between scenarios: CF disaggregation per type of GHG emission
considered
The authors thank to the Ministry of Economics and Competitiveness, which finances the research project called “Optimization of municipal solid waste
(CTQ 2013-48280-C3-2R)”, whose results are partially showed in this work, and the Madrid City Council, which provides the municipal information.
Improvements of air quality simulations of Tar-industrial
--------------------------------

particulate matter in the Iberian peninsula with Tecnologías Ambientales y
Recursos industriales
http://tarindustrial.etsii.upm.es

road dust resuspension and dust emission modules
aDavid de la Paz, aRafael Borge, bMichel Vedrenne
aChemical & environmental engineering department. (Technical University of Madrid,
Spain. (dpaz@etsii.upm.es)
bAir and Environment Quality, Ricardo Energy and Environment, London, UK
Abstract Meteorological Model
The Mediterranean Basin is one of the most sensitive zones in the planet to air pollution. Besides
anthropogenic emissions, the Iberian Peninsula is affected by natural particule matter (PM) sources. Dry WRF
Static data Static data
weather conditions allow road dust to be resuspended by vehicle-induced turbulence. These emissions + Dust sources
Meteorological + Emission factor by
may be as important as those directly emitted by vehicles (combustion and from road abrasion and tyre + Soil texture
+ Particle diameter variables
type Vehicle
+ nº Vehicle/day
and break wear processes). In addition, it is widely documented that crustal particles transported from + Satellite-data (Modis) (WV,WD, Rain….) + Average speed

the Saharan Desert may contribute significantly to ambient PM concentration in Madrid. This poster
summarizes the methodology and results of the implementation of a module to account for these Dust emisions Road dust
(DEM) Resupension
sources –not included in current emission inventories- into the Eulerian air quality model CMAQ (Figure
Static data
1). Comparison of predicted PM concentrations with ambient air quality measurements and satellite + Emission inventory
observation point out that this module clearly improves the performance of the standard CMAQ version. + Temporal profiles
+ Spatial surrogates SMOKE CMAQ
+ Chemical speciation
Methodology Emission Model Chemical transport model

Dust emissions from natural sources (DEM) Road Dust emissions from traffic
A domain consisting of 124 x 108 cells with 48 x 48 km2 spatial resolution, The modeling domain is centered in the region of Madrid and comprises 40
centred at 34°20'N, 9°18'E has been used in order to capture synoptic x 44 cells with 1 × 1 km2 spatial resolution (Figure 3). The calculation of the
features and general circulation patterns (Figure 2). The implemented dust contribution of PM resuspension (including suppression by rain) to PM10
sources spatial distribution follows the same scheme than the GOCART ambient levels was implemented by means of as a scaling factor inside the
model (Chin et al., 2000). Simulations were conducted for the entire emission model. Emissions from resuspension were calculated by vehicle
month of July 2007. The dust emission scheme incorporated by DEM is the type using mobility and traffic data for the Madrid urban area. It is
one described in Gillette and Passi (1988). This scheme provides an assumed that the vehicle weight is the most influential factor The results of
equation for the estimation of the vertical flux of dust (F): the air quality model were compared against observations aided by
?
statistical indicators of performance.
? = ? ? ?????? (???? − ??) ?? ???? > ??
E= K (SL)0.91 (W)1.02
F = vertical flux of dust E = PM10 emission in gVKT−1
C = empirical coefficient (1 µg s2 m-5) K = particle size multiplier (g VKT-1)
S = source function (unitless) sL = silt load (g m−2)
Sp = fraction of each size class W = average weight vehicle of the fleet (tons)
W10m = horizontal wind speed at 10m (m/s) 5 5
Ut = threshold velocity (m/s)
Dp = Diámetro partícula (m/s)
???????? (??ℎ ??/???)? = ??? (??ℎ ??/???)?
⍴ y ⍴a = Densidad de la partícula y el aire ?=1 ?=1
w = humedad ??? = ???????? ??????
Figure 2. Geographic distribution
??,0 1.02 + 0.2 ???10 ? , ?? ? < 0.5 of the dust source function Ratio EF
Figure 3. Domain simulation
??,? = VehicleType
∞ ?? ? > 0.5 (EFi/EFreference) (mg/veh·km)
The precipitation threshold is
⍴pgDp Passenger cars 1.00 65.30
0.0013 0.006 set at 0.254 mm/h. It is
??,? = 1+ Taxis 1.00 65.30
⍴a ⍴pgD1.56
p Light duty vehicles 3.56 232.73 assumed that resuspension of
1.928 1331 D1.56
p + 0.380.092 − 1 Trucks 19.86 1296.70 PM recovers after between 2
DEM is integrated through an off-line approach, which means that it can Motorcycles 0.09 5.84 to 4 hours from the time rain
be executed independently of the rest of the emissions. Dust particles Table 1. Resuspension average ceases (de la Paz et al., 2015).
(from 0.1 to 10 µm diameter) and were grouped into 5 size bins (de la Paz emission factors
et al, 2013). The resuspension scheme was initially tested at microscale using the OSPM
Conclusions street-canyon model. Predicted PM10 concentrations with resuspension
were much closer to observations than those without resuspension in
This paper presents the implementation and application of a dust module
specific streets where experimental data was available. Then it was
(DEM) into the WRF-SMOKE-CMAQ mesoscale modelling system (AQM).
implemented in CMAQ, substantially improving PM10 predictions
Two locations strongly affected by dust transport were the DEM module
(increasing annual average concentration up to 9 µg m-3) (Figure 5)
successfully captures the Saharan influence, improving thus the PM10
predictions of CMAQ alone. (Figure 4a). A reasonable reproduction of the a b1 b2
geographical patterns of the dust transport process (Figure 4b)
a b

Blida

Figure 5. (a) CMAQ [w resusp - w/o resusp]. (b1) Taylor PM10 (b2) Taylor PM2.5
Figure 4. (a) Temporal and (b) spatial AOT/PM10 comparison

Acknowledgements References
• Chin, Mian, et al. "Atmospheric sulfur cycle simulated in the global model GOCART:
 Madrid City Council for providing the meteorological and air quality data. Model description and global properties." (2000).
 The TECNAIRE-CM scientific programme is funded by the Directorate General • de la Paz et al. ”Modelling Saharan dust transport into the Mediterranean basin with
for Universities and Research of the Greater Madrid Region (S2013/MAE-2972) CMAQ”. Atmospheric Environment 70 (2013): 337–350
• de la Paz et al. “Implementation of road dust resuspension in air quality simulations
For more information, of particulate matter in Madrid (Spain)”. Frontiers in Environmental Science 04 (2015)
please visit our website • Gillette, Dale A., and Ranjit Passi. "Modeling dust emission caused by wind erosion."
www.tecnaire-cm.org Journal of Geophysical Research: Atmospheres 93.D11 (1988): 14233-14242.
Modelling the effects of local emission
abatement measures in hot-spots of Madrid
*C. Quaassdorff, R. Borge
Laboratory of Environmental Modelling. Chemical and Environmental Engineering
Department. Technical University of Madrid.
*christina.quaassdorff@etsii.upm.es

Abstract
This work aims at estimating NOx and PM10 emissions from road traffic with high spatial and temporal resolution (seconds and meters). This is useful to
understand the effect of local abatement measures and it is an essential input to microscale air quality models. For that, 1-hour representative traffic
scenarios are simulated with the traffic microsimulation model VISSIM in selected hot-spots. Measured traffic data (fluxes and fleet composition) are used
as inputs for the model to obtain speed-time profiles for each vehicle. These profiles are used to calculate specific emission factors for different vehicle
classes according to the VERSIT+micro model through the ENVIVER interface. This modelling system allows assessing possible effects of different local scale
interventions such as vehicle volume reduction, fleet composition and vehicle technology changes, etc. both on the total amount of emission amount and
their spatial distribution. According to this analysis, some local measures may bring about important emission reductions in both locations, up to 28% and
23% for NOx and PM10 respectively. Such measures may therefore constitute a valuable complement for city-scale policies and measures.

Introduction and Methodology
Since pollution levels exceed the legal limits in specific traffic-related urban locations in Madrid it is necessary to develop additional emission reduction
measures on hotspots and highly polluted micro-environments (Borge et al., 2014). However, local characteristics (urban typology, traffic conditions, street
geometry, etc.) play a very important role to accurately predict the effect of any particular measure at this scale. Therefore, there is a need to test
microsimulation models that may reproduce with great detail traffic activity in small areas and may provide reliable emissions (Quaassdorff et al., 2016)
able to feed CFD microscale air quality models (Santiago et al., 2013) and support the definition of effective pollution abatement options.
For this study two different highly polluted microenvironments (hot-spots) in the city of Madrid were analysed. The selected areas are a signalized
roundabout (Fernández Ladreda area) and a combination of three signalized junctions (Escuelas Aguirre area). In this places the emission abatement
measures assessed were:
1. Access restriction to heavy vehicles, which consists in the exclusion of Heavy Goods Vehicles (HGV) during peak hours.
2. Change in the technology of Public Transport buses, from diesel to Compressed Natural Gas (CNG).
Signalized roundabout Signalized junctions
Fernández Ladreda hot-spot Escuelas Aguirre hot-spot

Weekday 8:00-9:00h - morning rush hour Weekday 8:00-9:00h - morning rush hour
NO (g/h)
NOx (g/h)
x NO x (g/h)
NOx (g/h)
45.0 - 65.0 45.0 - 65.0

30.0 - 45.0 30.0 - 45.0

23.0 - 30.0 23.0 - 30.0

18.0 - 23.0 18.0 - 23.0

15.0 - 18.0 15.0 - 18.0

12.0 - 15.0 12.0 - 15.0

10.0 - 12.0 10.0 - 12.0

8.0 - 10.0 8.0 - 10.0

6.0 - 8.0 6.0 - 8.0

5.0 - 6.0 5.0 - 6.0

4.0 - 5.0 4.0 - 5.0

3.0 - 4.0 3.0 - 4.0

2.5 - 3.0 2.5 - 3.0

-27% 2.0 - 2.5
1.5 - 2.0
-2% -22% 2.0 - 2.5
1.5 - 2.0
-2%
1.0 - 1.5 1.0 - 1.5

0.7 - 1.0 0.7 - 1.0

0.5 - 0.7 0.5 - 0.7

0.2 - 0.5 0.2 - 0.5

0.0 - 0.2 0.0 - 0.2

M1: HGV restriction M2: Buses to CNG M1: HGV restriction M2: Buses to CNG NO (g/h)
x
NOx (g/h)

NOx (g/h)
NOx (g/h)
NOx (g/h)
NOx (g/h)
45.0 - 65.0
45.0 - 65.0
30.0 - 45.0
45.0 - 65.0 30.0 - 45.0
23.0 - 30.0
30.0 - 45.0 23.0 - 30.0
18.0 - 23.0
23.0 - 30.0 18.0 - 23.0
15.0 - 18.0
18.0 - 23.0 15.0 - 18.0
12.0 - 15.0
15.0 - 18.0 12.0 - 15.0
10.0 - 12.0
12.0 - 15.0 10.0 - 12.0
8.0 - 10.0
10.0 - 12.0 8.0 - 10.0
6.0 - 8.0
8.0 - 10.0 6.0 - 8.0
5.0 - 6.0
6.0 - 8.0 5.0 - 6.0
4.0 - 5.0
5.0 - 6.0 4.0 - 5.0
3.0 - 4.0
4.0 - 5.0 3.0 - 4.0
2.5 - 3.0
3.0 - 4.0 2.5 - 3.0
2.0 - 2.5
2.5 - 3.0 2.0 - 2.5
1.5 - 2.0
2.0 - 2.5 1.5 - 2.0
1.0 - 1.5
1.5 - 2.0 1.0 - 1.5
0.7 - 1.0
1.0 - 1.5 0.7 - 1.0
0.5 - 0.7
0.7 - 1.0 0.5 - 0.7
0.2 - 0.5
0.5 - 0.7 0.2 - 0.5
0.0 - 0.2
0.2 - 0.5 0.0 - 0.2
! EMT_BusStop
Bus stop
0.0 - 0.2 ! Bus stop
EMT_BusStop

Conclusions
A suitable combination of traffic and emissions micro-simulation models is needed to accurately define emissions and abatement options in hot-spots. NOx
emissions reduction is up to 27% in the case of HGV restrictions which is more effective than a change in the technology of Public Transport buses (2%
emission reduction). These results are promising as inputs for CFD models that may be used to design and test microscale air quality abatement measures.

Acknowledgements References
Funded by the Madrid City Council and the project TECNAIRE-CM Borge, R., et al. 2014. Emission inventories and modeling requirements for the development
(S2013/MAE-2972). of air quality plans. Application to Madrid (Spain). Sci. Total Environ. 466-467, 809-819.
Traffic modelling system VISSIM available by PTV AG and emission Quaassdorff, C., et al. 2016. Microscale traffic simulation and emission estimation in a heavily
modelling system VERSIT+micro/ENVIVER available by TNO. trafficked roundabout in Madrid (Spain). Sci. Total Environ. 566-567:416-427.
To the TARIndustrial group (ETSII, UPM) for the useful advice on Santiago, J.L., et al. 2013. A computational fluid dynamic modelling approach to assess the
emission abatement measures and ArcGis maps representation. representativeness of urban monitoring stations. Sci. Total Environ. 454-455, 61-72.
Assessment of 2D geometries in immune systems responses
using LIFT (Laser Induced Forward Transfer) as a high accuracy
bioprinting technique
aS. Lauzurica, a,bM. G. Fontela, aA. Marquez, bL. Notario,
bP. Lauzurica, aC. Molpeceres
aCentro Láser UPM. C/ Alan Turing 1, 28038 Madrid
bCentro Nacional de Microbiología-ISCIII. Ctra Pozuelo-Majadahonda Km. 2.

Majadahonda, 28220, Madrid, Spain

Abstract
The immune system is a very complex system that comprises a network of genetic and signalling pathways subtending a network of interacting cells. The
location of the cells in a network, along with the gene products they interact with, rules the behaviour of the immune system. In order to acquire a better
understanding of these processes, laser assisted bio-printing techniques emerges as one of the promising approaches to organize cells with high spatial
resolution in two and three-dimensional patterns to enable the study the geometry influence in the immune system interactions. In particular, laser assisted
bio-printing techniques using sub-nanosecond laser sources have better characteristics for application in this field, mainly due to its higher spatial resolution,
cell viability percentage and process automation.

Motivation work Results
1. Print different drops of sodium alginate, as close as possible, and with 1. Laser transferred parametric window
one single cell on each of different cell types.
2. Measure, by cytokine reporters, the activation due to the interaction
between different cell types.
3. Measure, by optical microscopy, the migration of different cell types
through different chemokine gradients.

Immune system network
· Immune response is ruled by the
communication between APCs and T
lymphocytes.
· The relative location of these cells in a
network, along with their gene
products, rules the behavior of the
immune system.
· Dendritic cells (APCs)/T lymphocyte 2. 2D geometry assessments Donor layer
interaction acts as an axis for the Nº cells control Lymphocytes Dendritic cells
coordination of different cell types
through the liberation of cytokines,
chemokines and growing factors that
we can track ‘in vivo’.

LIFT/BA-LIFT Geometrical distribution
LIFT (Laser Induced Forward Transfer) is a direct writing technique that Scale bars: 100µm
(Both cell types)
allows to print from low viscosity liquids to solid materials, and can be 3. Cell viability (flow cytometry)
used for additive manufacturing.

Conclusions
 LIFT technique is one of the most promising process for direct write
BA-LIFT (High precision cell printing) of living cells due to its high cell viability, high speed, outstanding
spatial resolution and scalability for high throughput cell printing.
 In this work we have presented a very simple BA-LIFT approach, using
Laser set-up and plasma treatment a polyamide layer for blister generation.
 Although an important number of LIFT applications for cell printing
Plasma conditions: are currently in development, specially in the field of tissue
· Glass vacuum chamber, 850±50 mtorr engineering, until now there are no specific application in the field of
· Time of plasma exposition: 15s. basics immunology.
 We have demonstrated that it is possible to use BA-LIFT to print free
patterns of lymphocytes and dendritic cells (antigen presenting cells),
Influence in hydrogel layer adherence
being possible to print single cell by controlling the cell density in the
liquid matrix and the droplet size.
 This approach is key to generate specific patterns of cells to study the
immune response ‘in vivo’, combining very novel techniques as
· Distance between transfer and
CRISPR/Cas9 that allow us to visually asses this response.
receiver substrate: 520µm.
Laser Direct Writing Techniques for Metallization
of Solar Cells
aDavid Munoz-Martin, aYu Chen, aMiguel Morales, aDavid Canteli,
aCarlos Molpeceres
aCentroLáser UPM, Alan Turing 1, 28031 Madrid, Spain
david.munoz@upm.es

Abstract
Front metallization is an expensive, fundamental step in the fabrication of solar cells. It comprises different processes including the printing of fingers and
busbars and several thermal treatments in order to cure and sinter the printed material. These two steps have been replaced by laser direct writing
techniques, what allows printing fully personalized metallization designs, with high throughput velocities and a lower thermal budget for the solar cell
structure.

Laser-Induced Forward Transfer (LIFT) & Laser Thermal Treatment
Metallization Curing & sintering
Technique LIFT CW laser local heating
Spectra Physics Explorer Spectra Physics Millenia
 Pulsed laser (15 ns)  Continuous Wave
Laser source
 532 nm  532 nm
 Pulse Energy up to 40 µJ  Power up to 20W

 Optical Scanner (ScanLab HurryScan)
Optical path  F-Theta Lens, focal 250 mm
 Focused beam diameter ∼20 µm

LIFT Laser local heating

LIFT printing of fingers and busbars Curing the printed material
• Commercial, high viscosity, non-Newtonian, micron-sized particles, • Lines (widths down to 80 µm and heights up to 40
silver paste (DuPont Solamet PV17F). µm) have been printed onto solar cell materials: c-Si
• Concrete-dot transfer regime: the protruding wafers and CIGS flexible solar cells.
paste reaches the acceptor substrate, allowing • Metallization finishes with a series of heating steps
the formation of a continuous pillar between
donor and acceptor.
• Highly textured surfaces provide high adhesion
forces and prevent the paste from spreading
along the surface.
CIGS
solar cell

Standard
Y. Chen et al. furnace
Results in Physics 6, treatment
D. Munoz-Martin et al. Applied 998-999 (2016)
Surface Science 366, 389–396 (2016)

Conclusions Laser Curing:
• The physics of LIFT of high viscosity fluids has been studied. High aspect Low power (∼0.5 W)
ratio lines (∼0.5) can be printed. & low speed (∼mm/s).
• All-laser based metallization process has been applied: LIFT printing and Several passes were
laser thermal treatment for curing and sintering. needed.
• Flexible CIGS solar cells have been metallized showing appropriate
functionality

Acknowledgements
This work has been supported by the EUROPEAN COMISSION – APPOLO (FP7-2013-NMP-ICT-FOF. 609355) and the Spanish MINECO projects SIMLASPV-
MET (ENE2014-58454-R) and HELLO (ENE2013-48629-C4-3-R)
Laser functionalization of surfaces
aDavid Canteli, aAndrés Márquez, aDavid Munoz-Martin, bGabriel
Faleiro, 2 bSamuel Santamaría, aSara Lauzurica, aMiguel Morales,
aCarlos Molpeceres

aCentro Láser, UPM, Alan Turing 1, 28031 Madrid, Spain (david.canteli@upm.es)
bEscuela Técnica y Superior de Ingenieros Industriales, UPM, José Gutiérrez Abascal 2,

28006 Madrid, Spain.

Abstract
The treatment of surfaces by physical or chemical methods is a very usual way to change their original properties. Although the most common use of a
surface functionalization is for tribology aims, modifications of the topography that leads to changes in the surface roughness, its hydrophilicity or
hydrophobicity, its light scattering behavior, biocompatibility or even aesthetic changes are possible. In this field, laser sources have proven to be a most
versatile and useful tool, being a clean and fast way to achieve any of those objectives. In this work we show the results of functionalization by laser
texturing of four different materials.

Tribology study in copper Colors in steel
Interaction Problem Solution Laser
1064 nm,
between Micro holes on the surface: irradiation
Friction and • Micro-dynamic bearing I0 I1 I2 ns pulses
surfaces heat effects • Reservoir for lubricant α
na
of steel
• Micro trap for wear debris no
Thermal
β nm process
Laser irradiation
Tribology functions Constructive Oxide film grow
355 nm,
and destructive in the surface
ns pulses
interferences MnCr2 O4 , Fe3 O4 , etc
• Fast (t < 1s)
• Clean (no chemicals)
Surface treatment for
Different oxide film
• Versatile (freeform designs) aesthetical purposes
thickness and
refractive index

Design Different colors
Spherical-cap shaped holes
200 um wide
3D image and profile of a hole
300 um deep
made by laser in a copper sample.

Design
Light absorption enhancement ‘Woman and bird at night’ from
the painter Joan Miró.
Higher light
Photovoltaic absorption Better
devices efficiency Scribing of metallic prostheses
AZO films Surgical Need to be Biocompatible
Higher light scattering Osseointegratable
355 nm, prostheses
ps pulses New strategy
1. Titanium alloy samples
by powder metallurgy

2. Surface texturing
Haze factor up to 73% in 300- Laser irradiation
1400 nm range.
532 nm,
Crystalline silicon Non-
Lower reflectance ps pulses thermal Electronic Scanning Microscope
image of the surface of a
1064 nm,
3. Enamel adhesion Titanium sample after laser
ps pulses (with Ti and Pd particles) treatment.

Conclusions
• Clean Technique suitable
Laser
Total reflectance from 30% to 9% • Fast for a wide range of
functionalization
Better than with chemical etchings • Accurate fields
of surfaces
• Versatile

Acknowledgments. This work has been partially supported by the Spanish Ministry of Economy and Competitiveness under project HELLO ENE2013-
48629-C4-3-R.
Laser crystallization of silicon and study by
finite element modelling
aDavidCanteli, aDavid Munoz-Martin, aMiguel Morales, aCarlos
Molpeceres
aCentro Láser, UPM, Alan Turing 1, 28031 Madrid, Spain
david.canteli@upm.es

Abstract
Laser crystallization of amorphous or microcrystalline silicon films to obtain high-quality polycrystalline films is one of the most promising methods for
diminishing costs in the microelectronic and solar cells sectors. During a laser crystallization process light is partially absorbed by the amorphous silicon,
heating the sample and, if the temperature rises high enough, causing the reorganization of the film structure into a crystalline one. In this work we show
results on the crystallization of non-hydrogenated silicon thin-films by a continuous wave infrared laser, as well as a study of the process with a simple
finite elements method (FEM) numerical model based in the dimensional non-linear heat transfer equation with a steady heat source.
Results
Laser source and samples Experiments results
Laser source Low laser intensity High laser intensity
• Continuous wave (CW) at 980 nm Low speed High speed
• Linear spot: 19,6mm (Top hat) x 2,16mm (Gaussian) Untreated Melting onset Liquid phase melting Dewetting onset
Allows the treatment of large areas

Samples (IEEE Journal of
Photovoltaics, Vol. 4, Nº. 6,
November 2014)
Silicon
Buffer layer
Laser
head Glass substrate

Sample • Heat resistant substrate
• Buffer layer
• Heated to 700ºC
Heating plate Prevent cracks
by thermal
stress
Motorized Process parameters:
axis • Power: 300 – 1300 W
• Speed: 1 – 25 mm/s

FEM modelling
Finite Elements Method (FEM) model.
• 2D stationary model
silicon film + glass substrate

• non-linear heat transfer equation
• Including phase change (1687 K)

Results from the FEM model of the laser process with different process parameters.

Phase change 1
exp
− T − Tm 2 FEM model results
π∆T ∆T 2
• Good agreement between calculations and experiments.
Source term S ?, ?, ? = ?? ?, ? 1 − ? ? ? ? ??? −? ? ? • Irradiation time is a key variable in the crystallization process.
2
? • The model can predict the start of the liquid phase crystallization process
Fluence profile ? ? = ?? ??? −2
?? can be used to study the onset of the dewetting of the silicon.

Conclusions
• A continuous wave IR laser source emitting at 980 nm has been used to crystallize amorphous silicon (a-Si) thin films.
• Laser-annealed films with high crystalline quality (grains of several mm long) have been obtained.
• A simple thermal FEM model has been developed in COMSOL Multiphysics to simulate the process by numerically solving the two dimensional non-
linear heat transfer equation with a steady heat source.
• The local temperature evolution in the irradiated area given by the FEM model shows good agreement with the experiment results.
• The model helps to determine the experimental parameters needed for crystallizing a-Si without damaging the film.

Acknowledgments. This work has been supported by the Spanish Ministry of Economy and Competitiveness under project HELLO ENE2013-48629-C4-3-R.
Numerical study on the effects of dimensionless
parameters on Laser Induced Forward Transfer
printability range
Miguel Moralesa, Iván Lópezb, David Munoz-Martina, Carlos Molpeceresa
aCentro Láser, UPM, Alan Turing 1, 28031 Madrid, Spain (miguel.morales@upm.es)
bETSI Industriales, UPM, José Gutiérrez Abascal 2, 28006 Madrid, Spain

Introduction Ba-LIFT
Laser-Induced Forward Transfer (LIFT) Blister actuated Laser-Induced Forward Transfer
• non-contact direct-write technique
• deposition of small volumes of material (even < pL)
• can print user-defined high-resolution patterns
• similar functionality to droplet-on-demand (DOD) inkjet printing
without nozzle clogging
• can print from solids to low viscosity fluids, including all
intermediate viscosities and also complex fluids.

FEM model – Two Phase Flow
Phase Field Equations = 0: ???
? Blister Movement Adaptive Mesh
? = 6 0 < ? < 1: ???ℎ ?ℎ???? Refinement
? = 1: ???
?? ??
+ ? & ?? + ? ?& ? 1−? − ?? & ?? = 0
?? ??

Navier-Stokes Equations
??
? + ? & ?u − ? & ?(?? + ??- ) + ?? = ?⃗34 ; ? & ? = 0
??

inertial forces ρ dv
Re =
viscous forces
=
µ Printability Map
inertial forces ρ dv inertial forces ρ dv inertial forces ρ dv 2

Re = Dimensionless
= =
ReParameters = = =
viscous forces µ viscous forces
We
µ cohesion forces γ γ = 40.79 mN/m μ= 5 mPaˑs (Z = 4,10)
ρ dv 2
Fix Parameters
inertial forces
inertial forces ρ dv
ρ dv 2 inertial forces viscous forces µ
Re =
We =viscous forces =µ = We = Oh == = Laser Energy E = 5.9 μJ
cohesion forces γ cohesion forces γinertia forces ⋅ cohesion forces ργ d
µ
Droplet diameter d = 10 μm
inertial forces ρ dv 2 Oh =
viscous =forces µ viscous forces1 =
We
Oh = = Z=
cohesion forces⋅ cohesion
inertia forces ργ dforces ⋅ cohesion
γ forces inertia Oh forces ργ d
1 viscous forces 1 µ
Z ==
Oh Z= = 1000
Ohinertia forces ⋅ cohesion forcesOh ργ d Non DOD printable
1
Z = γ = 10.40 mN/m μ= 5 mPaˑs (Z = 2,94) DOD printable
Oh
Too viscous
Z=1/Oh < 4
Weber number We

100

Satellite droplets
10
Z=1/Oh > 12

γ = 40,79 mN/m μ= 8 mPaˑs (Z = 2,56)
Insufficient energy for drop formation
1
0,1 1 10 100 1000
Reynolds number Re

γ = 40.79 mN/m μ= 0.6 mPaˑs (Z = 34,17)

Conclusions
Ba-LIFT DOD printability window
We > 4
4 < Z < 12

Acknowledgments. This work has been supported by the Spanish Ministry of Economy and Competitiveness under projects
SIMLASPV-MET (ENE2014-58454-Rand HELLO (ENE2013-48629-C4-3-R).
Crystallization behaviour of PHBV/WS2 inorganic
nanotube nanocomposites
aTyler Silverman, aMohammed Naffakh, bCarlos Marco and bGary Ellis
aETSI Industriales, UPM (tsilverman1@gmail.com)
bInstituto de Ciencia y Tecnología de Polímeros (ICTP-CSIC)

cooling (10ºC/min)
Abstract GO
C60 (5 wt.%)
SWCNT (5 wt.%) 4
5
26

Using tungsten disulphide inorganic nanotubes (INT-WS2) in organic-
inorganic hybrid composite materials provide the potential for improving
MWCNT (0.08 wt.%)
INT-WS2
12
21 PLLA
BN 25
Talc (5 wt.%) 21
thermal, mechanical, and tribological properties of conventional Tb2O3 10
La2o3 10
composites [1]. Recent results have raised new expectations, since it has CNF 1
been observed that well-dispersed INT-WS2 exhibit a much more HNT (3 wt.%)
MWCNT
1
21
PHBV
INT-WS2 24
prominent nucleation activity, improving the overall crystallization process BN 32
of poly(L-lactic acid) (PLLA) and poly(3-hydroxybutyrate) (PHB) than specific Thyamine
α-CD 18
28
Melamine
nucleating agents or other reported nano-sized fillers, such as carbon Lignine 12
16

nanotubes and graphene [2,3]. Similarly, good dispersion and interfacial
Talc
MWCNT 6
10
PHB
adhesion of INT-WS2 in a poly(hydroxybutyrate-co-hydroxyvalerate) (PHBV) GO
INT-WS2
9
36
matrix significantly influence the crystallizability of PHBV. These features 0 5 10 15 20 25 30 35 40
DTp (ºC)
may be advantageous for the enhancement of the mechanical properties
and processability of this new class of WS2-based composite materials. Figure 2. Increment in the ΔTp of PHB and PHBV systems containing
different kinds of fillers.
Results and discussion
The major issue in expanding commercial opportunities for PLLA, PHB and
related polymer materials (PHBV) is their crystallization behavior.
Incorporating well-dispersed nanoparticles like INT-WS2 into biopolymer
can modify the crystallization behavior. Therefore, it is of great interest to G=1/0.5
σe (erg cm-2)
0.5: Time
investigate the nucleation, crystallization, and structural development of required to
=
74.9
the matrix in INT-reinforced biopolymer materials. In this regard, the non- reach 50% 56.6
isothermal melt-crystallization of neat PHBV and its nanocomposites crystallinity 70.5
obtained with different INTs loadings, using a simple solution blending 58.1
method, was investigated by DSC using a Perkin Elmer DSC7/Pyris
differential scanning calorimeter. The results of revealed that the
incorporation of INT-WS2 in small weight percents (1.0 wt.%) dramatically
accelerated the crystallization rate compared to the pure biopolymer Figure 3. Variation of the rate of crystallization of PHBV/INT-WS2
(Figure 1). Nevertheless, for the same nanofiller loading, the improvements nanocomposites as a function of the crystallization temperature (Tc).
in crystallization temperature peak (Tp) for the PHBV(PLLA)-based samples
are lower than those observed for the corresponding binary PHB-INT In the same way, the isothermal crystallization behaviours and kinetics of
nanocomposites (Figure 2). In particular, incorporating 1.0 wt.% INT-WS2 neat PHBV and its nanocomposites were intensively studied. It was found
effectively increased the Tp of PHBV to a value comparable to that of BN, that the overall crystallization rates were reduced with an increase in the
and greater than all other NAs. BN is currently estimated to be more crystallization temperatures for both neat PHBV and PHBV/INT-WS2
expensive than INT-WS2 and, like talc, is not a nanoparticle filler, also nanocomposites. Moreover, at a given crystallization temperature (Tc), the
possibly leading to deficient mechanical reinforcement qualities that are overall crystallization rates were faster in the nanocomposites than in neat
observed with other nano-sized filler materials. PHBV and increased further with an increase in the INT-WS2 loading (Figure
3). The addition of INT-WS2 remarkably influences the energetics and
20
PHB kinetics of nucleation and growth of PHBV, reducing the fold surface free
0 0.1 wt.% INT-WS2 energy (σe) by up to 25.4 %.
0.5 wt.% INT-WS2
-20
1.0 wt.% INT-WS2
Conclusions
Endo

-40

PHBV
dQ/dT (a.u.)

-60
1 wt.% INT SEM The results obtained from the DSC crystallization experiments, along with
0.1 wt.% INT-WS2
-80
Tp
the good materials processability, economic and sustainability benefits,
-100
0.5 wt.% INT-WS2 encourage further investigations in order to evaluate the full potential of
1.0 wt.% INT-WS2
-120
these new bionanocomposites for their potential use in eco-friendly and
cooling (10ºC/min)
medical applications.
-140

Acknowlegments
40 60 80 100 120 140 160
T (ºC)

150
This work was supported by the Spanish Ministry Economy and
Competitivity (MINECO), Project MAT2013-41021-P. MN would also like to
130
acknowledge the MINECO for a ‘Ramón y Cajal’ Senior Research Fellowship.
110
References
Tp (ºC)

PHB
90 0.1 wt.% INT-WS2 1. M. Naffakh, A. M. Díez-Pascual, Thermoplastic polymer nanocomposites based
0.5 wt.% INT-WS2
1.0 wt.% INT-WS2 on inorganic fullerene-like nanoparticles and inorganic nanotubes. Inorganics
70 PHBV 2014, 2, 291-312.
0.1 wt.% INT-WS2
0.5 wt.% INT-WS2
1.0 wt.% INT-WS2 2. M. Naffakh, C. Marco, G. Ellis. Inorganic WS2 Nanotubes that improve the
50
0 5 10 15 20 25
crystallization behavior of poly(3-hydroxybutyrate). CrystEngComm 2014, 16,
f (ºC/min) 1126-1135.
Figure 1. DSC data of the dynamic crystallization of biopolymer/INT-WS2 3. M. Naffakh, C. Marco, G. Ellis. Development of novel melt-processable
biopolymer nanocomposites based on poly(L-lactic acid) and WS2 inorganic
nanocomposites.
nanotubes. CrystEngComm 2014, 16, 5062-5072.
Electrode modified with Thiolated DAB dendrimers
to form self-assembled monolayers with AuNPs and
HRP for biosensing
aEvelyn Ospina, aM.P. Garcia Armada, bB. Alonso, bC.M. Casado

aDep. Ingeniería Química Industrial y M. A., E.T.S.I.I., U. P. M. (ej.ospina@alumnos.upm.es)
bDep. Química Inorgánica, Cantoblanco, U. A. M.

Abstract
Hydrogen peroxide (H2O2) is not only the product of the reactions catalyzed by many of highly selective oxidases but also an essential mediator in food,
pharmaceutical, and environmental analysis [1]. In this work, Horseradish peroxidase (HRP), as a model enzyme, was immobilized onto AuNPs to develop
an electrochemical biosensor based on self-assembled monolayers (SAMs) of gold nanoparticles (AuNPs) and thiolated DAB dendrimers (DBA) on a glassy
carbon electrode (GC), for the electrocatalytic reduction of H2O2. The SAM is characterized by their high degree of orientation and stability due to the thiol
is covalently bonded to the corresponding metallic substrate, particularly gold [2]. This allow us to have a highly sensitive biosensor.

Scheme of hydrogen peroxide biosensor Direct electrochemistry of HRP

8,0E-05
Au/DBA/Au16/HRP
6,0E-05
Au/DBA/Au16
4,0E-05

2,0E-05

0,0E+00
i (A)

-2,0E-05

-4,0E-05
HRP
-6,0E-05

-8,0E-05

-1,0E-04
-1,5 -1 -0,5 0 0,5 1 1,5
E (V)

3,0E-05 Analitical Characterization
0 microM
2,0E-05
Sensitivity Detection Limit Linear Ranges
50 micro M
Variable Dimension (μA μM -1cm-2) (μM) (mM)
1,0E-05 150 micro M 1-(NHCH2CH2SH)4 0,190 0,065 0,02 – 0,60
3-(NHCH2CH2SH)16 0,293 0,015 0 – 0,60
0,0E+00
DBA 5-(NHCH2CH2SH)64 0,262 0,096 0 – 0,55
i (A)

-1,0E-05 5 nm 0,1829 2,014 0 – 0,2
16 nm 0,394 0,230 0 – 0,35
-2,0E-05 AuNPs 41 nm 0,348 0,030 0 – 0,7
-3,0E-05
Electro 5 cicles 0,3281 0,032 0 – 0,7
deposition 7 cicles 0,394 0,230 0 – 0,35
-4,0E-05 of Au 9 cicles 0,671 0,028 0 – 1,0
-0,8 -0,6 -0,4 -0,2 0 0,2
E (V)

These results demonstrated that the electrocatalytic
The reduction peak current increased with
behaviour of HRP was improved upon its immobilization
increasing concentration of H2O2, indicating
on the AuNP/DBA/AuNPs, allowing an efficient electron
that the catalytic reduction of H2O2 was
transfer between the redox active site of the enzyme
performed on the sensor and the immobilized
and the electrode surface [3], which indicated that
HRP retained its bioelectrocatalytic activity.
AuNP/DBA/AuNPs acted as a promoter of the direct
electron transfer.
References
1. M. Li, S. Xu, M. Tang, L. Liu, F. Gao, Y. Wang, Electrochimica Acta 56 (2011) 1144-1149.
2. P. C. Díaz, F. J. Arévalo, M. V. Fumero, M. A. Zon, H. Fernández, Sendors and Actuators B 225 (2016) Ackowledgements: We thanks the U.P.M and
384-390. the Dirección General de Investigación
3. A. Boujarkhrout, P. Díez, A. Sánches, P. Matínez, J. M. Pingarrón, R. Villalonga, Journal of Colloid (CTQ2009-12332-CO2) for the financial support
and Interface Science 482 (2016) 105-111 in this research
Poly(propyleneimine) Dendrimers: Combined
Experimental and Simulation Study
aSalvador León, bPilar García-Armada, bAndrés Fernández
aE.T.S.I.I., U.P.M (salvador.león@upm.es)
bE.T.S.I.I., U.P.M

Abstract
Dendrimers are multi-branched polymer molecules with interesting potential applications as fuel additives, pharmacological reagents, new materials or
ligands. Particular interest deserve the family of ferrocenyl poly(propyleneimine) dendrimers, thanks to its potential, in modified electrodes as
electrochemical sensors and biosensors, alone or decorated with gold nanoparticles. In this work, the geometrical features of poly(propyleneimine)
dendrimers are investigated by a combination of experimental SEM techniques and molecular simulation. In particular, a new mesoscopic molecular
dynamics simulations has been generated by fitting the energy parameters to pilot atomistic simulations on model compounds. This new model allows the
efficient simulation of the largest dendrimer generations over long time scales, and provides useful insight over the structure and dynamic behavior of
these systems.
Mesoscopic model
The dendrimer molecules are described by a series of two types of beads:
one type corresponding to the propyleneimine groups, and another for
the ferrocene units. Energy parameters are chosen to fit the
conformational properties of an atomistic dendrimer model compound.

Template of ferrocenyl poly(propyleneimine) dendrimer of
fourth generation.

Order 0 Order 2 Order 4 Order 6 Order 8
Ferrocenyl terminal groups
The inclusion of ferrocenyl groups at the end of the dendrimer branches
gives rise to changes in the conformation and structure of the molecules
Dendrimer structure (for instance an increase in the radius of gyration). As a consequence,
The mesoscopic simulations provide a the stratification of the groups distributions arises for lower generations,
description on the structural properties of the different and in particular at order 4 significant space holes arise.
dendrimer generations up to order 8. It can be seen that the radial Order 2
distribution of groups tends to be homogeneous for the smaller
generations (up to order 5), while the larger molecules display certain
degree of stratification in the inner core of the system.
Order 4

Order 6

Applications to electrocatalysis
This study helps to quantify the intra-dendrimer holes and estimate
the possible size of gold nanoparticles formed inside, when the
dendrimers are used as templates. Figures show the SEM
micrograph of platinum a wire modified with (left) an
electrodeposited film of ferrocenyl poly(propyleneimine)
dendrimer of 4th generation and the UHSEM of the same film
which has been used as template to for intra-dendrimer Au
nanoparticles.
Elongation of metallic nanoparticles embedded in a dielectric
matrix produced by irradiation with swift heavy ions
aAlejandro Prada, aO. Peña-Rodríguez, bJ. Olivares, cA. Oliver, cL. Rodríguez-Fernández,
dE. Bringa, aJ.M. Perlado, aA. Rivera
aInstituto de Fusión Nuclear, UPM , Madrid, Spain (alexpradav@yahoo.es)
bCentro de Micro-Análisis de Materiales, UAM, and Instituto de Óptica, CSIC, Madrid, Spain
cInstituto de física, UNAM, México D.F., México
dCONICET and Facultad de Ciencias Exactas y Naturales, UNCuyo, Mendoza, Argentina

Abstract
The aligned rod formation from spherical metallic nanoparticles embedded in a dielectric matrix produced by swift heavy ion irradiation has been known
for many years and it is interesting for many industrial applications. Although diverse experimental and theoretical techniques have already been
employed in order to explain this process, it is still not well-understood. One of the reasons that complicate this task is the lack of information concerning
the intermediate stages of the deformation mechanism. In this work we have studied the elongation of silver nanoparticles embedded in a silica matrix
irradiated with 40 MeV Br ions by means of in situ optical measurements, TEM and molecular dynamics simulations. The agreement between these
techniques is excellent. We have analyzed the nanoparticle deformation after several impacts with different ion energies, and studied the influence of
Ostwald ripening, internal stress and heat diffusion.
Introduction
 Metal nanoparticles are interesting for their many applications (photonics, spectroscopies, imaging, sensing, catalysis, cancer treatment and energy
production) due to their localized surface plasmon resonance (LSPR)
 Nanorods exhibit a LSPR split in two well-defined modes (in red and blue region).
 Ion irradiation with energies >0.1 MeV/u are a good manufacturing method to synthesize and elongate nanorods in the same direction.
 We study the formation of aligned silver nanorods during irradiation by means of in situ transmittance measurements, transmission electron microscopy
(TEM) measurements and molecular dynamics (MD) simulations.
1.2
1.6 13
6x10 ions/cm
2
0.4

Probability
Irradiation rv= 4.8 nm
fluence 0.9  = 2.0
0.2
1.2
Optical density

Optical density
Ion beam () = 0.32 0.0
1.0 1.5 2.0 2.5
Light source

Ion beam
0.8 0.6  = b/a
Sample Spheroid
Light Sphere
CCD array 0.4 0.3 Experiment
spectrometer Fit

Methodology 0.0
300 400 500 600 700 800
0.0
300 400 500 600 700 800
Experiments Wavelength ( nm ) Wavelength ( nm )
 2 MeV Ag implanted (5 x 1016 cm-2 ) + Annealing Spectra fitting MD simulations
 40 MeV Br ion beam up to a fluence of 3 x 1014 cm-2 + absorption  Mishchenko’s T-matrix codes.  MDCASK code.
spectroscopy  Parameters fitted: ɛ (aspect ratio  Simulation boxes 30 x 30 x 28,32 nm3
 HAADF-TEM mean value), σ(ɛ) (aspect ratio  Simulation time 75 ps (normal evolution) +
Before irradiation 8 × 1013 cm-2 1.5 × 1014 cm-2 standard deviation), spheres and 75 ps (Langevin force, accelerated evolution)
spheroids concentration, rv (the  Inter-atomic potentials: Feuston-Garofalini
volume-equivalent radius of the (silica), BMH (silica-silver), EAM (silver)
spheroids) .
 Computation times, more than 1
week.
8 100

3
Concentration (x10 cm )

3
-2

Results
Aspect ratio,  = b/a

6
12

10
4 Regions: 2 2
rv (nm)

 ( )

4
 Region I: pure elongation (cyan region) and absorption of rs
smaller particles by larger ones. 1
 Region II: elongation + dissolution and regrow of larger and 2 1 1
Spheroids
smaller particles. Spheres
 Region III: larger spheroids destroyed and regrow as small I II III IV I II III IV
spheres. 0 0 0.1 0
0 1 2 3 0 1 2 3
 Region IV: particles destroyed and regrow with similar sizes. 14 2 14 2
Fluence (x10 ions/cm ) Fluence (x10 ions/cm )
Pure elongation: 0 2 4
2.2 Experiments Br 40 MeV
 We simulated multiple ion impacts by MD. a
 Each ion induce a small elongation, a cumulative effect is Se= 7.24 keV/nm
Aspect ratio,  = b/a

necessary to obtain rods with larger ɛ. b
1.8 bb
 Small elongation produced by collective displacement of
atoms in track and the high stresses in the interface. a0 = 3.5 nm
1.4 Se =
6 8 10
Conclusions 6 keV/nm
 The final NP elongation depends on a complex competition 1.0 7 keV/nm
between single-ion deformation, Ostwald ripening and 8 keV/nm
dissolution. 0 4 8 12 16
The level of kinetic details obtained from the in situ 0.6
Number of impacts Published in:
measurements is unprecedented.
 A comprehensive theoretical model can be developed. O. Peña et al. Sci. Rep. in review (2017)
In  situ  monitoring  and  modelling  the  op8cal  proper8es  
of  silica  irradiated  with  swi;  heavy  ions  
aPablo  Díaz-­‐Núñez,  aO.  Peña-­‐Rodríguez  ,  aA.  R.  Páramo,  aA.  Prada,  aA.  Rivera,  a,bF.  Sordo,  
aJ.  M.  Perlado,  cM.  Crespillo,  c,dJ.  Olivares  
aIns8tuto  de  Fusión  Nuclear  UPM,  Jose  Gu8errez  Abascal  2,  E28006,  Madrid,  Spain    (pablodn87@gmail.com)  
bConsorcio  ESS-­‐Bilbao.  Edificio  Cosimet.  Paseo  Landabarri,  2  1ª  planta,  48940  Leioa,  Spain  
cCentro  de  Micro-­‐Análisis  de  Materiales  UAM,  28049,  Madrid,  Spain  
dIns8tuto  de  Óp8ca,  CSIC,  c/  Serrano  121,  28006,  Madrid,  Spain  

Abstract  
In  this  work  we  have  used  a  combina4on  of  in  situ  reflectance  to  study  structural  modifica4ons  in  silica  by  ion  irradia4on  (Br  ions  with  energies  ranging  
from  5  to  40  MeV)  and  a  methodology  based  on  Molecular  Dynamics  (MD)  and  Finite  Element  Methods  (FEM)  to  simulate  its  mechanical  response  to  swiG  
heavy  ions.  Reflectance  measurements  allowed  us  to  obtain  the  detailed  kine4cs  of  surface  modifica4on  and  reconstruct  the  refrac4ve  index  profiles  in  
the   irradiated   samples,   which   lead   us   to   very   accurate   results   for   the   track   radius   and   irradia4on   threshold   of   each   ion.   From   these   results   we   have  
observed   that   the   refrac4ve   index   increases   at   first   (for   low   and   medium   fluences)   but   decreases   for   higher   fluences.   Finally,   with   the   combina4on   of   MD  
and   FEM   simula4ons   we   obtained   the   local   densifica4on   generated   by   an   incoming   ion   and   the   strain   and   stress   fields   along   the   material   as   a   func4on   of  
the   irradia4on   fluence.   These   results   allowed   us   to   reproduce   track   satura4on   for   the   low-­‐medium   fluence   regime   and   the   final   decrease   for   the   high  
fluence  regime,  which  is  produced  by  the  plas4c  expansion  induced  by  track  irradia4on  during  irradia4on-­‐enhanced  viscous  flow.  

Experimental  Setup   Op8cal  Model  
Applying  Fresnel  equa4ons:  
•  Experiments  were  carried  out  with  the   !"#$ !%#$ •  A  two  layer  model  (vacuum/Irradiated  
!!" !!" !!"
5   MV   tamdem   accelerator   in   Centro   .$ &"'(()$ .$
SiO2)  allowed  us  to  obtaind  the  surface  
de   Microanálisis   de   Materiales   /$ /$
refrac4ve  index.  

%"
!$"
(CMAM).   0$ !#" *")+,-$ 0$
•  A   mul4layer   model   allowed   us   to  
•  Silica   samples   were   irradiated   with   Br   !#"
8"**(*'& reconstruct  the  refrac4ve  index  profile  
-./01,&
and   F   ions   in   the   electronic   sRoping  
regime  (2-­‐8  keV/nm).   .$ /$ 0$ !'#$
!&234& From  the  surface  refrac4ve  index:  
!"#$%&'()*+,&

•  Nearly-­‐normal   reflectance   was   •  A   Poisson   func4on   was   used   to  
,*& measured  with  an  experimental  setup   $$$$$
calculate  track  radii,  and  
.1&;<

!1$
3$ 3$

!5$
!2$

!.$

!/$
&"'(()$

556&.**.7&

!4$
90 :+ '0,+%*(/,%,*& mounted   inside   the   irradia4on   •  Using   Szenes’   model   we   obtained   the  
chamber  in  the  visible  spectrum.   *")+,-$ threshold   sRoping   power   for   track  
forma4on  

Results  
Measured  reflectance  (R)  for  F  at  5    MeV   Surface  refrac4ve  index  calculated  from  R   Poisson  kine4c  amorphisa4on  for  F  at  5  MeV  
0.044 1.48 0.04
Silica (F 5 MeV) Silica (irradiated)
F 5 MeV
0.042
Unirradiated 1x10 cm
13 -2 0.03
Refractive index

0.040 1.47
Reflectance

14 -2
3.5x10 cm
|(ε - ε0)/ε0|

0.038 0.02 Silica
F 5 MeV Quartz
0.036 1.46 Br 5 MeV
Br 10 MeV 0.01 Δεn = Δεmax[1 - exp(-σφ)]
0.034 Silica (virgin) Br 15 MeV
Br 25 MeV
0.032 Br 40 MeV 0.000 1 2 3 4
500 600 700 800 900
1.45 11 12 13 14
10 10 10 10 14 -2
Wavelength ( nm ) Fluence ( cm )
-2 Fluence ( x10 cm )

Track  radii  obtained  from  the  Poisson  kine4cs   Refrac4ve  index  profiles  reconstructed  with   Comparison  of  the  experimental  and  theore4cal  
 and  fiRed  with  the  Szenes’  model   the  mul4layer  model.   surface  refrac4ve  index  calculated  with  the  FEM  
5 model  
Br 5 10 15 25 40 MeV Silica Unirradiated
4 12 -2 125
0.06 0.13 0.2 0.3 0.5 MeV/amu 5.0x10 cm
4 F 5 MeV
F 5 MeV 13
2.0x10 cm
-2

100
|(ε - ε0)/ε0| ( x10 )
Radius ( nm )

0.26 MeV/amu 3 13
5.0x10 cm
-2
-2

3
Sn ( eV/nm )

14 -2
3.7x10 cm
1.89 keV/nm

75
2
2
50
Silica
1
1 Bromine 25
Fluorine
0 0
0 2 4 6 8 10 0.0 0.5 3.5 4.0
0
4.5
Se ( keV/nm ) Depth ( µm )

Conclusions  
•  In  situ  reflec4vity  provides  a  unique  tool:  fast,  non  destruc4ve  and  easily  controllable  with  automa4c  process.  
•  Thanks  to  this  technique,  we  obtained  detailed  informa4on  of  all  the  refrac4ve  index/density  changes  produced  in  the  material  as  a  func4on  of  the  
fluence  with  a  high  degree  of  detail  and  accuracy,  not  only  at  the  surface  but  also  for  all  the  track  profiles.  
•  The  silica  surface  densifica4on  is  explained  by  means  of  MD  and  FEM  simula4ons.  The  proposed  model  shows  good  agreement  with  the  experimental  
results  and  iden4fies  three  regimes:  i)  the  compac4on  effects  at  low  fluences,  ii)  track  inclusion  expansion  effects  at  medium  fluences  and  iii)  irradia4on  
enhanced  viscous  flow  at  high  fluences.  
Influence of grain boundaries on the radiation-induced defects and
H
a
retention in
a
nanostructured
a,b c
and coarse-grained
a
W
M.Panizo-Laiz, G.Valles, C.González, I.Martín-Bragado, R.González-Arrabal,
aN.Gordillo, dR.Iglesias, aC.L.Guerrero, aJ.M.Perlado, aA.Rivera.
aInstituto
de Fusión Nuclear UPM, José Gutiérrez Abascal, 2, 28006, Madrid, Spain
(miguel.panizo.laiz@gmail.com)
bDepartamento de electrónica y Tecnología de Computadores, Universidad de Granada.
cIMDEA Materials Institute, C/ Eric Kandel, 2, 28906, Getafe, Madrid, Spain
dDepartamento de Física, Universidad de Oviedo, C/ Calvo Sotelo, s/n, Oviedo, Spain

Abstract:
The influence of grain boundaries (GBs) on the radiation-induced defect evolution and on H retention at 300 K have been studied experimentally and by
computer simulations. Coarse grained tungsten (CGW) and nanostructured tungsten (NW) samples were implanted with H and C ions at energies of 170 keV and
665 keV respectively. Three sets of experiments were performed: (i) H single implantation, (ii) C and H co-implantation and (iii) C and H sequential implantation.
Computer simulations were performed by using the Object Kinetic Monte Carlo methodology parameterized by Density Functional Theory data. The same
experiments were simulated in monocrystalline tungsten (MW) and NW.
Experiments [1] and OKMC simulations [2]
Sample code Type of W Type of study H irradiation C irradiation Comment
Energy (keV) Fluence (cm-2) Energy (keV) Fluence (cm-2)
MW-H Monocrystalline Computational 170 5x1016 665 5x1016 -
CGW-H Coarse-grained Experimental 170 5x1016 665 5x1016 -
NW-H Nanocrystalline Comp. and exp. 170 5x1016 665 5x1016 -
MW-Co-CH Monocrystalline Computational 170 5x1016 665 5x1016 C and H simultaneous implantation
CGW-Co-CH Coarse-grained Experimental 170 5x1016 665 5x1016 C and H simultaneous implantation
NW-Co-CH Nanocrystalline Comp. and exp. 170 5x1016 665 5x1016 C and H simultaneous implantation
MW-Seq-CH Monocrystalline Computational 170 5x1016 665 5x1016 H implanted after C implantation
CGW-Seq-CH Coarse-grained Experimental 170 5x1016 665 5x1016 H implanted after C implantation
NW-Seq-CH Nanocrystalline Comp. and exp. 170 5x1016 665 5x1016 H implanted after C implantation
OKMC simulations [2] Ebinding Experimental analysis: RNRA [2]
Cluster ? ?? ??
(eV) ??( ??, ??) ??
HV (HV) 1,20
Emigration
Defect H2V (HH2V) 1,19
(eV)
H 0,205 H3V (HH3V) 1,07

SIA 0,013 [3] H4V (HH4V) 0,95

V 1,66 [3] H5V (HH5V) 0,87
H6V (HH6V) 0,52
[4], [5]
Results
SRIM vs OKMC OKMC: Number of H per vacancy Experimental vs OKMC

Conclusions:
•GBs have a clear influence on the
distribution of vacancies
•H retention is highly affected by the References:
GBs and the vacancies concentration [1] R. González-Arrabal et. al./ J. Nucl. Mater. 453
•The size of HnVm clusters slightly (2014) 287-295
depends on the presence of GBs [2] G. Valles et al. / Acta Materialia 122 (2017) 277-286
•GBs act as preferential paths for H
[3] C. S. Becquart et. al. / J. Nucl. Mater. 403 (2010) 75-88
diffusion.
•NW is very promising for PFM [4] C. Guerrero et. al. / J. Mater. Sci. 51 (2015) 1445-1455
applications. [5] C. González et. al. / Nucl. Fusion 55 (2015) 113009
Water absorption in polypropylene-cellulose
ecocomposites
aA. González-Defix, bM.U. de la Orden, cC. González-Sánchez, aF.R. Beltrán,
aV. Lorenzo, aJ. Martínez Urreaga.
a Dpto. Ingeniería Química Industrial y Medio Ambiente, Grupo “Polímeros, Caracterización y
Aplicaciones”, E.T.S.I. Industriales, UPM, Madrid (joaquin.martinez@upm.es)
b Dpto. Química Orgánica I, Facultad de Óptica y Optometría, UCM, Madrid.
c Dpto. Ingeniería Química y TMA, Universidad de Oviedo

Abstract
Polymer composites made from cellulosic reinforcements are low cost materials with low density and good mechanical properties, and are more
sustainable than the conventional composites made from mineral reinforcements, such as glass fibers, because they are obtained, at least in part, from
renewable resources. Therefore, cellulose-reinforced composites, sometimes called ecocomposites, are increasingly used in different fields of application,
including construction and automobile industries.
There are also some disadvantages. Due to the hydrophilic nature of the cellulose, these materials can show an important water uptake, which leads to
swelling and results in a reduction of the service life. In this work the effects of cellulose treatments, composition and coupling agent on the kinetics of the
water uptake in polypropylene-cellulose ecocomposites have been studied.

Materials Methods
Cellulose: E = eucalyptus kraft pulp; EB= bleached eucalyptus kraft pulp; EBR = Composites with 30, 40 and 50 wt. % cellulose and 1.5 wt. % coupling
bleached and refined pulp. Polypropylene (PP): Moplen HP 648U homopolymer agent were melt compounded in a Berstorff ZE25 twin-screw extruder.
(Basell). Coupling agents: MAPP = maleic anhydride modified PP polymer Water uptake was measured gravimetrically at 25-55 ºC.
(Epolene E-43, Eastman); PEI = polyethyleneimine (Lupasol PR 8515, Basf) Absorbed water was characterized using IR spectroscopy.

Results 6 Effect of treatments
of cellulose fibers
Effects on water uptake
Water uptake (wt. %)

Effect of coupling agent
4 8
10
Effect of celulose content
Water uptake (wt.%)
6
8 2
Water uptake (wt. %)

Bleached and refined 4
6 Bleached
0 Untreated
4 2 No coupling agent
0 200 400 600 800 1000
coupling agent: maleic
2 Time (h) coupling agent: PEI
0
30 wt. %
40 wt. % 0 200 400 600 800 1000
0
50 wt. % Time (h)
0 200 400 600 800 1000
Time (h)
Water absorption FTIR difference spectrum showing
3590 cm-1
studied by IR “free” water
how the water is absorbed in the
Water absorption: spectroscopy
ecocomposite with 30 wt.% of EB
and 1.5 wt. % MAPP.
Fickian model

3160 cm-1
Bound (to
cellulose)
water

3400 cm-1
¿liquid water?

Conclusions:
 Water uptake increases with the cellulose content, the treatments of the fibers
(such as bleaching and refining) and the use of hydrophilic coupling agent.
 Water absorption can be accurately described by a Fickian model in some cases.
 IR spectra show the presence of free water, water bound to cellulose and,
probably, liquid water in the ecocomposite.
 The undesirable water uptake can be reduced by choosing the cellulose fibers and
the coupling agents.
Definition of a secondary metallurgy process for
the modification of oxide inclusions by Ca
treatment in Al-killed high speed machining steels
aIzaskun Alonso, bMartin Berg
aSidenorInvestigación y Desarrollo, S.A. (izaskun.alonso@sidenor.com)
bKTH Royal Institute of Technology (martinb7@kth.se)

Abstract
The need of constant improvement of mechanical and machinability properties of the steel is increasing. Advances in steelmaking during the last six
decades have resulted in steel grades with very low level of impurities, giving to the steel excellent mechanical properties as fatigue resistance. But
machining of high-cleanliness steels is generally associated with high energy consumption, an increased cutting tool wear, and high manufacturing costs.
Consequently, the remaining issue is assessed as to optimize today’s steel grades with respect to the combined machinability and performance
requirements. In conclusion, non-metallic inclusions are to some extent necessary for a proper machinability performance.

In steelmaking calcium is commonly used to modify inclusions composition, as well as their shape and size. In aluminum-killed steels, the inclusion
population will generally include hard and abrasive alumina (and spinel) inclusions and manganese sulphides. Taking advance of the strong affinity of
calcium to oxygen and sulphur, calcium treatment´s objective is to modify those oxydic inclusions mainly to calcium aluminates (CaO-Al2O3) and the
sulphur in the steel is associated with these inclusions to calcium sulphide.

The multivariable system given by the industrial conditions in secondary metallurgy is so broad, that the industrial implementation of effective Ca-
treatment is a difficult task that requires systematic investigations. Thereby the topic of the present study concerns the mechanism and kinetics of the
transformation process of alumina (and spinel) inclusions in steel when calcium is introduced by wire feeding or powder injection. To clarify the
mechanisms, several types of model experiments are being performed on both laboratory test and industrial scales. The phases formed due to the
reaction associated with calcium addition will be examined by SEM-EDS. Results in the plant will be assessed in terms of castability of the liquid steel and
machinability behavior of the final product.

Relations of steel properties to metallurgical treatments
Strength and
Hot formability
fatigue strength
Rolled bars
Rolling & heat treatment
Ductility Cold formability
Rolling mill

Hardenability Weldability Continuous Casting
Metallurgy of
liquid steel Vacuum Degassing
Surface quality Corrosion resistance
Ladle Furnace

Cleanliness Machinability Electric Arc Furnace

Raw material: scrap

Tailor made Ca treatment for improved machinability
Graphite extension

Experimental setup for the Ca addition in steel lab scale Graphite cup
Alumina rod with Ca addition

experiments has been designed to investigate on the effect of Graphite screw

Ca addition procedure, for different oxygen contents. Graphite sample holder

Alumina crucible

Liquid steel
Ca addition

Ca treatment

Holding time
DETERMINATION OF THE INTERFACIAL
BEHAVIOUR IN GRAPHENE NANOCOMPOSITES
BY MULTISCALES MODELING
aGuillermo Fernández Zapico, bJuan Manuel Muñoz Guijosa
ag.fzapico@alumnos.upm.es, bJMguijosa@etsii.upm.es,

Abstract
Graphene is an emergent material that is causing significant impact in numerous research fields. Its excellent mechanical, thermal, optical and
electrical properties make it an attractive material for multiple applications; the composite materials are one of them. From the theoretical point of
view, the addition of graphene over a polymer matrix improves considerably the mechanical properties. However, from the experimental
standpoint, the improvement is not so big. This phenomenon is generally explained by two causes: one, the weak interface between matrix and
nanoparticle, and two, the graphene orientation.

In order to predict the mechanical behavior of any material, this project builds a tool that considers the interface behavior between matrix and
nanoparticle as well as the graphene’s dispersion and orientation un-alignment. To understand the material´s behavior at macroscale, this tool
analyzes its behavior at nanoscale. This tool allows closer approximation to experimental data than other models theoretical models and could
avoid the use of expensive and complicated characterization techniques.

Multiscale Modeling, from Nano to Macro

STEP 1 STEP 2 STEP 3
Nano scale modeling Material homogenization at nano Macro scale. Ensemble of nano scale
scale level. elements

Graphene

Matrix

At the nano scale level a graphene polymer An unique material with the properties
composite unit cell is studied. The of the matrix and graphene is created. Introduction of the graphene orientation,
nanoscale model takes in account the It also takes in account the damage at disaligment and dispersion. The program
interface damage and the adhesion the nano scale level for different updates itself the mechanical characteristics
parameters (G, ) solicitations. of the composite
Below a typical damage map of the
material is shown
Matrix Graphene Graphene
1Graphene´s orientation 2 Damage Maps:
D(?? ,?? , ??? )

??
Matrix ??
???
?
Obtain ?? , ?? and ???
Graphene

Matrix Matrix 3 characteristics
Mechanical

Graphene
actualization
Graphene Graphene
?? = f(D)
Matrix Matrix
?? = f(D)
??? = f(D)

References
[1] M. Sharon y M. Sharon, Graphene. An Introduction to the Fundamentals and Industrial Applications, Scrievner Publishing.
[2] Guo y Y. Zhu, «Cohesive-Shear-Lag Modeling of Interfacial Stress Transfer Between a Monolayer Graphene and a Polymer Substrate,»
Journal of Applied Mechanics MARCH 2015, Vol. 82
[3] L. Gong, I. A. Kinloch, R. J. Young, I. Riaz, R. Jalil y K. S. Novoselov, «Interfacial Stress Transfer in a Graphene Monolayer Nanocomposite,»
Advanced Materials, 2010, Vol.22(24), pp.2694-2697
Project Management Coaching Model: First  PMQ
insights
aBallesteros Sánchez, Luis I., aOrtiz Marcos, Isabel.
a Universidad Politécnica de Madrid (luisignacio.ballesteros@upm.es)

Abstract
The Project Manager is more and more a key agent in the development of business, growth and capabilities of organizations, particularly in a changing,
complex and challenging environment (PMI, 2007; PMI, 2013). He has the responsibility of assisting needs related to project team management, ensuring
that results and quality of the project meet requirements and stakeholders satisfaction. Having a critical role in the organizations, to be an effective Project
Manager requires a grate understanding of different fields that must be coordinated, as well as strong personal skills (Ahsan & Ho, 2013).
Since personal competencies have demonstrated to be better predictor of successful Project Managers (Clarke, 2010; Muller & Turner, 2010; Skulmoski &
Hartman, 2010), the academic research has focused on understanding what are these personal competencies. However, less research has been developed
to better understand how these competencies can be strengthened. Executive coaching has gained popularity during last decade as a tool for developing
better behaviors and attitudes in the workplace, but there is still an important need to enrich the theoretical framework and better understand how it
works, specially in the context of project management.
The aim of this research is to develop and test an specific project management coaching model, which has provided a baseline for the development of soft
competencies of Project Managers. The model is being validated by conducting an experiment with active Project Managers to analyze its effectiveness,
success factors and the impact of coaching on different competencies. First results reveal interesting insights.

Research Questions
• WHAT are the key personal competences of Project Managers?
• HOW is possible to strengthen these competences effectively?
• HOW MUCH and what competences are more strengthened
through a project management coaching model?

Project Management Coaching Model
• A coaching process design with objectives description, inputs,
steps, activities, tools, templates and outputs.
• A definition of key personal competencies: for both the Project
Manager and the Project Coach.
• A set of good practices: based on success models in Project
Management.
• Definitions and principles of PM Coaching

Design of the Experiment
Sample and Data Collection
• 30 Project Managers (15 EXPeriment + 15 CONtrol)/32 OBServers
• 11 Companies from different industries: 120 coaching sessions.

First insights
Main Results
New PM Personal Competences
Framework
Based on relevant literature review and main
challenges and objectives gathered from 120
Project Managers sessions.
Effective Project Management
Coaching Model
High levels of effectiveness demonstrated.
Improvement of behaviors
• Disseminating of information.
• Expressing positive expectations.
• Promoting team learning.
• Ensuring that expectations and
responsibilities are clear to team members.
• Consolidating opportunities.
• Including experts in meetings and influencing
• Accepting responsibility for faults.
Cultural risks in projects developed in South and
East countries.
An approach to identify risks in international cooperation projects.
aRodríguez‐Rivero, Rocío; aOrtiz‐Marcos, Isabel

aDepartment of Organization Engineeering, Business Administration and Statistics. ETSII‐UPM 
(rocio.rodriguez@upm.es)

Abstract
Nowadays most projects develop in international contexts where cultural issues are very important. A deep analysis of the risks linked to culture allows
companies a better understanding of how to achieve the project success.
Hofstede’s Cultural Dimensions and Meyer’s Culture Map have defined how every society possesses features that distinguish it from each other. The
existence of such theories invites to research which are the specific risk categories associated with working with different cultures.
Based on the combination of both studies, this research establishes its own cultural dimensions. Through literature review, cultural risks are identified and
organized according to the defined dimensions in the Cultural Risk Breakdown Structure (CRBS).
So the aim of this research is not only to create a comprehensive risks register of these cultural risks but also a set of project manager competences to
improve client satisfaction and to take decisions in a cross‐cultural environment.
RBS
Cultural Risk Breakdown Structure (CRBS)
Cultural Approach: Hofstede and Meyer Cultural Technical External Organizational
Project 
Management
Seven Cultural Risk Dimensions as a synthesis of both cultural Figure 1. Addition of cultural category to PMBoK RBS. CRBS model
approaches, Hofstede and Meyer, to represent the culture in a complete
frame, useful to apply to international team works in every kind of Geographical Areas: Africa, China and Middle East
business. These dimensions are: Three areas selected since they are fast‐developing regions in which many
new projects are carried out and they are characterized by centuries‐old
Cultural Risk Dimensions
cultural and socio‐political customs. The three areas score high values of
G. Hofstede
Power Distance
Power distance – Long‐term  Power Distance and Uncertainty Avoidance.
Leadership orientation
Individualism
Masculinity 
Ubuntu Guanxi Wasta
Uncertainty Avoidance
• Two traditional cultural • Confucianism roots: • Extremes of wealth

Middle East
China
Africa

Long‐Term Orientation Trusting Individualism
regions (North Africa & benevolence and and poverty
Indulgence Cultural Risk  Sub‐Saharan Africa) and conformity • High variety of
a very complex country • Relationships are languages
Dimensions (South Africa) highly appreciated • Importance to family
E. Meyer • 48 countries + 6 islands (guanxi) networks
Communicating • Main common patterns: • Recognition of social • Importance to the
Evaluating Uncertainty  respect for authority standing: dignity and working relationships
Persuading Persuading avoidance  and elders, collectivism, prestige. Face (mianzi) • Polychronism: multi‐
Leading – Deciding family orientation and • Reciprocity and tasking & spontaneity
Deciding need for consensus empathy (renging)
Communication
Trusting
Disagreeing Figure 3. Analysis of the geographical  areas studied
Scheduling
Methodology Number of 
Identifed Risks
Africa China
Middle
East
Figure 2. Cultural Risk Dimensions  • More than 60 JCR reviewed Power distance –
7 4 4
Leadership
Conclusions to improve risk management in • Projects developed in three areas
Long‐term 
multicultural contexts • Seven cultural risk dimensions orientation
3 1 2

Individualism 2 3 4
Design of a set of competences to project managers working in Results
Uncertainty 
multicultural contexts. • 69 Risks associated to not consider avoidance – 9 2 4
Leadership
‐ Motivate and activate team work
culture Deciding
Power distance– Communication 6 4 4
Leadership ‐ Define a precise organization chart
‐ Know the country institutions and their legal frames
• 15 To Power Distance (22%)
Persuading 2 1 2
‐ Identify the most influentialstakeholders and take them into account
• 15 To Uncertainty Avoidance (22%)
Trusting 1 2 2
Long term orientation– Time management
‐ Have the capacity to manage unexpected events
• 30 in Africa (43%) Figure 4. Number of identified risks
Scheduling
‐ Understand the differences in time conception to scheduling

Team building Next step: application of CRBS model to
Individualism ‐ Train the team using clearly defined roles
ADAPTATION CAPACITY

‐ Promote the idea of team sucess: failure /succes depends on the team
Human Development Projects
Defining
Uncertainty avoidance– ‐ Guarantee the stability of the team´s work Study of Cooperation Projects funded by UPM
CRBS

Deciding ‐ Distribute responsibilities
‐ Adapt to the local conflict management style • 152 Projects from 2005 to 2014 by 29 UPM’s Cooperation Groups
‐ Transmit confidence in the technologies used
• Lessons learned from their final reports categorized in seven groups:
Communication
Communication ‐ Communicate in a precise way
Proper beneficiares  Political Cultural  Activities’  Knowledge Relationship with Second
‐ Meet directly with area managers
‐ Know the local language
definition environment environment implications transfer local partners stage

Negotiation Figure 6. Clusters of lessons learned from final reports of UPM’s Cooperation Projects  
Persuading ‐ Adapt business skills to the local context
‐ Increase team members´ self‐confidence

Trust building
Bibliography
Trusting ‐ Possess networking skills • Hofstede, G., & Hofstede, G.J. (2005). Cultures and Organizations.
‐ Demonstrate commitment
‐ Accept local customs Software of the Mind, Second edition. Mc‐Graw Hill.
Figure 5. Competences required to manage CRBS  • Meyer, E. (2015). The culture map. Public affairs. New York.
Open Data assessment of Spanish cities

aRaffaele Sisto, bJavier García López

a PhD Student Technical University of Madrid (raffaele.sisto@alumnos.upm.es)
b PhD Student Technical University of Madrid (javier.garcialope@alumnos.upm.es)

Abstract
Local, regional and national governments are acquiring a commitment to a horizontal strategy in Open Government. This idea of governance aims to
promote public administration transparency, open data and citizens’ interaction in decision-making processes. Most of Administrations improve citizen
information through portals of Transparency, Participation and Open Data which offer free access to public sector data to serve citizens or businesses.
Their objectives are: to obtain an updated vision about the status of these initiatives with a special focus on Spanish and Italian cities that are countries
with the highest maturity levels; to identify best practices within their efforts and strategies; to demonstrate gaps as lack of standardization; to assess the
quality of open data regarding incorrect formats, useless or outdated data. It assesses the best open data city portals for understanding which one is more
valuable and which are the technologies more adequate to citizen use.

Open Data assessment of Spanish cities
Methodology
The three areas of Open Government were assessed. The study carried out Data quality. How the data may be used and how reliable the data is. Five
analysis and research of the most important Open Data websites through features were evaluated: Data available is open source, free to everyone and
navigation and user experience as conventional features. It also analyzes only updated; Data sources are official ones; Format based on the 5-star open
the official websites of public administration and local authorities. Further, data;
the study is focused on Spanish provincial capital and cities over 200,000 Data accessibility. This evaluation is based on six features: Aggrupation
inhabitants who have promoted a specific open data portal. downloaded (RDF); SPARQL/API Service; Languages; Accessibility; Online and
The assessed cities were evaluated according to the next criteria: downloadable data.
Participation analysis using two dimensions, (i) the existence of a specific Data typology. A holistic approach to data is an essential element in smart
participation councillorship and (ii) the publication of online activity. cities making available data from all sectors of the city.
Transparency analysis (based on two main axes), (i) the existence of a specific Data visualization. Not only data scientist should allow accessing open data.
transparency councillorship and (ii) transparency indicators (according to Thereby, the study analyzes the key visualization tools that allow better data
Spanish and Italian law). accessibility: dashboard, data pre-visualization before downloading, readable
Open data portals analysis using four dimensions: format, relevance.

Conclusions for Spanish Cities Final suggestions
The following summary table shows the final result of the evaluation. This study observes a lack of a deep analysis on the Open Data benefits for an
Open Government strategy. The two major uses of Open Data portals are not
properly represented in the majority of cities evaluated:
Citizens empowering and transparency. The main issue is not having a
standard on visualization and organization. No example includes an easy
module to be used by citizens to ask for the relevant information for them.
Data business models. No analysis of the data value is appreciated in any
example before deciding data published.
Each administration level (national, regional or local) has begun to develop
Open Government strategies with different scopes and methods. This is
The main impression is that Spanish cities have decided to open their data in getting worse due to their distribution of responsibilities and some objections
a very heterogeneous way. They have many gaps in both form and content in the dissemination of certain information which could be understood as
because they have not followed a common strategy. It has been found cities compromising. All this has led to a heterogeneous state of the art in content
that only have transparency website and have not opened their data (for and form. Below, It has been summarized some opportunities for
example Murcia) or many other cities have a specific website but its content improvement and recommendations to achieve the principle described in the
is limited (for example Tarrasa only have jpg or pdf files). On the other hand, previous code of good governance:
cities like Santander, Madrid and Valencia, are as advanced in this field as • Municipal Strategic Plan for Open Government
London (UK) or San Francisco (USA). A vast majority of municipal websites • Data qualification
analysed have direct links to any specific fields related to Open Government • Encourage Citizen Participation
but only 36,5% of them (23/63) have a municipal Open Data website. • Open Data indicators and Dashboard
33 different format types have been found. Only 10 of them represent 83% of
all published datasets. The three formats most published by the 11 selected Each official Open Government website of the assessed Spanish cities visited
cities are CSV (24.2%), XML (11.7%) and JSON (11.4%). Mainly the used during May and June of 2016. Please contact the authors for references and
formats are focused to promote their interoperability. results regarding Italian cities.
Mapping City Metrics Methodologies

aJavier García López, bRaffaele Sisto

a PhD Student Technical University of Madrid (javier.garcialope@alumnos.upm.es)
b PhD Student Technical University of Madrid (raffaele.sisto@alumnos.upm.es)

Abstract
Smart cities as new socio-technological model of the cities, indicators methodologies and city metrics rise as an interdisciplinary tool and they have
become an essential instrument to evaluate, measure and represent how the use of ICT influence the city management and the improvement of the
citizen’s quality of life.
The outcomes of this research have the goal to: (i) contribute to the development of a standard adaptable methodology for every context and city
morphology, to improve their strategies by optimizing their resources; (ii) detect limitations, common gaps and future opportunities of improvement; (iii)
identify the indicators’ legitimacy and reliability; and (iv) state the need of a standardization of the measurement and assessment.

Mapping City Metrics Methodologies
Methodology
It has been developed a qualitative and quantitative analysis of over 80 different methodologies based on indicators systems, indexes, comparison standards
and rankings with and international reach within the smart cities context.
Previously to the evaluation process, an intensive search of these methodologies has been done that have been published over the last five years and that
matched these characteristics. Once this was done, a first discard of 29 of them because they did not achieve a holistic vision, since they did not address
several of the main recognized domains of a smart city. Likewise, the methodologies that are only accessible with a previous payment have been discarded
and/or the ones that do not have the list of indicators published. From this filter a quantitative and qualitative review has been done to, eventually, dismiss
those that were not applied to cities, ending up with a total of 33 finalist methodologies. On this final selection, a detailed analysis has been developed
identifying comparison criteria such as the number of indicators, domains, quantity and type of the cities studied, type of sponsors, objectives definition,
methodology of indicators construction, contextualization, availability and quality of the database, improvement proposals and development over time.
Results and conclusions
It can be confirmed that the amount of indicators in each methodology is very heterogeneous. Its
number vary from the 20 of the “Arcadis Sustainable Cities Index 2016” to the 450 of the “Rapporto
Smart City Index 2016” of the consulting group Ernst&Young. In most cases, the indicator selection of
each methodology is very related with the measurement goal, but the way of adding indicators to
reduce them to a synthetic value is very heterogeneous, or in some cases, non-existing.
There is no agreement on the domains where the indicators are distributed. The recurrent areas with
the most indicators associated are environment (45), economy (39), transport and mobility (37), and
government (33), since these are the four main areas and the more consolidated as references to
evaluate the smart city performance and where there is more availability of data.
All the methodologies above analyze the “smartness” of a city but differ in their objectives, metrics,
scopes and therefore its results. Normally, they are conditioned by the availability of information and
the dataset homogeneity and reliability. Following, four inherent determinants to be considered are
summarized:
• Scale aspect of the indicator: when no data is available on a local scale, the data is extrapolated
from the national scale. This situation means an abstraction that distorts the reality of the assessed
cities.
• Competence aspect of the indicator: when what the indicator measures is not a direct
responsibility of the local administration.
• Semantic aspect of the indicator: when there is no direct correlation between the definition and its
objective.
• Time aspect of the indicator: when the data update that nourishes the indicator may distort
the ability to compare it with other cities
Finally, a set of needs is proposed in the form of improvement opportunities:
• The need to contextualize each measurement;
• The need to know what, how and for what you are measuring;
• The need to not limit to administrative units, by adding the geographical
and metropolitan scale to the measurement due to its interrelation.
Please contact the authors for references and more results.

Supporting Institution Spatial scope
Entrepreneurship policy agenda in the EU
explained: a contribution to understand the
European entrepreneurial path and its way
forward
Alberto Arenal, Cristina Armuña
(UPM (a.arenal@alumnos.upm.es, cristinaag@alumnos.upm.es)

Abstract
Since the last quarter of the twentieth century entrepreneurship has emerged as a pivotal theme in the policy agenda all over the world (Wennekers et al.,
2002) but it is not up to the second half of 1990s when EU policymakers start to develop an entrepreneurship policy as such (J. Z. Acs et al., 2010).
Despite nowadays there is no doubt of the relevance of entrepreneurship as a key driver for EU competitiveness and economic growth, and the interest of
promoting entrepreneurial ecosystems across Europe, there is a lack of understanding of the evolution of the key areas, topics and concepts that have
shaped EU entrepreneurship policy agenda.
Through text mining techniques this research analyses the most relevant EC Communications and a comprehensive selection of articles from relevant
journals in the field since 1990s. Results offer a granular analysis of the emergence and development of entrepreneurship policy in the EU, establishing
patterns and temporal relations between Academia and Policy approaches.

Entrepreneurship Ecosystems in the EU: relevant milestones of the policy domain
Domains ofthe Entrepreneurial Ecosystem Evolution of the Policy domain in the EU: high level and specific Entrepreneurship policy
framework
Santer Prodi Barroso Juncker
MAASTRICHT LISBON AGENDA EUROPE 2020 (2010)
TREATY (1992) (2000)
A new Start Lisbon Digital Single Market
Ecosystem Ag. (2005) (2015)
Domain: POLICY
1995 2000 2005 2010 2015
Luxembourg European Charter for Entrepr. Europe´s
Geography: UE Small Business SBA
Next
Employment Small Enterprises (2000) Act (2008) reviewed 2020 Action
Summit (1997) Plan leaders: the
Fostering Action Plan: (2011)
(2013) start-up
Entrepreneurship in The European Modern Social and scale-
Europe: Priorities agenda for SME Policy Business up initiative
for the Future Entrepreneurs (2005) Initiative (2016)
(1998) hip (2004) (2011)
6 Domains of the Common Entrepreneurial Ecosystem
Framework. Based on the entrepreneurial model of
Babson College (fine border) and EIP-OECD (thick Objective (RQ):
border). Own elaboration.
Deeper understanding of the evolution of key the areas, topics and concepts which have
shaped the European Union entrepreneurship policy agenda by using text mining techniques
Methodology
UNSTRUCTURED DATA TEXT MINING PROCESS RESULTS
Corpus 1 • CLUSTERING
708 EU Policy Documents Factor Finding EC
Pre- Cluster
Tf-If analysis Communications / papers
Corpus 2 processing analysis
(SVD) related to each other
1110 Academic Papers from Text mining techniques • CLASSIFICATION
Q1&Q2 Journals in the field
of Entrepreneurship Text mining and text analytics are a subset of data mining techniques focused on turning Basis for arranging similar
unstructured text data into high quality information or knowledge. EC documents/ papers in
They cover a range of interrelated objectives: search and information retrieval, document a specific category
Source: Official Journal of the
European Union and Web Of clustering, document classification, information extraction, natural language processing • CONCEPT EXTRACTION
Science (T.R.) and concept extraction The entrepreneurship
policy agenda
Preliminary conclusions and further work
Preliminary conclusions Further work (in progress)
 Lisbon Agenda (2000) could be considered the pivotal moment when EU policy  Increasing the granularity in the analysis of the entrepreneurship
makers start to develop an Entrepreneurship Policy policy agenda (relevant themes put in a temporal framework)
 There are three general stages in the evolution of the EU Entrepreneurship Policy:  Establishing an empirical based classification of terms (concepts)
A. Firstly, primarily focus on the diagnosis but hardly any relevant concrete and revealing relationships between terms and key factors that
initiatives are not possible at individual word level
B. Second, under the principle “Think Small First”, concentrate in the regulatory  Going deeper in the evolution and, in case, the relation between
framework for SME the Academia and the EU Policy makers. An additional level of
C. More recently, narrowing the scope and tunning the focus on concepts as analysis is possible considering a third corpus, for example, Grey
entrepreneurial ecosystems, social entrepreneurship and high growth literature (Think Tanks, consultancy firms…).
entrepreneurship
Exploring the factors that contribute to
transform entrepreneurial intention into action.
Reflections on potential entrepreneurs' needs
and entrepreneurial education programs
aCristina Armuña, bAlberto Arenal
aUPM (cristinaag@alumnos.upm.es) bUPM

Abstract
Potential entrepreneurs are defined as those individuals with the intention of launching a business venture. In this phase, known as “stand-up”,
opportunities, knowledge and skills play a determinant role to the conception of a business, which can lead or not to the “start-up” phase when the firm is
effectively launched (GEM,2016). Considering entrepreneurs as value creators with positive impact in economic growth, there is a high interest in
supporting those potential entrepreneurs (at their stand-up phase) to become real entrepreneurs (start-up phase). In this context entrepreneurship
education is considered an influential tool to shape entrepreneurial attitudes (Liñán, 2011). Following the TPB model (Ajzen, 1991) this research explores
the main factors affecting entrepreneurial intention on a sample of potential entrepreneurs (91 students of three introductory entrepreneurship education
programs - EEP), which could contribute to orient entrepreneurship education practices to impact in their intention and future behaviour.

Early Start-up stages: from intention to action
Stand-up Start-up
N
DE
L M ERCAD

O
EVOLUCIÓ
Nascent Entrepreneurs
The intention is considered the best single predictor
of behaviour (Krueger & Brazeal, 1994). Recent Actively involved in setting up a
meta-analysis confirm the relation between business they will own or co-own
intentions and subsequent behaviours in the (Kelley et al., 2016)
Potential Entrepreneurs entrepreneurship field (Schlaegel & Koenig, 2014). Behaviour
The stage "when individuals or groups
are equipped with the entrepreneurial Too few people have the entrepreneurial mindsets &
attitudes and skills to attempt the Potential role of skills needed to set up their own business (EC,2016) Contents
creation of new company" (European entrepreneurial
Commission, 2016) Delivery process
education One of the most influences tools to improve
Intention entrepreneurial attitudes (Liñán,2011) (Cooper et al., 2004)

Exploring the TPB to understand potential entrepreneurs’ intention
The Model: Among the different intention models used to TPB: prediction of Intention (motivations to do something) and Behaviour (the fact
explain entrepreneurial actions, the Theory of Planned Behaviour of doing something) from the analysis of three personal variables: the Attitude
(TPB), first proposed by Azjen (1991) has been widely used in towards the behaviour (one's perception about the behaviour as positive or
studies related to entrepreneurship for the last 20 years and is negative), the Subjective norm (one's perception about what people around are
considered highly consistent (Lortie & Castogiovanni, 2015; Micaela going to think about the behaviour) and Perceived behavioural control (PBC)
et al., 2014). (one's perception about his/her own capabilities to the actual behaviour).
α KMO
The Sample 53.8% Model&Results Regression Factor Analysis
R2=0.246 Intention (6) 0.889 0.863
Students that have coursed an 46.2%
EEP at Tongji University Attitude (5) 0.821 0.818
REGIONAL 0.15
(Shanghai). ATTITUDE PBC (6) 0.746 0.770
ENTR. CULTURE 0.62
Female Male 0.28
0.25 SN (4) 0.769 0.742
Of the total sample of 130 students, the analysis 0.20
considered the 91 individuals who had not already R2=0.511 Scales Reliability
NEAR ENTREP. PBC
experience on entrepreneurial ventures.
INTENTION
51.60%
39.60%
Standard Coefficients Beta SUBJ.
8.80% NORM •In EEP, the combination of personal
developments and applied knowledge for
Conclusions and further research business development are considered the
Europe Asia Others
•Aligned with research in the field (Lortie & two main axis to build entrepreneurial
The Questionnaire Castogiovanni, 2015), results suggest Attitude is oriented programs (Vesper&Gartner, 2001).
The questionnaire includes 23 observed variables the best single predictor of Intention in the TPB •Further research focused on EEP programs
(items have been built as 7-point likert-type scales) model with a role for the individual’s perceived and techniques is needed (Liñán&Falyolle,
grouped as follows: capabilities to become entrepreneur. 2015) and the incorporation of elements
•Focused on the attitude, the model explores the focused on the factors that impact on these
• Intention (explained variable): 6 variables (Liñán,
role that PBC and Subjective norm play, and potential entrepreneurs could contribute to
2011; Liñán&Chen, 2009)
additionally includes the perception about the promote the stand-up stage to the start-up
• Attitude: 5 variables (Liñán,2011;Liñán&Chen, positive regional culture towards one. According to the study, the participation
2009) entrepreneruship and the influence of knowing of entrepreneurs in the programs could have
• PBC: 6 variables (Liñán, 2011;Liñán&Chen, 2009) entrepreneurs. The PBC and the near positive effects on students’
• Subjective Norm: 4 variables (Autio, 2011) entrepreneurs are the most weighted. entrepreneurship intention.
• Regional Entrepreneurial culture’ perception: •Entrepreneurship Intentions using the TPB have •A larger students sample is needed to
single-statement been mostly addressed on general students validate results and explore EEP
samples, with an opportunity to go deeper in characteristics (specific contents and
• Near entrepreneurs (friends and acquaintances): results on these specific EEP students methods, for example) that could positively
single-statement (Lortie&Castogiovanni, 2015). impact on potential entrepreneurs intention.
Big Data for Sustainability in the Digital Era:
Methodological Framework towards a Conscious
and Smart Governance
a Gema del Río Castro, b Camino González, b Ángel Uruburu
a Universidad Politécnica de Madrid (gd.rio@alumnos.upm.es)
b Universidad Politécnica de Madrid

Abstract
The advent of Big Data and ICT paradigms is heralded to trigger a revolution of science, enabling a data-driven economy and digitally savvy society.
Combining Big Data with traditional methods brings unimaginable capacities for forecasting, understanding complex issues, raising awareness,
enabling new value ecosystems, hence holds promises for deriving value, promoting better decision-making and accountability. As game changer,
Big Data offers a far-reaching potential on the sustainability arena, for addressing intractable societal, environmental and governance challenges
in an unpredecented fashion, particularly towards pursuing the Sustainable Development Goals (SDGs). Notwithstanding, because of their complex
multi-dimensional nature and nascent state-of-play, the nexus between both domains remains unexplored, and little is known about how to seize its
opportunities while minimizing pitfalls, or how to refine data into insights to create shared value. This research aims at contributing to bridge current
gaps and maximizing the potential of the Data Deluge for Sustainability.
Stakeholders’
Technology adoption THE BIG DATA REVOLUTION
Competitivity No agreed definition. Usually known
Good
Big Data Governance
by its 5V’s: high volume, velocity,
Ecosystem Social issues variety, veracity and value.
Alternatively, the 3C’s, new ecosystem
(5V’s, 3C’s) Environment of data crumbs, capacities &
Digital communities [1] (Letouzé, 2015).
paradigms BD4S SUSTAINABLE DEVELOPEMENT GOALS
17 ambitious goals & 169 targets set
#BD4SDG the Global Agenda 2016-2030 [2]
ICT4S Sustainability
Digitalization #Data4Good building momentum towards a
5P’s: people, planet, sustainable development focused on 5
#Data4Dev
prosperity, peace, pillars: People, Planet, Prosperity, Peace
partnerships & Partnerships.

Figure 1. Scope of the research theme. Sources: own contribution. SDGs icon from [3] WBCSD (2016).

Aim & Scope Research Questions Methodology
 What features describe the intersection amid Mixed inductive-deductive research,
The research aims to explore the link
Big Data & Sustainability ? framed within 2 stages.
between Big Data & Sustainability (BD4S)
1st stage is focused on profoundly
to fully understand its hallmarks and shed  What are the untapped opportunities and
understanding the BD4S context to
light on how can be better leveraged for challenges for harnessing the BD4S nexus?
formulate a conceptual model, from
pursuing a conscious corporate governance.  How and to what extend could the Big Data
literature review & analysis of ongoing
The ultimate goal is to devise a Ecosystem contribute effectively and responsibly
pilot projects/ cases.
methodological framework which helps to to the SDGs (BD4SDG)?
2nd stage is dedicated to conceive &
drive a sustainable digital transformation  What tools & approaches could be formulated to validate, via case studies, a construct
through the SDGs. enable the adoption of BD4S towards a conscious (methodology) to facilitate the adoption
corporate governance ? of BD4S on the corporate arena.

Discussion: BD4S, hype or promise? Big Data for Sustainability (BD4S) is becoming an emerging global trend, holding both
promises and challenges. BD4S is already being explored for tackling diverse social, environmental or governance issues related to some SDGs,
from biodiversity conservation, smart mobility to healthcare, gender equality or education, mainly through a governmental perspective,
excessively focused on quantitative monitoring. However, Big Data capabilities (namely descriptive, predictive, prescriptive, diagnostic),
remain unproven on the SDGs realm. Perils on security, privacy, digital inequality, data blindness, apophenia or lack of causation still prevail.
Thus, action is needed to overcome current challenges and harness the power of BD4S from a corporate perspective, by formulating innovative
methods and responsible strategic directions for embedding sustainability into the core of business and global strategy.

References:
[1] Letouzé E. (2015). Big Data & Development: General Overview. Data-Pop Alliance White Paper Series.
[2] UN (2015). United Nations Sustainable Development Goals. Available from www.un.org/sustainabledevelopment/sustainable-development-goals/
[3] WBCSD (2016). SDG Business Hub. Available from http://sdghub.com/
Unwarranted Variations in Healthcare and the 
Role of Business Intelligence
a,b Sina Lessanibahri, aLuca Gastaldi , bCamino González Fernández
aPolitecnicodi Milano (Sina.lessanibahri@polimi.it)
bUniversidad Politécnica de Madrid

Abstract
Nowadays healthcare organizations are constantly forced to perform better with less resources. Consequently, healthcare managers are continuously
searching for ways to utilize the resources as efficient as possible, without negative impact on the quality of care delivery. On the other hand, substantial
cross‐regional variations are shown by extensive prior investigations in different countries. Existence of such variations indicates that the best care has not
been received or that resources have not been appropriately deployed in health care organizations. Business Intelligence (BI) has been recognized as a
potential remedy for this shortcomings. In this thesis we try to leverage over advanced analytics specifically associations rules mining, to identify the
associations and trends in the unwarranted variations across the health care systems’ performances. Moreover, as a methodological contribution we
propose an improved algorithm for association rules mining that will facilitate the process of finding interesting associations among healthcare related
indicators and variables.
Key words: Health care; Unwarranted Variations; Business Intelligence; Association Rules Mining; Frequent itemsets mining

among them will be much easier for medical experts, which is crucial for
Research Background healthcare datasets that usually have an overwhelming number of variables.
BI and Healthcare Following Figure depicts the computational effort improvement of the
algorithm on a test dataset, compared to Apriori, one of the most commonly
In healthcare, researchers and decision makers try to leverage over the
aggregated data from data warehouses and centralized repositories to used ARM algorithms.
2500
gather knowledge and enhance both healthcare quality and patient
CPU Time (S)

2000
outcomes. Decision makers can analyze these data to discover for
instance the substantial changes in hospitalization rates—especially for 1500

expensive medical treatments. The implementation of BI in healthcare 1000 Apriori
proceeds relatively slowly and in an ad‐hoc way. This suggests that 500 Apriori + Width‐Sort
Successful implementation of BI in healthcare relies on understanding 0
and analyzing the peculiarities and complexities of the domain. 0.25 0.2 0.18 0.1 0.05 0.02 0.01 0.005

Minimum Support treshold
Unwarranted Variations
Detecting the Unwarranted Variations
It is well known that healthcare systems suffer from medical practice
variations. Existence of such variations indicates that the best care has In order to measure the unwarranted variations we used National Health
not been received or that resources have not been appropriately System's Hospitalization Minimum Basic Data Set (MBDS) that contains records
deployed in health care. Previous research on care variations initiated of all patients in Spanish public hospitals. The variations for 4 performance
by Wennberg and Gittelshon (1973) revealed the existence of indicators and 6 Major Diagnostic Categories were calculated, adjusting for the
geographical variations in various surgical operations. Unwarranted risk factors such as age and severity of illness. The provinces were further
Variation are defined as variations that cannot be explained by clustered into 4 groups based on their performances. Following figure depicts
patients’ illness or patients’ preferences. These are the variations that the performances of the 4 clusters. The value of an indicator close to the center
are explained by differences in health system performance. We argue of the chart, indicates better performance of the indicator comparing to the
that there is a need for studies that consider several high impact and country average.
Mortality
resource intensive conditions as well as performance indicators
simultaneously, to provide a holistic overview for health professionals
and policy makers.
Cluster 1
Research Questions Length of Stay Readdmission
Cluster 2
RQ1. How to develop/improve specific BI analytic techniques that are Cluster 3
more appropriate for the inherent complexities of the healthcare? Cluster 4
RQ2. How can BI detect and measure the unwarranted variations
considering the idiosyncrasies of the healthcare organizations?
Infection
Findings Following figure shows the provinces that belong to each of the for clusters. By
comparing the performances’ of the healthcare systems in different regions and
The initial findings for each of the  research questions are reported in  clusters, it is possible to propose implications for policy makers to reduce the
the following. unwarranted variations.
BI Analytics in healthcare
This phase of the study involves a methodological contribution to
Association Rules Mining (ARM). ARM is one of the most popular
exploratory data mining techniques to discover interesting and
previously unknown correlations from data. Even though originally
developed for market basket analysis, recently there has been an
increasing attention toward using ARM for medical datasets. We
propose a novel approach called width‐sort to mine the Maximum
Length Frequent itemsets as well as itemsets with the pre‐specified
length. Mining the maximum length itemsets is beneficial to decrease
the computational effort. Moreover, by mining the maximal length
frequent itemsets, the tedious task of finding the interesting rules
Increasing the trading prediction by mining
aggregated human texting messages
a Yang Liu , b Joaquín Ordieres Meré

a yang.liu00@alumnos.upm.es
b j.ordieres@upm.es

Overview
A novel approach is proposed to predict intraday directional-movements of the stock in the trading market based on the text of breaking financial
news or social media by event data. This work is an effort to put more emphasis on the text-mining methods and tackle some specific aspects thereof that
are weak in previous works. The research is a specific market, namely, the trading and stock market, which seems not to have been researched in the
previous works based on predictive text-mining. In order to accomplish the above, this work produces a multi-layer algorithm that tackles each of the
mentioned aspects of the text-mining problem at a designated step. This research is that will help financial managers and traders use these to predict the
financial activities and provide them the business decisions to avoid the financial risks.

The figure 1 : The workflow of Fasttext The figure 2 : The workflow of deep neural network

Sentiment analysis Model creation
The first layer is termed Sentiment Integration Layer by deep neural
This step is observed the time series of the company‘s event based on network, which integrates sentiment analysis capability into the algorithm by
the financial news or the social media, which analyses problem of co- proposing a sentiment weight that reflects their sentiment. Additionally, this
reference in text mining that is sentiment classification. Fasttext is inspired layer reduces the dimensions by eliminating value in terms of sentiment and
method that focuses on the meaning of words, which solves to obscure by thereby improves prediction accuracy. The second layer encompasses a deep
ambiguity and play on words. This way can classify polarity, subjectivity and neural network model creation algorithm. It updates the models with the
intensity in the corpus. According to the results of the classification, then it most recent information available, the deep learning algorithm is extensively
changes to the word representation. Thus, prediction accuracy increases evaluated using real market data and news content across multiple years,
significantly at this layer that is attributed to appropriate noise-reduction which is a good fitting capability to evaluate and select the financial news.
from the feature-space. The third layer is that it uses the recursive neural network explore patterns
and of financial news through the recursive and pooling operation.
Measuring general elections in Spain via Twitter, based on
geolocated user comments
aFranco Riquelme, José N. bOrdieres Meré, Joaquín B.
a,bDept. Ingeniería de Organización, Administración de Empresas y Estadística, Escuela Técnica
Superior de Ingenieros Industriales - UPM. Email: jn.franco@alumnos.upm.es ;
j.ordieres@upm.es

Abstract
Social networks provide new insights into current political processes and also generate a “data deluge” that reveals underlying trends in the geography and
demography of the regions that they where are analysed. Research on social platforms has increased ostensibly -especially those based on Twitter- and
inspired by the fact that more and more users express their likes and dislikes. In order to retrieve this information, several techniques are applied for this
purpose, such as statistics, text mining, machine learning, among other disciplines. Nevertheless, information needs to be enriched with geolocation,
which gives us a spatial view of where events take place, thus providing an automated analytical view of the points where events of interest occur.

In this research, we explore the impact of a specific political event in social media: Twitter. In this case, the general elections in Spain held on December 20,
2015, and June 26, 2016. It is intended to establish relative support concerning each party in different provinces of The Spanish geography in the previous
period, during and after the realisation of the elections. Through the correlation of the results between both dates, it is sought to explain the evolution of
the same, comparing the official election results and those obtained from the tweets, to determine the trend that affects the main political parties in the
election year elected.

The aim of this study is to focus on the growing influence of Twitter in environments such as the investigation of the intention of votes in the elections of a
region or country. The purpose is to deduce the influence of this social network when choosing a particular candidate or political party and through these
results, establish new research methods based on application models, replicable even in other fields of social sciences and to develop new applications
such as: models of organizational behavior for decision making; predictive models for technological products estimating their relative success among users;
models that estimate the relationship of a product between a marketing campaign and its effect on potential customers, all based on the knowledge
generated from social networks.

Study case: The 2015/2016 Spanish
General Elections
Marketing applications, predictive models,
new methods generated from Social
Networks

Figure 2. Forthcoming applications

Figure 1. Research description
Revisiting the past and addressing the future
of public infrastructure project
a,bHetemi, Ermal; aOrdieres M., Joaquin; cDominguez, María L.
aUniversidad Politécnica de Madrid (e.hetemi@alumnos.upm.es; hetemi@kth.se)
bKTH Royal Institute of Technology
cADIF Administrator of Railway Infrastructures

Abstract
Large infrastructure projects, public or private, are highly risky, involve large investments, are characterized by long-term planning, with inadequate budget
and time contingencies. The project scope or stakeholders’ expectations change significantly over time due to uncertain costs, schedules, and benefits
making it a ‘norm' throughout the project see (Flyvbjerg 2014; Cantarelli et al. 2010). Studies report different cost overrun levels for the very same project
typology, this as a consequence of discrepancy between cost overrun definitions and reference points, hence producing a significant variability in the cost
overrun percentages that have been reported in the normative literature, (see (Love et.al. 2013) for further consideration).The extant literature on
infrastructure projects identifies four main explanations: technical, economical, psychological and political explanations, see (Cantarelli et al. 2010) . The
majority of researchers built their studies along the assumption that forecasting errors are one-sided and even malevolent (Flyvbjerg 2016), i.e. the existence
of deliberate underestimation of costs and an overestimation of project benefits respectively.

Exploring the relationship between project
scope and lock-in within large
infrastructure project
The main purpose of this study is to understand infrastructure project
performance through technical and economical lenses, main importance is
dedicated to the former. The study elaborates project management reality with
foci in contract management, whilst investigating scope changes deriving from
secondary data taken from high-speed rail projects; i.e. adopting relationship
analysis on project scope at the initial level and throughout the decision points,
as well as investigating possible influences related to time, frequency and
typology of change producing scope variation.
Cost changing factors Figure 2. EU Framework – Effect of cost-changing events on Key Cost elements;
Source: European Commission (1998). Understanding and Monitoring the Cost-
The existing literature gives a partial explication when generally assuming that Determining Factors of Infrastructure Projects: A User’s Guide
theories on optimism bias or strategic misrepresentation (Flyvbjerg 2013; Notes: Large dot denotes a major effect – potentially 20% change for affected
Flyvbjerg et al. 2009; Flyvbjerg et al. 2005) address the issue. This when cost elements Small dot denotes a minor effect – typically 5% change or less for
having in mind that even after special offices similar to National Evaluation each cost element affected.
Office to curb “opportunistic behaviour” were set in place in western world
since 2006, still performance of infrastructure projects is in the shadow, i.e. Methodology
performance of infrastructure project is not enhanced. § Synthesizing existing literature
§ Research agreement signed with ADIF
Administrator of Railway Infrastructures, state-owned company
that answers to the Ministerio de Fomento. Adif plays a leading role
Design
Changes in promoting the railway sector.
Exchange Unexpected
Ground
ADIF is responsible of:
Rate
Conditions
§ Administrating rail infrastructures (tracks, stations etc.),
§ Managing with railway construction, and
Relative
Poor Project § Distributing capacity to rail operators traffic etc.
Price
Cost Management
Changes
Changing
§ Dataset shared on all existing and ongoing High Speed Rail (HSR)
Factors projects; compiling of 36 HSR lines.

Land
Inappropriate
Acquisition
Contractors
Costs

Funding
Force Majeure
Problems

Figure 1. Cost-Determining Factors of Infrastructure Projects
Source: European Commission

Data processing analysis
§ Entity Relationship Model (EER) developed based on logical data process,
see (Chen 1976);
§ Data integrated and handled via phpMyAdmin – a tool written in
PHP Language,
§ EER model – representing the current stage on data processing;

Figure 3. EER model – Integrated data based on important semantic information
Mapping descriptions:
1:n ([line-subline]; [subline-contract]; [contract-incidents]; [contract-
certifications]) 1:1 ([contract-contract type]; [contract-contract winner])
International manufacturing network:
operational issues and reshoring
aAhmed Sayem, bMiguel Ortega Mier, cAndreas Feldmann
a PhD Researcher: UPM, Spain and KTH, Sweden (ahmed.sayem@upm.es)
b Professor, UPM, Spain.
C Asso. Professor, KTH, Sweden.

Abstract
Globalization is not a new phenomenon. The globalization of production activities -renowned as manufacturing
offshoring- has started in the first quarter of twentieth-century. Since then, manufacturing firms are continually
increasing attention to locate their facilities in order to achieve competitive advantage. Though this strategy is still
ongoing, companies face challenges with managing and operating activities in their manufacturing network.
Consequently, there has been a growing interest in the phenomenon of reshoring or backshoring. There are research,
focused on the configuration of international manufacturing and the operation of individual plants within network or
supply chain. However, there has been little discussion on the operation of network entails facilities of an individual
company. This research seeks to address the operational issues of manufacturing network that have implications on
network’s performance, thus to examine the phenomenon of manufacturing reshoring.
Background
------------------------------------------------------------------------------------------------------------
Symptom: Manufacturing relocation /Reshoring /Back-shoring
Specific problems:
i) Cost differentials are not fixed over time,
ii) Lead-time does not remains the same as estimated,
iii) Flexibility in supply chain is difficult to achieve or maintain,
iv) Quality problem.

In Brief
----------------------------------------------------------------------------------------------------------
 “Manufacturing strategy is a plan for moving a Company from
where it is, to where it wants to be” (Miltenburg, 2009).
 International manufacturing configuration options are dependent
upon the perspectives –say, internal and external.
 Reshoring or relocation is fundamentally a location decision.

Main message
---------------------------------------------------------------------------------------------------------
In response to the changing world scenario, companies are no more
benefitted from the existing model.
Now the relevant issues are:
 how and why to choose specific configuration.
 Coordination of manufacturing networks
 Quality, and the flexibility delivery.

Research outcome (1) Research outcome (2)
--------------------------------------------------------------- ------------------------------------------------------------------------------
International manufacturing strategy: International manufacturing network performance:
 Which contingent factor(s) is  Practices related to coordination, and
accountable for the relocation of  Flexibility, and delivery quality are highlighted as
particular configuration type. the key indicators of IMN performance.

The European Doctorate in Industrial Management (EDIM) is a four-year state-of-the-art doctoral research program for doctoral training addressing managerial
issues of significance for the future competitiveness and sustainability of the industrial companies of Europe.
The mission of the program is to generate cutting-edge knowledge at the intersection of engineering and management.
Application of Optimization and Simulation in
the Aeronautic Industry
AR. Pulido, bT. Borreguero-Sanchiran, cA. García-Sánchez, cM. Ortega-Mier
aCorresponding author affiliation (raul.pulido@upm.es)
bAirbus
cUPM

Abstract
We present a simulation model we have developed for evaluating the robustness of a baseline scheduling for an
airframe assembly line. To begin with, we have identified and quantified the main causes of the disruptions.
Then we have created a discrete event simulation model of the production line to take everything into
consideration, and then run several experiments to evaluate different production planning obtained with the
different methodologies and the impact of failures in the deliveries of finished products. Also, different scenarios
in terms of failure quantity and typology have been studied.

Step 1
DAY 1 DAY 2 DAY 3 DAY 4

• We create a MILP model to solve the flexible Job Shop scheduling iSlot
SHIFT 1 SHIFT 2 NIGHT SHIFT 1 SHIFT 2 NIGHT SHIFT 1 SHIFT 2 NIGHT SHIFT 1 SHIFT 2
12.0 16.0 20.0
8.00 0 0 0
12.0 16.0 20.0
0.00 4.00 8.00 0 0 0
16.0 20.0
0.00 4.00 8.00 12.00 0 0
12.0 16.0 20.0
0.00 4.00 8.00 0 0 0

problem with operators. Each process can be performed on Preliminary Test
Cutting
4
8
1
1 1

alternative machines (shared by multiple products) and requires not
Workbench assembl 16 1 1 1 1
Workbench assembl 8 1 1
FC-A-RH Equipping 60 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

only a machine but also some operators to carry out the process.
Furnishing 48 1 1 1 1 1 1 1 1 1 1 1 1
Soft Cutting 12 1 1 1
Final test 8 1 1

We minimize the number of operators and the number of airframes Final Operation 12 1 1 1

in the plant. We obtain a schedule of the production. Fig. 1. Schedule of production from the MILP model.

Step 2
• We build a discrete event simulation model of the airframe
plant. We built all the internal process, and simulate the
machines required, the workers, and the entities in the plant.
• We simulate the production plan obtained by the MILP model.
Fig. 2. Logic representation of the model.

Step 3
• The simulation model take into consideration the effect of:
-Machine failures.
-Errors that require extra work.
-Absence of operators.
-Final test rejection (High and Low).
• We create different scenarios, with extra shift, workers and severity
of failures.
Fig. 3. 3D representation of the model.

Step 4
• We test the different options to mitigate the problem for all
the scenarios.
• We run the all the scenarios 30 times, and evaluate which are
the best strategies to mitigate the effect of the failures and
Fig. 4. Confidence interval of Option B absence of personal.
Conclusions
• The use of the simulation could help us to avoid the desire to always use the option with less workers,
because the cost of extra workers that are non-planned is regularly higher that the ones that are planned.
• The robustness test could help us to quantify better which schedule is more robust against the possible delays
due extra workers, failures and delays.
Mandatory Convertibles:
An Efficient Form of Capital Issuance
aHuerga A, bRodriguez-Monroy C
Angel Huerga (e-mail: angel.huerga@gmail.com) Dpto. de Organización, Administración de Empresas y
Estadística. ETS de Ingenieros Industriales. Universidad Politécnica de Madrid (UPM), España.
Carlos Rodríguez-Monroy (e-mail: cmonroy@etsii.upm.es) Dpto. de Organización, Administración de Empresas
y Estadística. ETS de Ingenieros Industriales. Universidad Politécnica de Madrid (UPM), España.

Abstract
Mandatory Convertibles represent only a small fraction of all the securities issued by corporate or financial institutions, however, they reach nearly a 30%
in volume of all the convertible securities issued every year and its popularity increases steadily over time. Mandatory convertibles can reduce the
shareholder dilution compared to a straight capital increase. Mandatory convertibles were commonly issued by financial institutions amid the global crisis
as a means to increase capital and could be considered as an antecessor of some types of AT1 (Additional Tier One Capital Instruments) and Contingent
Convertible Bonds. This research examines the scarce academic literature that studies this very particular security.

Mandatory Convert

Higher Potential Total
Return of MCS on Stock
Total Return

Price Downside Scenario

30000 Lowertrigger 100%
Annual Amount Issued USD Millions Lowerconversionratio 1
25000 Uppertrigger125%
Upperconversion ratio 0.8
20000 Lower PotentiaTotal
Return of MCS on Stock
Price Upside Scenario
15000
Stock Price
10000

5000

0
2012
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011

2013
2014
2015
2016

Delta of a Mandatory Convertible

Gamma of a Mandatory Convertible Close to Maturity
ICTs and their impact on the profitability of SMEs in the
main tourist areas of Guatemala
aRivera, Sigfrido, bRodríguez-Monroy, Carlos
asgrr@alumnos.upm.es
bcrmonroy@etsii.upm.es

Abstract
The primary objective of the study is to evaluate the effect of the use of Information and Communication Technologies (ICTs) as a tool for the improvement
of the profitability of small and medium-sized enterprises (SMEs) in the tourist sector of Guatemala, on the basis of a diagnosis of the current situation of
the use of ICTs and e-commerce for SMEs in 2015, in the main tourist areas of the country.
In this research, in the first place, the theoretical factors that led the SMEs to make use of ICTs were determined and discussed; subsequently, the main
variables that explain and differentiate tourist sector enterprises on the basis of the degree of implementation of the use of ICTs were identified, in order
to then determine their degree of competitiveness within the national tourist sector.
Finally, the conclusions and recommendations, that make it possible to verify what the theory exposes with respect to the use of ICTs matches the results
obtained with the SMEs evaluated.
The research aims to be a reliable source of information for public and private entities of the Guatemalan tourist sector, that will allow them to have a
clearer perspective on the use and adoption of new ICTs, in order to help them develop new comparative advantages and differentiating features, as they
face an economic landscape in an increasingly globalized and competitive world.
ICTs’ impact on profitability of SMEs
Tourism is an economic and social phenomenon that contributes to the development and well-being of nations, whose expansion benefits multiple
industries, representing about 9-10% of the world Gross Domestic Product (GDP) and generating 1 of every 10-11 direct and indirect jobs. It is also the
main source of foreign currency for one-third of the developing countries and for over one-half of the least developed countries, where it can represent up
to 40% of their GDP.
In the last decade, new ICTs have had an enormous impact on the tourist sector in the entire world. This impact is still in its initial phase in Latin American
countries, where the potential of e-tourism is particularly promising. For these reasons, countries such as Guatemala, which is located in Central America,
have a high potential for diverse types of tourism (cultural, archaeological and adventure among others). Guatemala considers tourism a vital component
of its development strategies and policies in general, and is investing in its growth because it is considered a key factor to promote socio-economic
progress, due to its linkages to other economic activities in favor of the dispossessed.
To maximize its tourist potential, it is necessary that companies in the tourist sector of Guatemala (mainly hotels, tour operators and travel agencies),
promote and commercialize their own tourist offer on-line, based on the collaboration between the public and the private sectors, thus improving the
visibility of Guatemala as a country of destination and collaborating with the local tourist industry in order to react in a better way to the market demand,
meeting the needs of potential tourists through a fast and secure access to information, the preparation of customized trips and facilitating on-line
payment.
Main results
To analyze the current situation of new ICTs applied to the Guatemalan tourist sector, it is important to observe how the pioneering companies in adapting
to these technologies have strengthened their competitiveness and have increased their market share. Most of these companies are located in the more
developed tourist areas, which are shown as blue circles on the map below. They have improved their dominant position in the tourism market, while
those that have not been able to adapt to the new technologies have been losing weight within the tourist market and are being increasingly relegated to
the point of endangering their own continuity. The majority of these companies are located in areas of scarce tourist development and are shown as red
circles on the map.
Finally, it was possible to verify as indicated by the theory, that if companies in the tourist sector implement e-commerce services and new technologies, in
all their amplitude, they would be favored with: more innovation and competitiveness, access to a potential market of millions of people, taking advantage
of all possible points of sale, direct customer contact and personalized marketing. These, together with other advantages, make e-commerce and the use
of ICTs great impact instruments the future development of the tourism industry.

Indicator Micro % Small % Medium % Large %
Proportion of firms that use computers 42,91 74,83 87,50 100,00
Proportion of employees who use computers regularly 28,09 30,78 25,83 33,18
Proportion of firms that use Internet 66,09 88,74 96,88 100,00
Proportion of employees who use Internet regularly 27,74 30,56 24,68 33,18
Proportion of firms present in the web 34,95 62,25 93,75 100,00
Proportion of firms that receive orders by Internet 8,65 26,49 28,13 100,00
Proportion of firms that place orders by Internet 5,88 18,54 21,88 100,00
Proportion of firms that use Internet classified by access mode:
Narrow band 19,72 43,05 40,63 0,00
Fixed broad band 6,92 10,60 31,25 100,00
Movile broad band 39,45 35,10 25,00 0,00
Proportion of firms that use Internet classified by type of
activity:
Sending or receiving e-mails 57,09 86,75 96,88 100,00
Using instant messaging or discussion forums 33,56 60,93 75,00 100,00
Obtaining information about goods and services 27,34 48,34 84,38 100,00
Providing access to other financial services 5,88 12,58 34,38 50,00
Providing services to customers 26,99 45,70 78,13 100,00
On-line sale of products 5,88 12,58 34,38 50,00
Internal or external hiring 21,80 39,07 65,63 100,00
Training of personnel 13,84 24,50 43,75 100,00
Innovation Management in Knowledge
Intensive Service Firms
Phd Candidate: a Isaac Lemus-Aguilar
Supervisors: b Antonio Hidalgo, c Raffaella Cagliano
a Universidad Politecnica de Madrid / Politecnico di Milano –
isaac.lemus@upm.es/isaac.lemus@polimi.it
b Universidad Politécnica de Madrid
c Politecnico di Milano

Abstract
This study aims to understand how is innovation managed in knowledge intensive service firms (KIBS). A exploratory multiple case study
was conducted in different consulting firms, mainly located in Spain and Italy. Main results show the Identification of the tools used to
support multiple innovation processes inside these firm, identification of the impact of information technologies in the generated
innovation output of these type of firms and identification of higher innovation capabilities for digital and open innovation. As a conclusion,
Information technologies are transforming radically the way in which service firms innovate, changing the nature of their offerings to
clients, new ways of work with their employees, the type of collaboration with other firms and even their own business models.

Theoretical Foundation Methodology
Industrial Facts A qualitative research strategy was used based on case studies.
KIBS are organizations that use and build knowledge as their The first phase of the study was conducted in consulting firms
primary raw material for their value-adding (Hidalgo & Albors from Italy between June and September 2015. The second phase
2008). Looking for a KIBS context, we found out that innovation in of Study was realized in consulting firms from Spain between
consulting firms has missed specific attention in academic April and June 2016. We contacted around 60 companies via the
research. university external relation offices, Linkedin and personal
networking. We selected companies from all ranges that had
Consulting firms usually help other companies to be innovative,
important projects and operation in both countries. In total, 20
but they also need to innovate themselves in order to survive in
consulting firms participated in our study (8 from Italy and 12
the market. However, consultancies possess particular
from Spain).
characteristics that do not correspond to generalizations made in
previous studies (Hertog et al. 2010; Rubalcaba et al. 2012; Chae Preliminary Results
2012; Furseth & Cuthbertson 2013; Hidalgo & D’Alvano 2014).
Consulting firms are facing changes with:
The open innovation stream has mainly focused in manufacturing
firms (Lichtenthaler 2011; Huizingh 2011; West & Bogers 2014; a) Extended use of open innovation. Becoming “Innomediaries”
Bogers et al. 2016), only few studies have investigated service and reinventing their role from “problem solvers “to “solution
firms (Chesbrough 2011; Mina et al. 2013; Janeiro et al. 2013), seekers” (Chesbrough et al. 2007; States et al. 2016).
therefore it is still unknown the use and effects of open b) Creating innovation ecosystems (Adner 2006; Tamoschus et al.
innovation in this context. 2015; Bogers et al. 2016). Large consulting firms have a big asset:
they have access to several clients. They can connect ideas from
companies in very different areas that otherwise wouldn’t have
been able to connect.
c) Coopeting (Bouncken & Kraus 2013; Ritala et al. 2014). It is
getting common to join forces with competitors not just to hand
part of a client’s projects, but to complement capabilities that
enable companies to explore and exploit competitor’s
knowledge.
d) Organizational Redesign: Using dynamic capabilities to
transform themselves , creating of an organizational structure in
Research Gaps charge of innovation, creating new business models, creating
The research explores how firms organize to manage innovation incubators and spin-offs, acquiring innovative SMEs, doing other
from an open innovation and innovation management sectors activities such as research centers and venture capital for
perspective as well as its relationship with service innovation and external and internal projects (Teece et al. 1997; Hobbs et al.
dynamic capabilities. It investigates the innovation processes in 2008; Teece 2010; Vanhaverbeke & Chesbrough 2014).
the context of consulting firms based on the open interaction
among main actors identified during the literature review from an Our results show that there are 7 different innovation processs
internal (management, employees) and external (customers, that affect the typology of the innovation management system in
competitors) perspective (Chesbrough 2010; Chesbrough 2011; consulting firms and that can be categorized in 3 groups: internal
Rubalcaba et al. 2012; Garriga et al. 2013; Furseth & Cuthbertson (management-, technology-, and employee-driven innovation),
2013; Chesbrough & Bogers 2014; Vanhaverbeke & Cloodt 2014; mixed (employee involved in user innovation, open innovation),
Tamoschus et al. 2015) external (policy-, user-driven innovation). Due to the analytical
To fill in the research gaps the following research questions were approach used in this type of organization to structure work to be
formulated. done and the role of the client in consulting services, most of the
1. How do consulting firms manage open innovation? case studies present a management-driven innovation and
employee involved in user innovation processes. This can be
2. Which processes, practices and capabilities do they use for contrasted with other innovation processes identified in
open service innovation? professional service firms such as client-driven product
3. What are the consequences of using open innovation in innovation and internally-driven process innovation (Smets et al.
consulting firms? 2012; Morris et al. 2015).
Determinants of governance structure in
multipartner alliances: a multiple case study
Frano Barbica,b, Antonio Hidalgoa
aUniversidad Politecnica de Madrid (frano.barbic@gmail.com)
b Politecnico di Milano, Italy

Introduction Methodology
This article contributes to the debate on the relation We use multiple case studies of multi-partner
between trust, risk and control in the management alliances formed between Italian SMEs to examine
of multipartner alliances. these relationships.

Multi-partner alliances are form of interfirm The authors conducted 32 interviews with
collaboration where multiple members, that managers of firms that belonged to 20 alliances
interactively work to accomplish joint goals, are from different industries that were studied during
governed by a single overarching contract. period 2014–2016.

Research context

Results
The table represents a
Aim
summary of results:
The aim of this paper is to explain how managers’ where different colors
perceptions of trust in partners and alliance risk, at of the rings represent
the start-up phase of an alliance, may influence the different levels of
choice of governance structures in multi-partner governance structure
alliances. complexity.

Conclusion
Analysis of the governance structure of multi-partner
alliances suggests that the structural preferences of
partners are based on their perceptions of different
types of risk (relational and performance risk), and the
overall objective is to minimize the total risk.
However, in several cases, the governance structure
reflected those proposed by instructions of the funding
agency. Instead of efficiency consideration, a
“bandwagon effect” was a driver of the adoption of
A theoretical framework for choice of governance structure in the collaborative alliance form.
multi-partner alliances
Network Cooperation - an emerging model of
connected collaboration for the Sustainable
Development Goals 2030 agenda'
aManuel Acevedo Ruiz, aAna Moreno Romero, aCarlos Mataix
Aldeanueva
a ETSI Industriales. Universidad Politécnica de Madrid

Abstract
An emerging approach to development strategy, Development 2.0, is placed to serve the new international
development agenda marked by the UN Sustainable Development Goals 2030. The international development
Cooperation sector needs a profound overhaul to adapt to the new strategy and remain relevant for the SDGs. This
research proposes a new, highly networked Cooperation coherent with Development 2.0. The model is contrasted with
examination of real cases and organizations to generate some tentative conclusions.

Research Purpose In a Representational (2-D)
To explore networking characteristics/behavior of international development Cooperation systems.
network, nodes connect with
Objectives: O1 the design and characterization of a networked Cooperation model in line with Development 2.0 nearby nodes or with the center.
trends and the SDG 2030 Agenda. O2 enhanced understanding of the conditions that can influence the The network acts on behalf/for
implementation of networked Cooperation systems. its members (advocating,
Hypothesis: H1 The adoption of networked structures/processes in Cooperation systems contributes to improve negotating, providing services,
their performance (in programs, projects, etc.); H2 The use of specific network strategies is a main factor for organizing events, etc.).
successfully implementing networked Cooperation schemes. H3 ‘Enabling’ or 3-D networks are more appropriate for
Cooperation’s work than ‘Representational’ or 2-D networks, as they stimulate participation and collaboration.

Theoretical Framework
Development 2.0 refers to new ICT-enabled models than can transform the processes and structures of
international development (Heeks). A Development 2.0 approach derives from the interaction between the
paradigms of Human Development (Sen) and Network Society (Castells), and needs to support the successful
In an Enabling (3-D) network
adaptation of Human Development strategies and practices to a Network Society context. From a research
any node can connect to any
perspective, it seeks a conscious and sustained dialogue between Web 2.0 models/ways of thinking, and the
other one. The network seeks
broader structural concerns within development studies (Thomson). Innovation is a driver of change for
to strengthen capacities of its
Development 2.0 practices (including in re-examining North-South development relations). Development 2.0
members to achieve their own
flourishes in Open environments that facilitate wide access to information/knowledge, sharing and participation
objectives. Most actions result
(Smith & Elder).
from collaborative activities
Methodology carried out by members and
supported by network resources
• Literature / institutional review
• Questionnaire to development/ Cooperation professionals in (quantitative and qualitative data) about the
inclusion of networking in Cooperation systems and the management of development networks
• Case studies on three development networks
• Analysis of external evaluations of three large development programmes Contrasting Hypothesis
H1 Questionnaire informants indicated positive relation in
Conclusions their work between incorporating network
structures/practices and results. Still limited real networking
practices, but positive perception of its potential outcomes.
Programme evaluations showed that their projects
supporting/creating networks obtained better outcomes,
particularly in capacity building. (H1 appears refuted)
H2 ‘Conscious’ institutional networking is a relatively recent
trend. Proactive practices with network strategies/managers
show clear positive relation for adoption of network
practices. Explicit network strategies still a minority (when
existing, they`re still largely implicit). (H2 is well refuted)
H3 2D and 3D networks represent extremes of ‘operational
centrality’. Most networks examined (case studies,
evaluations) are closer to 2D profiles. 3D networks not
necessarily more appropriate, thus H3 to be re-considered,
as development objectives are wide-ranging and should be
the key criteria determining network characteristics.
Capacity building objectives seem better served by 3D
profiles, while 2D networks better for policy
influence/service provision. 3D networks are more mature.
Network effects in Dev 2.0 systems
Network Cooperation model – key attributes • Network Cooperation to generate beneficial effects in all
• Incorporation of institutional networks as primary project delivery mechanisms. elements of these systems, (i) structures (arquitectures),
• High degree of ICT mainstreaming (tools and capacities) into Cooperation agencies (ii) resources (inputs), (iii) processes, (iv) outcomes
• Emphasis on knowledge management as the underlying basis for Cooperation process (outputs), as well as the development context
• Design of development projects as networked initiatives • Network effects at distinct Cooperation levels: (i) micro
• Expansion of networked social capital generation (projects); (ii) meso (organizations); (iii) macro (systemic,
• Extension and promotion for multi-stakeholder actions eg. UE, UN System, OCDE-DAC)
'Process analysis of a 4 axis single point micro
cutting machine'
aLópez-Estrada,Luis; bFajardo, Marcelo; c Vizán, Antonio.
aUPM-ETSII-Department of Mechanical Engineering (imt.luis.lopez@gmail.com)
bUPM-ETSII-Department of Mechanical Engineering (marcelofajardopruna@gmail.com)
cUPM-ETSII-Department of Mechanical Engineering (avizan@etsii.upm.es)

Abstract
The current advances when creating micro technology demand that the machining tool market evolves in order to develop devices capable of achieving
parts that require high precision for smaller dimensions, in order to compete with the current technologies available, the machine in this paper its
presented.
Current technology for machining micro parts rely mostly on micro milling, but giving the nature of the process, its control and the geometries that can be
achieve require a considerable complex system, in order to reduce these weaknesses, a 4 axis micro-machining device, capable of moving in a cartesian
system, with the advantage of using a single edge tool, and has ability to move the tool orthogonal to the cutting path has been developed. This special 4-
axis machine is capable of placing the velocity of cutting tangentially along the tool trajectory during the machining process.

The Machine
Characteristics
The design of the machine aims to simplify and facilitate the processes of
metal cutting in micro-machining, the advantages of using single point
cutting techniques are widely superior to the use of micro-milling ,
smaller geometries can be achieved with less complicated tool design.

Elements
The main components of the machine is a precise CNC controller capable
of achieve 1 micron accuracy in all Axis (X,Y,Z,C).
Two vision cameras to monitor the process and to measure both , the 4 Axis single point micro cutting device
geometry of the piece and the wear of the tool. High precision CNC and monitoring artificial vision system.
A high precision dyno meter sensor that is capable to measure all the
forces along the 3 cartesian axis in order to analyze the process and
being able to develop a function of parameters in terms of the values
acquired.

The Process
The design of the machine aims to simplify and facilitate the processes of
metal cutting in micro-machining, the advantages of using single point
cutting techniques are widely superior of those in micro-milling , smaller
geometries can be achieved with less complicated tool design.
Schematic of the single point micro cutting process
A function in terms of the cutting force, the specific cutting pressure and
the area of the cutting section is obtained to have a greater control over in the machine
the process. The forces involved changes depending on the different trajectories in
Current tests show that competitive geometries, that rival the ones the process, changing the behaviors of the tool and the manufactured
achieved with current techniques, are viable. piece.

Single point tool, and geometry demonstration.
With the vision system we are able to monitor in real time the process , not just by analyzing the forces involved , the results are
comparable with current technologies. Scan the QR code to see a video of the process.
Cutting  parameters  identification  in  end  
milling  operations
aE.  Leal-­‐Muñoz, aA. Vizan
a  Department  of  Mechanical  Engineering,  Universidad  Politecnica de  Madrid,  Madrid,  
Spain. (erardo.leal.01@gmail.com)

Abstract
Nowadays, it is easy to find manufacturing systems highly automated. Actions are focused on the introduction of automatic systems and smarter
manufacturing systems in order to increase the productivity. In this sense, this work presents a methodology that allows the online identification of cutting
parameters considering system dynamics. The methodology has been prove using a cutting force model and an experimental validation has carried out.
Results shows a highly accuracy and the methodology is able to be used to the online correction of cutting parameters.

Ø End  milling  is  characterized  by  low  radial  depth  of  cut  and  most  of  time  
only  one  flute  is  in  the  cutting  zone.

Ø The  proposed  methodology  allows  the  estimation  of  both  radial  and  axial  
depth  of  cut.

Ø Machining  process  dynamic  is  considered  in  the  estimation  procedure.

Ø Machining  test  were  carried  out  with  variation  of  radial  depth  of  cut,  axial  
depth  of  cut  and  both  depth  simultaneously.    

Ø Results  shows  a  high  accuracy  in  the  estimation  of  the  depths  of  cut  with  
an  average  error  less  than  1.5%.

Fig.  1.  End  milling  process

Dynamic  response  of  the  system
Real system can be divided into two subsystems: Process dynamics and measuring system. The
measuring system include the force sensor, signal conditioning and data acquisition. Process
dynamics are related with the tool, workpiece, feed drives, cutting conditions and spindle rotation
speed. This dynamic response of the process produce a delay that affects the measuring of the
time associated with the estimation of the axial depth of cut.
Modal analysis of this two subsystems was carried out. Results shows that the most significant
influence is by the process dynamics. Within the factors of influence, the spindle speed is the one
that most affects the system delay. The value of system delay is used to correct the time
measured in the force signal.

Results
Experiments in a real machining process was carried out. The experiments was designed to
produce a variation of radial depth of cut and axial depth separately and one experiment in which
both parameters vary.
In figure 3.A the results for axial depth of cut are exposed and in figure 3.B results for radial depth
of cut variation are shown.

(A) (B)

Fig.  3.  Results  for  axial  depth  of  cut  (A)  and  radial  depth  of  cut  (B) Fig.  2.  Scheme  of  the  estimation  method
Development of a 3PRS‐XY Machine Tool to 
Single Point Cutting Process in Microcutting
aFajardo, Marcelo; bLópez‐Estrada, Luis; cVizán, Antonio

aUPM‐ETSII‐Department of Mechanical Engineering (marcelofajardopruna@gmail.com)
bUPM‐ETSII‐Department of Mechanical Engineering (imt.Luis.lopez@gmail.com)
cUPM‐ETSII‐Department of Mechanical Engineering (avizan@upm.es)

Abstract
This work presents the development of a machine tool to be used in a single point microcutting process. This machine tool is composed by a 3‐PRS parallel
mechanism that positions the cutting tool in the Euler angles α, β and in the Cartesian Z position. Since to define a point P into space completely is
necessary provide its 6 variables Px, Py, Pz, α, β, γ, this machine tool use a linear XY table to provide the 2‐DOF in X and Y that the parallel mechanism
can’t, the P point to be defined is the tool‐tip end. Because of the single point cutting process is necessary to be able of rotate the cutting tool along its
longitudinal axis to maintain the cutting edge always oriented in the feed direction. In summary, the developed machine tool has 6 DOF that fulfill the
requirements of the process. The behavior of the machine tool is analyzed used a Simulink® model and is compared with experimental results. The model
considers the mechanical properties of the joints, stiffness and damping. The developed machine tool has its main application in the machining of small
parts.

Introduction The Machine ‐ Tool
The growing requirement of microproducts and the development of
applications that requires parts and components increasingly smaller
Characteristics
has allowed the development of specific solutions to satisfy these The developed machine (Figure 1) is constituted by a 3‐PRS parallel
requirements. Some of these solutions are new and others are inspired mechanism and a X‐Y linear table. The parallel mechanism is used to
in conventional processes as EDM, LBM, USM, cutting processes (e.g. position the tool‐tip in the workspace, the X‐Y linear table and the
milling, turning, etc). This work is focused in cutting processes that “spindle” rotational axis allow compensate parasitic motions and
involve one single cutting edge and mainly in the machine tool required provides the process feed.
to that process.
The goal of this work is to develop a machine tool with a structure that
allows using single‐edge cutting tools in a single‐point cutting process.
Due to the properties of the process, the mechanism must be able to
rotate the cutting tool to keep the orientation toward an orthogonal
cut. The cutting tool must be rotating simultaneously around the three
cartesian axes in order to follow a defined cutting path. To achieve each
position in the cutting path a combination of two structures is
developed, a parallel mechanism composed of three PRS limbs (3‐PRS)
and two linear axes in a XY‐ Table. Because of the DOF of the parallel
mechanism, an additional DOF to turn the cutting tool around its
longitudinal axis is considered. The developed machine tool then, has 6 (a)                                                       (b)
DOF. Figure 1. Developed Machine‐Tool (a) CAD Model (b) Actual Machine
State of art in machine tools for micro cutting
In microcutting the process that had been studied the most is Simulink® Model
micromilling. However, this process uses a multi‐tooth cutting tool that Using Simulink® a virtual model (Figure 2) is developed to analyze the
requires a complicated manufacturing process to get an adequate behavior of the 3PRS‐XY machine‐tool onto several work conditions.
cutting tool. It means more and more expensive tools for smallest Then, the modeled performance is compared to the actual one since
dimensions. In the other hand, processes that use simplest cutting the virtual model consider internal mechanical properties of the joints
tools as microturning, microgrinding or fly cutting have restricted as damping coefficient and spring stiffness.
applications. Microturning can produce micropins until 100µm, In the tool‐tip a model of the cutting process is used as an input to
microgrinding, depending on grain size of grinding wheel, is capable of study the static and dynamic behavior of the machine tool for a single
produce micropins and microgrooves, fly cutting is a method basically point cutting process.
to cut grooves.
Since the cutting process has been widely studied in sizes that would
be considered as macro scale, it is an attractive viewpoint that this
knowledge would be able to be used in smallest scales. There are
machines tools to microcutting process that are a miniaturization of
theirs similars in macro scale manufacturing as in micromilling and
microturning. However, in microcutting some phenomena than in
major scales has no effect in the process in micro‐scale has an
important role or at least more incidence in the process. These
phenomena could be grouped in three sections, work material, cutting
tools and machine tool. Hardness and grain size of material and cutting
tool; homogeneity, sharpness and run out in the cutting tool and;
positional accuracy, stiffness, damping, thermal stability in the machine
tool, are some of the characteristics that have more influence in
micromachining than in conventional manufacturing. Figure 2. Simulink® block diagram
Simulation of the Simulink® Model
With the Simulink® model we are able to analize the behavior of the machine tool previously to programm the CNC installed in the
actual machine (Fagor 8070) and determine if the input parameters are valid. Scan the QR code to see a video of the machine tool
simulation and the proposed single point cutting process.
Characterization the physical and chemical properties of epoxy
resins for additive manufacturing by photopolymerization
aE D N. Niño, bJ L Endrino and A aDíaz Lantada

aUniversidad Politécnica de Madrid (deydannv@gmail.com)
bCranfield University

Abstract
The photopolymers used in additive manufacturing may benefit from the incorporation of nano-particles towards enhanced physical and chemical
properties. In this work, we study this phenomena by means of the manufacture of a set of substrates with different geometries and bulk
functionalizations. The test probes were obtained using laser stereolithography (3D printing Form1+ system by Formlabs). The substrates were modelled
using a computer-aided design software and obtained layer by layer with a layer thickness of 100 microns. "Clear" epoxy resin (a mixture of esters of
methacrylic acids, oligomers and methacrylate monomers and photoinitiators) was used for the manufacture of the substrates. Additionally, nano-
additives of Cu at different concentrations (5%, 10% and 15%) were incorporated to the resin and their impacts in the final physical chemistry properties
were analysed. The procedure, which leads to 3D functionalized geometries, shows the potentials of bulk functionalization of resins and the viability of
additive manufacturing with the functionalized materials.

Methodology of Manufacture Process
Table 1. Mass measurements of the nano-additives and solutions
Nano-additive type Nano-additive mass (g) Solution mass (g)
Cu Nws 1 0.0098 6.003
Cu Nws 2 0.0666 6.831
Table 2. Mixture Features (a) (b) (c)
Additive percent Mixture time (minutes)
Resin type Additive type
(% w/w) CPM-Vacuum machine Magnetic stirrer bar
Without additive N/A N/A N/A
5.08 30 N/A
Copper Nanowires (Cu Nws) 1
9.97 30 N/A
Clear
5.08 60 60 (d) (e)
Copper Nanowires (Cu Nws) 2 9.97 60 60 Figure 1. Resin types used in the
15.37 60 60 substrates manufacturing. (a) Cu Nws
solution, (b) Clear resin and resin mixed
Substrates Manufacturing with Cu Nws at (c) 5.08% w/w, (d) 9.97%
w/w and (e) 15.37% w/w

(a)
(a)
(a) (b)
Figure 2. 3D printer Form 1+. (a)
(b)
Frontal view and (b) resin tank
(b)

(c)
(a) (b) (c)
Figure 5. Substrates for testing traction manufactured
in: (a) clear resin and resin mixed with Cu Nws at (b)
(d)
5.08% w/w and (c) 9.97% w/w
(c) (d) Figure 4. Substrates for testing fatigue manufactured in: (a) clear
Figure 3. Circular substrates resin and resin mixed with Cu Nws at (b) 5.08% w/w, (c) 9.97% w/w
manufactured in: (a) clear resin and (d) 15.37% w/w
and resin mixed with Cu Nws
at (b) 5.08% w/w, (c) 9.97%
Preliminary Results (a) (b) (c)
Figura 6. Circular substrates manufactured in:
w/w and (d) 15.37% w/w Cu Nws 10% (a) clear resin and resin mixed with Cu Nws at
Cu Nws 5%
14000 Reference (b) 5.08% w/w and (c) 9.97% w/w

12000

10000
Intensity (a.u.)

8000

6000

4000

2000

0
0 500 1000 1500 2000 2500 3000 3500 4000

Raman shift (cm-1)
Figure 7. SEM and EDS in the substrate Figure 9. Raman Spectroscopy of substrates manufactured
with Cu Nws at 5.08% w/w to 4000X in clear resin and resin mixed with Cu Nws Figure 8. SEM and EDS in the substrate
with Cu Nws at 9.97% w/w to 4000X
Cell scaffolds design using pentamode
structures based on Bravais Lattices
Graciela Fernández Méjica, Andrés Díaz Lantada

graciela.fernandez.mejica@alumnos.upm.es
adiaz@etsii.upm.es

Abstract
Development of new “metamaterials” allows the design of materials with specific physical properties depending on the final application of that material. A
very recent set of mechanical metamaterials are pentamodes, materials somehow imitate the behavior of liquids but being solid
Nature offer variety of possibilities in which the metamaterials design can be based on, as the 14 Bravais Lattices. With a design based on Milton and
Cherkaev pentamode structure, it is confirmed these 14 structures have pentamodal behavior,.
Due to the dimensions that some 3D Printing technologies allows to obtain (as two photons lithography), this metamaterials can be synthesized in the
range of microns and be used to develop cell scaffolds and bioreactors for cells strains growth and, also, for actuators with some specific modal behavior.
These devices could have different physical properties along their structure and could be synthesized in one-shot.
Functional Graded Structures
Basic Design
• Elemental / column / pattern

• Minimun number of branchs Combined Structures

• Nodes only where atom exists

• Parameterized structure

Mechanical Properties Frecuency analisys
Pentamode  B/G > 103 [1]

These 14 structures have pentamodal behavior, which can be divided in
two groups. The first group, with a very remarkable pentamodal behavior
and the second group with behavior also pentamodal, but not so
remarkable. [2]

Cell Scaffolds & Bioreactors Actuators
Functionally graded scaffolds constitute a key Structure manufactured in a
development in the areas of tissue engineering and Form 1 3D Printer (25µm layer
biofabrication, as their mechanical properties could thickness)
be tuned to be used to simulate the human body
transitions from a rigid to a flexible environment,
and then synthesize human tissue of different
human body parts, for example cartilage and bone
tissue at once. [3]

References
[1] Kadic, M., Bueckmann, T., Stenger, N., Thiel, M., & Wegener, M. (2012). On the practicability of pentamode mechanical metamaterials. Applied Physics Letters, 100(19),
191901; 191901.
[2] Méjica, G. F., & Diaz Lantada, A. (2013). Comparative study of potential pentamodal metamaterials inspired by bravais lattices. Smart Materials and Structures, 22(11),
115013; 115013.
[3] Diaz Lantada, A., Fernandez Mejica, G., de la Pena, M., Tellez, M., Garcia Ruiz, J. P., & Alarcon Iniesta, H. (2016). Tissue engineering scaffolds for osteochondral repair.
Microsystems for Enhanced Control of Cell Behavior: Fundamentals, Design and Manufacturing Strategies, Applications and Challenges, 18, 331-349.
Modelling of the mechanical behaviour of the
photopolymerization processes of resins intended for
additive manufacturing (AMT) using laser-
stereolithography: The influence on part quality
ade Blas, Adrián; aFernández, Guillermo; aDíaz, Andrés
aDepartamento de Ingeniería Mecánica (ablasromero@etsii.upm.es)

Abstract
This study presents a simplified model of the polymerization process of photo-cured epoxy resin, with the aim of finding theoretical values for the
mechanical properties of the material in its transition stage from liquid to solid state, approximating these properties to a single figure at the instant of its
gelation in the time denominated as tg. By evaluating a phenomenon observed during the fabrication of samples using a stereolithography additive
manufacturing printer, it is hoped to fit a model that will replicate the mechanical forces undergone by the manufactured part. The introduction of new
hypotheses to simplify the case under study will be tested by simulating these cases using the finite element method, taking the values obtained from
previous publications that used experimental and analytical analysis so that the congruence of the results will be constant throughout the model. Finding
these theoretical values will help develop future criteria for the feasibility of manufacturing the parts using laser stereolithography and therefore have a
direct influence on the quality of the end part.

1. Study aim
We are aware that the rise in production using SLA printers has proportionally led to the need to find a simple method to allow the user to know the
feasibility of manufacturing the end part, and should it not be feasible, to reconsider the design and apply corrective factors to the thickness and reduce
critical stress situations as already shown in previous publications Stava et al.
Therefore, we propose a model that simplifies the epoxy resin polymerization photoelastic process and the dependency on the time slope of the
mechanical properties of the material to be used in the simulation (E(dt), ν(dt))[7] for a constant value of these properties in its gelation state (E(tg),
ν(tg)), the point at which we suggest the material reaches a critical situation (breaking strain).

Phenomenon under study
By making two calibration test pieces CP01 and CP02 (Figure 1.c) printed with
SLA3500 (3D Systems Inc, Rock Hill, SC 29790, USA) and Accura® 60 epoxy resin
(3D Systems), to enable the team to create a database of critical situations or
dimensional errors in the manufacture of the parts by SLA ‘free surface’, it was
possible to observe a phenomenon that up to then had been recorded in the
literature.

Hypothesis and process simulation
Taking an in-depth review of the literature as the basis, the factors that will
influence the process giving rise to this deformation can be defined: a)
maximum linear creep of the epoxy resin εmax; b) the volumetric action of Fig. 1 a) ‘Curling’ effect of a cantilever beam due to creep
gravity; c) surface support in the lower section due to the variation in density through ‘free surface’ polymerization processes; b) calibration
(∆ρ= ρl- ρg); d) fixed constraints of the element due to its design. test piece and deformation observed; c) CP01/02 specimen of
study and E11 testing element.

2. Computer implementation
In order to validate the suggested model a simulation was
performed on a section (12 mm x 12 mm x 0.6 mm) separated
into 6 layers of 0.1 mm (Figure 2.a) representing the successive
layers that the process would polymerize, each layer having a
minimum of two octahedral elements (Figure 2.b) on the
Scheme 1 Layout of the procedure used through finite element simulation length of the thickness. The successive layers were modelled
with a solid joint while the fixed displacements were applied to
3. Results and conclusion the opposite sides of the sections
To evaluate the suitability of the Modulus of Elasticity in the state of gelation a) b)
Eg, we proceed to trace a trend line for the experimental value obtained by
Guees et al. for the initial elastic modulus of a just-polymerized strain of epoxy
resin SL 5170 and its dependency of the time after the hatching, until 20 first
minutes.
A study has been conducted that started out from
an observed phenomenon. It has been possible to Parameter Value Units c)

evaluate and model the behaviour of Accura® 60 Еg 2.1 MPa
(3D Systems) epoxy resin during the process of νg 0.3 -
laser stereolithography Additive Manufacturing ρg 1.17 Kg/m3
εg 0.016 -
with an SLA 350 printer by simplifying a time-
tg 49 s
dependent case in a linear elastic stationary Fig. 2a) Representation of the geometric model to be studied:
model. interspersed displacements imposed on the layers; b)
This project was financed with the support of the European project: Tool-less symmetric meshed model for the FEA; c) deformation of the
manufacture of complex geometries’, funded by the European Commission model through the action of the loads and imposed constraints
under the broad agreement 633192 - H2020-FoF-2014-2015/H2020-FoF-2014
Similarity index as a tool to compare impact 
history records
aAntonia Pacios Álvarez, M. Consuelo Huerta Gómez de Merodio, 
Jesús Alonso Álvarez, José Antonio Parra‐Hidalgo, Luis López Lázaro
aDepartment of Mechanical Engineering (antonia.pacios@upm.es)

Abstract
To compare several impact time histories obtained from experimental tests ‐in situ or laboratory‐ or finite elements simulations, existing correlation
methods can be used. However those methods proved to be sensitive to some parameters, magnitude for example, but unaffected by phase shift or shape
difference. Based on the knowledge of physical behavior of the phenomenon and the characteristic parameters, a method to quantify the similarity
between two impacts is proposed. Three indicators are defined: the impact level (the impulse), the stiffness (duration of the impact), and the system
response (maximum acceleration). A "dispersion" function is defined for each indicator that shows the variation of an impact against a reference one. With
the objective to integrate the multiplicity of indicators the function of similarity proposed is a simple addition and the Similarity Index (SI) assigns a
numerical value to the discrepancy between two time histories over a period of time and can be calculated as the weighted addition of the indicators.

Main Objective
While the certification of a security glass ask to determine the human impact resistance in laboratory conditions (Fig. 1), when there is the need of
proving some fixation elements or new system, in situ testing has to be performed (Fig. 2). Records from impact performed at different pendulum heights
(continuous lines) are presented in Fig. 3. Even thought they look similar, not always is possible to quantify how close time histories are or if FEA
simulations (symbols line) are accepted. Disparity of results ask for a tool that allows to state the similarity between impact history records.

Several testing configurations and experimental and simulation representative graph a

b

Fig. 1. Lab testing: impact pendulum according to  Fig. 2. In situ testing: approach to impact  Fig. 3. Comparison of experimental 
UNE‐EN 12600 and portable impact testing,  UPM pendulum, examples from Bellapart results and FEA simulation

Proposal for Similarity Index Determination
Previous work shows that duration of the impact, máx. acceleration and contact area are the most representative parameter to reproduce an impact. The
first two and the impulse are considered the best indicators to reproduce the event. A value function (Fig. 4) is defined that associates a number between
0 and 1 to each dispersion value, 0 for maximum and 1 for no dispersion; lineal, cosine and exponential functions were tested. Similarity Index (SI) can be
calculated as the weighted addition of the indicators.
Characteristic Indicators Value function (ax, (1/a)x, cos x) Evaluation of weight factors
Indicators considered (Fig. 3) 1 Weight coefficients kT, kF and kI are the
(a) STIFFNESS: Duration of impact in Hz (T)
given values to obtain Min ( (ISIND ‐1)2 )
(b) SYSTEM BEHAVOIR: Max. acceleration (Fmax)
(area)   IMPACT LEVEL: Impulse (I) 0 To integrate the multiplicity of indicators
Max. Deviation the function of similarity proposed is a
(T, Fmax, I) simple addition.
Fig. 4. Example of value function  IS = kT * INDT + kF *INDF + kI * INDI
IS = 0.25∙INDT + 0.1∙INDF+ 0.65∙INDI
Results
COMP.1          COMP.2          COMP.3          COMP.4          COMP.5          COMP.6

As it can be seen in Fig. 3, the impact registers from height of 190 and 200 mm
used for comparison 2, black and orange lines, are identical; in Fig. 5 the value
for SI is close to 1, as it was expected. However impact registers from height of
300 and 450 mm used for comparison 5, green and orange lines, being very
similar, present an SI= 0.32 since the influence of INDI is relevant (KI=0.65).
When only a single indicator is used to represent an impact the Impulse (INDI) is
the most representative.

Fig. 5. Values for SI when comparing couples  of impact records
Conclusions
The function of similarity presented as an weighted sum of three indicators (frequency, force and impulse) is the best approximation to correlate the grade
of similarity as it can be observed from the max. values and shape of the temporal registers. The SI has been used to compare series of data from a
database of 134 soft human impacts time histories, for different impact heights, 3 glass plates configurations and 4 support conditions with good results.
Human Impact on glass: study of the excitation
level measured on a plate versus energy applied
by the pendulum
a Mª Consuelo Huerta,  Antonia Pacios‐Álvarez, 

José A. Parra‐Hidalgo, Jesús Alonso, Sergio Pardo,
a ETSII – UPM (mariaconsuelo.huerta@upm.es)

Abstract
Nowadays it is too difficult to establish the “fracture load” during an impact on glass plates due to their brittle fracture. Therefore, in many cases it is
performed an impact test in situ, using a pendulum similar to the pendulum from standard UNE‐EN 12600. This paper presents the existing
discrepancies between the actual excitation level of the glass plate and the estimated one calculated from the potential energy of the pendulum.
These discrepancies are produced due to the distribution of the initial energy of the pendulum between the glass plate, the support frame and the
pendulum head, and the energy dissipation during the impact mainly caused by the damping of the pneumatic. In order to avoid these discrepancies,
a method is proposed which focuses on the impulse measured by an accelerometer located on the pendulum.

Main Objectives 
To establish a procedure to represent in situ tests the actual impact level  Test setup
applied to glass plates it is necessary to define an easy and accurate procedure, 
with high repeatability, that allows to determine the load applied to the plates Measurements of 
head pendulum: 
time
This work presents the use of the register of acceleration from impactor head  • Initial height (h0)
as a measurement of maximum load and impulse. • Final height (hf)
• Impulse (I)

Data from materials and experimental work
Glass plates Results from modal test
• Tempered glass plates (T) of 10 and 5 mm thickness ; Laminated glass plates  GLASS PLATES IMPACT 
(L) out of annealed panes of different thickness (5+5, 4+4, 3+3 mm). PVB  T104S L554S HEAD
interlayer.
F, frequency (Hz) 35.9 36.24 13.27
• Continuo four sides supported (4S); two sides supported and four point 
ζg, ζp damping (%) 0.84 1.78 3.86
fixtures.
Data from impact test
• Nominal impulse, as the head pendulum theoretical velocity without 
dissipation (In)
• Pendulum impulse, as the head pendulum initial and final velocity (Ip)
• Impulse on the glass, as the force integral (Ig)
• Initial and final energy (E0, Ef)
• Head energy loss (Ep)
Comparisson of the considered impulses

Pendulum head T104L L554L
Equivalent damping 2.90 4.09

Proposed model to define glass impact level
H1: Energy dissipation during the contact stage  H2: Estimation of energy distribution
• Element damping (ζp, ζg ) • Partition coefficient: E0= Eglass + Emachine = r E0+ Emachine
• Equivalent damping from the energy loss (Ep) o Optimization procedure: 
o Design variable:   r  
o Target function: min((mp (v0,g +vf,g) – Ig)/ Ig)2 )

• Ep = E0 – Ef = 0.5 mp( v0,g2 – vf,g2 )‐>   r  E0 = 0.5 mp v0,g2

T104S L554S
Partition coefficient 0.71 0.70

Conclusions
• Damping values obtained from modal test  support the hypothesis that  energy dissipation occurs during contact stage.
• In the test performed according to EN 12600, impulse registered with the accelerometer in the pendulum head is equal to 70% of initial energy. 
• Those results can be incorporated to numerical models introducing damping, and considering initial velocity as the 80 % of nominal velocity .
• To determine the impact level from the impulse is a more accurate measurement than being determined by the initial height; specially when using  
values obtained from in situ test. 
Smartwatch’s usefulness for hand - arm
vibration exposure assessment
aL. Sigcha, I. Pavón, J. M. López, G. De Arcas.
aInstrumentation and Applied Acoustics Research Group (I2A2).
Universidad Politécnica de Madrid. (ignacio.pavon@upm.es)

Abstract
This study analyzes the opportunities offered by wearable technology for occupational risk assessment and specifically the use of smartwatches for hand-
arm vibration exposure assessment. The results suggest that this kind of devices could be able to be used in specific hand-arm vibration assessment tasks.
It has been found some technical limitations but also it has been identified opportunities for improvement in the near future.

Introduction
Exposure to hand-arm vibration is a major cause of occupational diseases. The vibration exposure risk assessments are usually made from taking discrete
time measurements, this imply inaccurate the risk assessment and difficulties for the comparison between measurements and the limits established by
law. In order to improve the vibration risk assessing procedures is possible to use wearable technology (smartwatches) for risk assessment, exploiting the
advantages of the smart devices instead the traditional vibration measurement systems.

Materials and methods RESULTS
The scheme of the figure was used to implement a measurement system Frequency and amplitude results (Lab. test).
following the guidelines of the ISO 5349-1:2001 standard. The system
calculates the frequency weighted acceleration ahv and the value of The results from laboratory test in the table show that the differences
exposure to vibration A(8). The test device used was a Sony 3 (SWR50) between ahv measurements made with the two devices are small when
smartwatch with a Bosch sensor (BMX055). the acceleration amplitudes are low, but they increase at high amplitude
vibration levels especially in the frequencies of 50 and 100 Hz.
Acceleration RMS Frecuencia (Hz)
Wh Filter
X (ahvx)
Amplitud (mVrms) 6,3 12,5 25 50 100
Vibration Acceleration RMS
Wh Filter ahv A(8)
Source Y (ahvy) 200 -0,04 -0,04 -0,03 -0,05 0,021
Acceleration RMS 400 -0,02 -0,04 -0,05 -0,09 0,039
Wh Filter
Z (ahvz)
600 -0,01 -0,01 -0,02 -0,1 1,636
Test 1: Evaluation of frequency and amplitude 800 0,01 0 0,95 2,76 1,499
1000 0,25 0,47 1,07 2,23 0,855
(Laboratory test).
The test was performed using a shaker vibration bank fed by sinusoidal Common drill machine results (Outdoor test).
signals at octave intervals. A reference vibrometer and the smartwatch The test show small differences for low amplitude levels that increase
were coupled simultaneously on the shaker vibrating surface and then 10 when reaching the signal saturation. It can be observed that in some points
ahv measurements were made at different levels and frequencies. there is a higher difference between the measurements of the devices,
these errors are produced mainly by the variability of the sample rate and
the fitting to the hand drill.
8
8

7
7
ahv reference m/s2

ahv reference m/s2

6
6

5
5

4
4

3
3

2
2

1
1

0
0
0 2 4 6 8
0 2 4 6 8

Test 2: Evaluation of the analysis system using a ahv smartwatch over the ahv smartwatch over the
machine m/s2 wrist m/s2
common use machine (Outdoor test).
A simulated test was made using a hand drill, measuring simultaneously
the weighted acceleration ahv with the reference vibrometer and the
CONCLUSIONS
• The device has limitations to be used as a precision instrument, mainly
smartwatch. To perform this experiment, an electric drill was used to make
due to the limitation of the sampling rate (Fs = 250 Hz) and the
several drills in different materials for a time at least 1 minute.
dynamic range (±2g) in which the sensors of smart devices are set
during manufacturing.
• The results show that system is suitable for evaluation of machinery
with low ahv acceleration levels. This situation can improve in the near
future when smart consumer devices get higher performance in their
sensors.
• Nowadays these devices could be used to obtain better
representativeness of the exposure levels because they can be worn
during the entire working day.
• The cost reduction by using devices and sensors with relatively low
prices compared to the specific vibration measurement equipment can
make this kind of devices a tool to develop a safety culture and
education of workers, company, managers, etc.
Bending strength study of repaired coaches
structures.
aDavid Rincón, bEnrique Alcalá
aINSIA, Universidad Politécnica de Madrid, (david.rincond@alumnos.upm.es)
bINSIA, Universidad Politécnica de Madrid

Abstract
Buses and coaches accidents involve a high number of passengers, so the safety of these vehicles are a major issue. The high investment made on them by
the transport companies makes their repair a suitable solution after some accidents. However, there is no regulation to evaluate whether a repaired coach
can maintain the characteristics of its initial condition or not.
The objective of this study is to quantify the loss of energy absorption capability of the profiles used in the construction of these vehicles and to model the
behavior of a repaired profile with FEM models.
In order to accomplish the objective, several test have been carried out by means of bending tests and calculating moment versus angle curve for three
different load cases: 5, 10 and 20 degrees of previous permanent deformation.
The results show a high loss in energy absorption capability after a repair of 20 degrees of permanent deformation, and a methodology for simulate these
results with FEM models have been developed.

Introduction Methodology
Rollovers are the most severe accidents of This research consists of the following two phases: The first one is the experimental study of
coaches and their consequences have required the behavior of the steel rectangular hollow sections by means of bending tests. The second
specific countermeasures to diminish the risks in one has consisted in emulate the behavior found in the tests with FEM models as faithfully as
case of accident., as the UNECE Regulation 66 it can be. The sequence of both tests and F.E.M. models are shown below.
Rev00. Since then, several improvements to the Sequence of tests
Reg. 66 requirements have been established,
being the most relevant of them the
consideration in Revision 01 of the 50% of the
mass of the passengers as consequence of the
Tests

mandatory installation, and use, of the seat belts.
Coaches that have been involved in a rollover are
often repaired straightening their structures in
repair benches. To date it has not been
considered in any regulation, the establishment Post
of repair or inspection criteria for structures of Accident Final
Initial State Straightened Straightened
coaches after experiencing an accident. Deformation Deformation
State
It has been considered a need to evaluate,
experimentally in a first step, the loss of energy
absorption capabilities of repaired rectangular
F.E.M.

hollow sections. A methodology to simulate this
loss of mechanical properties in FEM models is
mandatory, in order to investigate this change of
behavior in profiles with different section without
the cost of having to run tests.

Results
5o Same Direction 10o Same Direction 20o Same Direction Absorbed Energy Loss at 30o
5o Same 10o Same 20o Same
Direction Direction Direction
Moment (Nm)

1324 J 1159 J 657 J
-12 % -23 % -56 %
5o Opposite 10o Opposite 20o Opposite
Direction Direction Direction
1342 J 1245 J 788 J
Angle (degrees) Angle (degrees) Angle (degrees) -11 % -18 % -48 %
Test F.E.M.
5o Opposite Direction 10o Opposite Direction 20o Opposite Direction Conclusions
Moment (Nm)

The loss in energy absorption capability
of rectangular hollow section profiles is
really significant above 10 degrees of
previous plastic deformation.

The methodology proposed to model
the behavior of repaired profiles with
FEM is a good approximation.
Angle (degrees) Angle (degrees) Angle (degrees)
Design, fabrication and test of a short
mechanical model of the orbit corrector
magnets for the High-Luminosity LHC
aJ.A. García-Matos, aP. Abramian, aJ. Calero, aP. Gómez, aJ.L. Gutiérrez,
aL. García-Tabarés, aD. Lopez, aJ. Munilla, aF. Toral, bN. Bourcey,
bP. Fessia, bS. Izquierdo-Bermudez, bJ.C. Pérez, bE. Todesco
aCIEMAT (jesusangel.garcia@ciemat.es), bCERN

Abstract
The LHC will be upgraded in the framework of the High-Luminosity LHC (HL-LHC) project. Among other equipment, the HL-LHC requires the MCBXF
magnets. They consist of two nested superconducting dipoles acting on the horizontal and vertical plane. Their main design challenge is the mechanical
structure necessary to hold the large torque -around 100 kN∙m per unit length- arising when both dipoles are powered. Due to the expected radiation
dose, a mechanical clamping is required. An innovative support structure based on stainless steel nested collars is proposed, achieving the clamping
between inner and outer collars by means of a large key-slot. FEM simulations are carried out to assess the stress distribution at each load scenario. As a
proof of concept to validate the feasibility of the assembly, a short mechanical model of 150 mm has been produced and successfully tested. The results
presented here states that the proposed structure works as expected, enabling its use in the full length prototype.

Introduction To evaluate this proposal, mechanical calculations have been made using
2.5D FEA in ANSYS Workbench®. Every load step from magnet assembly, to
The design and construction of the first MCBXF prototype is developed by cooldown and powering has been simulated, as well as the necessary
CIEMAT in the framework of a collaboration agreement with CERN. MCBXF pressing to close both group of collars. Final design simulations show that
purpose is to introduce small turns in the particle beam in the direction and all the materials are kept below their yield limits at all scenarios.
intensity that is necessary to compensate for possible deviations in the
trajectory. To this end, they consist of two nested superconducting dipoles Fabrication and test
acting on the horizontal and vertical plane. A short mechanical model of 150 mm length using aluminium dummy coils
has been assembled as a proof of concept of the support structure. This
2D Magnetic
field map way not only the assembly but also the FEA simulation models used to
determine coils pre-stress, assembly clearances and materials integrity at
each load scenario could be also assessed and validated.
A whole set of dedicated tooling has been designed and produced to this
end. The nominal gap between the cradles that push the collars together is
0.3 mm for both dipoles and according to FEA simulations, closing this gap
0.1 mm and reaching 0.2 mm should be enough to assemble pins and keys
properly. The pressing force necessary to achieve this should also not be
Coils 3D view
greater than 50 Tm according to simulations.
Mechanical design and simulation Azimuthal pressure on Azimuthal pressure on
inner/outer collar nose at inner/outer collar nose at
The nested stainless steel collars support structure proposed for this pins/keys insertion press release
magnet must hold the Lorentz forces of the individual magnets (mainly Simulation 108 / 110 MPa 74 / 81 MPa
outwards the aperture, perpendicular to the field direction) and also the Test (average) 106 / 124 MPa 72 / 89 MPa
torque due to the combined operation. In order to properly monitor the test, four displacement gauges with
micron accuracy are placed at the cradle end faces, two at the front and
two at the back. Three collars laminations are also instrumented with strain
gauges to monitor the pressure seen by the collar noses. The results
B obtained from the test agree with the simulations at all levels, from the
strain seen by the gauges, to the pressure and displacement needed to
F close the collars. Although further testing is required, these results prove
that the proposed structure works as expected, enabling its use in the full
length prototype. B C
A
Forces and torque scheme over Assembly
displacement distribution (x30) elements
To achieve this, 3 mm thick male and female collars are placed around the
coil, flipped 180º at each layer respect to the previous pair. Collar packs are
obtained, alternating male-female, clamped together with rivets and to the
opposite pack by means of pins (inner dipole) and keys (outer dipole).
Interference between coils and collars noses has been implemented in
order to assure the necessary pre-stress in the coils to avoid quenches.
Thus, it is necessary to press each collar set to get some play with respect A: Assembly clearances. B: Strain gauges at outer collar nose.
C: Setup for outer collar assembly test.
to its nominal position in order to introduce pins and keys and get the final
Conclusions
assembly. The clamping between inner and outer collars is achieved by
means of a large key-slot, close to the pins. HL-LHC orbit correctors are not very high field magnets, but result in a
challenging design, because of the large aperture and nested
configuration. Detailed mechanical analyses have been performed to
validate the current design. The short mechanical model test results
Inner/outer collar
assemblies : validates the feasibility of the assembly and the results of the mechanical
Pressure at collar simulations.
noses (Mpa)
References
J. A. García-Matos, F. Toral and P. Fessia, "Magnetic and Mechanical Design
of the Nested Orbit Correctors for HL-LHC," in IEEE Transactions on Applied
Superconductivity, vol. 26, no. 4, pp. 1-5, June 2016.
Acknowledgments: This work has been partially funded by CERN contract KE-2292 and the Spanish Ministry of Economy, Industry and
Competitiveness under projects ref. FPA2013-47883-C2-2-P and Resolution on May 24th, 2016.
Extreme Value Theory:
Applications for adverse weather conditions
aCristina Sáez, aCarolina García-Martos, aMaría Jesús Sánchez
aTechnical
University of Madrid
(csaezbonales@gmail.com - garcia.martos@upm.es - mariajesus.sanchez@upm.es)

Abstract
Extreme Value Statistics has a long tradition in the field of mathematics, but it was not until the 1950’s that it has emerged as one of the most important
statistical disciplines for the applied sciences. The reason for its success is the enormous variety of applications that demand extreme value estimations in
diverse fields such as natural phenomena (typhoons, floods, air pollution), materials design in engineering, risk analysis for insurance companies, finances
or traffic flow.
We present a comparison of several methods for estimating extreme wind speeds in areas with adverse weather conditions. These methods assume
independent observations, which is an unrealistic assumption in practice, so it is usual to assign a minimum separation time to ensure the independence
of extremes. An innovate method, based on GARCH models that removes the constraint of independence, is currently being drawn up.

1. Motivation and aim 3. Data
For the construction of a wind farm, one of the renewable energies that are 10-minute observations of wind speed at a height of 40 meters from
undergoing greater development, it is crucial to determine the most suitable September 2009 to August 2013 in the province in Cagayan,
wind turbines according to environment conditions as it has important Philippines.
repercussions on the costs of the structure and its safety.

IEC 61400-1 Classification of Turbines

The aim of this work is to provide a good estimation of reference wind speed
that allows to decide the wind turbine class that best suits our needs.

2. Methodology
 Method r-Block Maxima (r-BM)
Divide data in blocks, select r-extremes in each block and fit Generalized
Extreme Value distribution (GEV).
 Method Peaks-Over-Threshold (POT) Strong
gale on
Select values above a certain threshold and fit Generalized Pareto distribution. Beaufort
Scale
 Method Independent Storms (MIS)
Identify storms independent of each other, select maximum value for each
storm and fit GEV distribution.
 GARCH approach: Ongoing project
Fit a GARCH-type model to all the values making no assumption about its
distribution and model its tail using method POT.

4. Results
Method r-BM Method POT Method MIS

Pre-parameters: r=15 val./block; Min. sep=60h Pre-parameters: Threshold=24m/s; Min. sep=60h Pre-parameters: Minimum threshold = 22m/s
Data: 60 values  13 independent values Data: 43 values  4 independent values Data: 54 independent values
Reference Speed: 31.85 m/s  Turbine Type III Reference Speed: 43.53 m/s  Turbine Type II Reference Speed: 33.64 m/s  Turbine Type III
Confidence Intervals Bootstrap: [18.6, 37.4] Confidence Intervals Bootstrap: [29.5, 45.2] Confidence Intervals Bootstrap: [30.2, 37.1]
Pros: Easy to implement. Pros: More effective use of extreme values. Pros: Effective use independent values
Cons: r-maximum value need not be an extreme. Cons: Pre-set threshold. Have to ensure Cons: Have to ensure independence 
independence. NEXT STEP: GARCH approach
Local forecasting of electric demand: Madrid

aMiguel Carretero, aEduardo Caro, aJesús Juan, aJavier Cara
aLaboratory of
Statistics - Technical University of Madrid
(eduardo.caro@upm.es)

Abstract
In this work, the short-term electric load forecasting method that is currently used for the Spanish mainland, is implemented for a local zone. A
regression-spline model has been designed to implement the relationship between temperature and energy demand, including a different model: local
temperatures, inclusion of minimum daily temperatures, and the interaction with working and non-working days.
To check the method's performance, a real data set from the Spanish electricity system has been used and the developed algorithm has been tested
employing the MSE as the performance metric, analyzing the electric zone of Madrid.

Motivation and aim
For any transmission system operator, it is important to have a proper
prediction for the short-term future consumption. This is particularly
valuable in the electric power framework: since the electricity cannot be
efficiently stored, the electricity system operator must meet generation
and demand on a real-time basis.
The aim of this work is to enhance the quality of the short-term load
forecasts applied to a local zone.

Temperature model
The non-linear relationship between temperature and electric load has
been modeled using a spline-regression model Wahba (1990) and Wood
(2006):

where

Hourly mean squared error for several models tested, leading an average MSE of 4.8.
In this model, the maximum and minimum daily temperatures have
been considered. Additionally, the regressors of temperature have been
duplicated, in order to consider the effect of temperature over working
and non-working days, respectively.

Non-working days model
In order to model properly the relationship between holidays and the Madrid’s electric
load, it has been included a set of regressors using dummy variables. For each public
holidays, a set of 0/1 variables is included, considering the day of the week and the
following/previous days. This model leads to a set of around 300 variables.

Results
The implementation of the forecasting model to an smaller electric zone has proved to be
accurate and reliable.
As an illustrative example, the figure below provides the forecasted (green line) and the
actual load demand (blue line), denoting accurate results.

Effect of January 6th over Madrid’s demand.
Real and forecasted electric load for a week using the developed model.
Advanced variance reduction methods in
Monte Carlo calculation
AM.Magán, F. Sordo, bJ.M.Perlado
AESS-Bilbao/Instituto de Fusión Nuclear (mmagan@essbilbao.org)
BInstituto de Fusión Nuclear

Abstract
Neutron particle transport problems, often feature a several orders of magnitude range of variance in the result. When using Monte Carlo methods, this
results in a situation where many millions, or even billions of histories must be run in order to obtain just a few tracks in certain parts of the problems.
Variance reduction methods aim to solve this problem in a number of ways. In particular weight windows cause particles to split in several tracks with
lower weight, in order to keep track density more homogeneous. MCNP in particular features a weight windows generator, but it is severely limited,
specially when trying to improve the global solution rather than a particular area. Thus, we have implemented an in-house weight window generator
featuring far more control, and power, and ease of use thanks to modern programming language.

Global Variance Reduction in MCNP
Concept
In an analog Monte Carlo calculation, the density of track, directly correlated to uncertainty, mimics the flux density. This is usually highly inconvenient, so
GVR (Global Variance Reduction) aims at "flattening said track density. Thus, a good initial approach to reduce the disparity is decrease the
weight proportionally to flux. An exponential transform can be used to adjust how agressive we wish to be in GVR.

Implementation
Using mesh tallies in MCNP, we can get a volume distribution of the flux in, either in cartesian or cylindrical coordinates with an analog (i.e: No GVR)
run, and use that to generate a first window. That window can then be used to get a solution that reaches farther, and iterate until we cover the
entire geometry.

Power made easy: The mesh tally class.
MCNP uses fortran and ancient programming, making for inputs that are virtually human-unreadable. However, we are not bound to its tools. By using
object oriented programming, we have implemented a mesh tally class.
• Easy input/output.
• Tally modification: Merging, energy and space collapsing possible.
• Possibility to use geometrical masks to customize the GVR, i.e. define high interest zones/energies.
• Artificial mesh tallies (guesstimates, deterministics calculations, etc) can also be transformed into mesh tallies and used.

Tried & True: Benchmarked for ITER IO
The method has been used in two benchmarks: One simple geometry, conceptual exercise, and another complex problem featuring a full-blown model
of the ITER tokamak.

Out in the wild: Applied in the ESS and ITER projects
Currently this method is being used in two of the most important experimental nuclear facilities in the world: the European Spallation Source and the
ITER project, with excellent results.
Monte Carlo study of detection systems of explosives
and illicit substances using a D-D neutron generator

Lenin Cevallos, Eduardo Gallego, Karen A. Guzmán-García, Gonzalo García
Departamento de Ingeniería Energética, ETSI Industriales, UPM
Contact: lenin.Cevallos.robalino@alumnos.upm.es
Abstract
Prompt gamma neutron activation analysis (PGNAA) using a Deuterium-Deuterium (D–D) neutron generator, is one of the most promising method to detect
explosives and illicit substances. Using Monte Carlo methods with MCNP6 code, a model of a DD Generator (DD-110 Adelphi Technology [1]) has been built
with one or two NaI(Tl) detectors of 1.5” x 1” size in order to determine an optimal configuration. Several calculations have been performed to analyse the
detection power for samples of different explosives: RDX, Ammonium Nitrate, land mines, military and homemade explosives, have been selected. Different
models have been made varying moderator and shielding thicknesses (polyethylene and lead) and we conclude that the configuration with 10 cm thickness
of polyethylene, is the best configuration to obtain a neutron thermal flux appropriate for detection of these simulated samples of explosives.
Introduction
D-D Neutron generator: Fusion of light nuclei is the most common approach used with particle accelerators to generate neutrons; the most common of
these reactions used for producing neutron are listed in the Table 1. While other reactions also exist and have been used with accelerators these are the
ones that have been found to be the most useful, the easiest to employ, and the most efficient for the purpose [2]. Polyethylene ((CH2)n), is an effective
neutron shielding material for neutrons and displays excellent attenuation behavior because of its hydrogen content (14% by weight) and its density. In
polyethylene, the hydrogen capture of thermal neutrons is through the 1H(n, γ)2H reaction which has a cross section of 0.33 barn for neutrons in thermal
equilibrium at room temperature (En = 0.027 eV) [3].
Materials and Methods
Is in development a system of detection based through the use of the reaction Deuterium-Deuterium (D-D), the when generate neutrons in the reaction,
through these system has been simulation with the use of the code Monte Carlo, the process is the detection of explosives, identificating the elementals
compounds of the sample in the reaction of the thermic neutrons, as the issue gamma fast of 10.82 MeV in the nitrogen [4]. The main problem, is the
differentiation of the materials elements considered parasites, in the identification of the amount of hidden explosives [5].

Q Value Minimum Product
Reaction
[MeV]
+3.269
Energies [MeV] Results
Neutron
+17.589 energy Neutron flux
(MeV) (n cm-2 s-1)
-1.644
2,50E-02 1,33E+05
+15.031 1,00E+00 3,97E+05
-1.850 2,50E+00 1,33E+05
TOTAL 6,62E+05
+4.361
Table2.- Calculated neutron Figure 5.- Neutron dose equivalent rate (µSv/h) due to the
flux at the sample position explosive detection system at points 0.5 1.0 1.5 and 2.0 m,
Table1: 1.-
Table Common
Common nuclearnuclear
reactionsreactions used touse Figure 1.- MCNP geometry of the explosives
particle accelerators
to produce neutrons
produce neutrons. [1]. in particle accelerators detection system based on a D-D neutron for a 1,0E+11 n/s neutron corresponding to figure 2.
generator and 1.5” x 1 NaI detector generator output.

Figure 6.- Gamma ray spectrum from an activated Nitrate
Ammonium sample through MCNP6.
Figure 4.- Neutron flux as a function of distance
from neutron generator (cm)
Conclusions
An explosive detection system based on a D-D neutron generator
The determination of the has been conceptually designed incorporating reliable and effective
Figure 2.- Scheme neutron dose equivalent rate (µSv/h) due to the H, C, N and O in samples radiation shielding of the gamma-ray detector using MCNP6. The
explosive detection system at points 0.5 1.0 1.5 and 2.0 m,
corresponding to figure 1 configuration.
through of use of a detection of explosives using thermal neutrons is primarily based on
source of neutrons, the identification of nitrogen through its 10.82 MeV capture ɣ-ray.
provide an invaluable The 2.5 MeV neutrons produced from the D-D neutron generator
tool in the identification with an output of 1.0E+11 n/s could be used to detect explosives
of substances in samples with an NaI detector.
and baggage. A neutron
References
beam interacts with the [1] http://www.adelphitech.com/products/dd108.html
[2] Chichester David L., Idaho National Laboratory, INL/EXT-09-17312, page. 1-81, (2009).
sample and the amount [3] Yasin Tariq, et al, e-Polymers, Vol. 059, ISSN 1618-7229, page 1-7, (2008).
of H, C, N and O. [6] can [4] Bergaoui K., et al, Applied Radiation and Isotopes, ELSEVIER, Vol. 94, page 118 – 124,
be determined through (2014).
[5] Kuznetsov A.V., ., et al, Applied Radiation and Isotopes, ELSEVIER, Vol. 61, page 51-57
the emission of gamma (2004).
rays and their energy [6] Gill Miller T. et al, Proceedings of the SPIE, Vol. 2511, page 14-24 (1995).
Figure 3.- Gamma ray spectrum from an activated RDX sample spectra. Acknowledgement. This PhD research is being made possible thanks to a grant
through MCNP6. SENESCYT (Secretaria Nacional de Educación Superior Ciencia y Tecnología) of Ecuador.
Fukishima Daiichi Unit 2 Accident
Analysis with MELCOR 2.1
a R. Bocanegra, b V. Di Marcello, b V. Sánchez-Espinoza, a G. Jiménez
a Grupo de Ciencia y Tecnología de sistemas avanzados de fisión nuclear, ETSII UPM
(r.bocanegra@upm.es)
b Karlsruher Institut für Technologie (KIT)

Abstract
A Fukushima Daiichi Unit 3 MELCOR model from Technical Research Centre of Finland (VTT) was modified to simulate the Fukushima Daiichi Unit 2
accident. Several simulations were performed using three different modeling approaches. The base model (Case 0) includes only the basic modifications to
reproduce the accident. The intermediate model (Case 1) includes an improved alternative injection curve. In the third model (Case 2), the Reactor Core
Isolation Cooling (RCIC) system fail is modified. Results are compared against Tokyo Electric Power Company (TEPCO) measured data obtained during the
accident. It is concluded that there is a strong dependency on parameters which still have many uncertainties, such as the RCIC two-phase flow operation,
the alternative water injection and the suppression pool failure criteria, which affect the reactor final state.

RPV Pressure
8
Pressure (MPa)

6

4
TEPCO
2 Case 0
Case 1
0 Case 2
0 20 40 60 80 100
Time (h)

RPV Liquid Level Reactor Core Final State
8
6
4
2
Level (m)

TAF
0 BAF
-2 TEPCO
-4 Case 0
-6 Case 1
-8 Case 2
Case 0 Case 1 Case 2
60 70 80 90 100 No core melting Partial core melting Partial core melting
Time (h) No RPV failure No RPV failure RPV failure at 96 h

Alternative Water Injection Hydrogen Generated
25 800
Case 0 Case 0
Case 1 Case 1
20
Case 2
Hydrogen (kg)
Mass Flow (kg/s)

600 Case 2
15
400
10

5 200

0
0
70 75 80 85 90 95 100
70 80 90 100
Time (h) Time (h)

PCV Pressure CONCLUSIONS
1600
TEPCO
Case 0 • Pressure agree with the RPV measured data.
1200
Pressure (kPa)

Case 1 • Containment still have not been reproduced
Case 2
800 correctly.
• The results show a strong dependency on:
400
RCIC two-phase flow operation
0 Alternative water injection
0 20 40 60 80 100
Containment integrity failure criteria
Time (h)
Towards the definition of a local damage
criteria for containment safety analysis
Kevin Fernández-Cosials, Rafael Bocanegra, Samanta Estévez-Albuja, Zuriñe Goñi,
Gonzalo Jiménez
Grupo de Ciencia y Tecnología de Sistemas Avanzados de Fisión Nuclear (ETSII, UPM)
(kevin.fcosials@upm.es)

Abstract
Nuclear Power Plant (NPP) containment analysis has been done historically under the lumper parameters approach. This was useful when the first NPPs
where designed back in the late 60s. Those analyses provide an average pressure and temperature the containment must withstand during a Design Basis
Accident (DBA). However, recent analyses have been proven the previous ones not only inaccurate but also useless in terms of temperature. Recently, these
analyses have been remade in the UPM Safety Group with the current computational capabilities and with 3D models of the containment building. As a
consequence, a new type of damage have to be defined for local surpassing of the equipment temperature limits. As a result of these works, the local
damage criteria has been developed. It is may be time to reconsider this criteria for future design of NPPs.

3D GOTHIC Model
Model in GOTHIC
The thermal-hydraulic code GOTHIC has been used to create a 3D
containment model. The model has been developed following three
steps. Firstly, building a detailed CAD 3D model. Then, modifying this
CAD model to obtain a simplified CAD 3D model, and finally,
developing the GOTHIC 3D model.
Definition of a new Local Damage Criteria
A new definition of damage for containment analysis is proposed
based on the qualification criteria for safety equipment and
3D CAD
instrumentation and is based on the temperature limitation (148.9 °C)
detailed model
in the equipment and qualification criteria.
3D CAD Mass and Energy Release
simplified model The postulated accident consist of a DEGB LOCA in the RCS cold
leg. The M&E release data, used as input, was obtained from a
PWR-W MELCOR model developed at UPM. The NPP taken as reference is a 3
GOTHIC 3D Model loop PWR-W with a power generation of 2696 MWth.

Simulation Results Room max temperature evolution
180
Post-processing software 160 148.8 °C Max temp. limit

ProTON 140
Temperature [°C]

120
Base case results 100

The averaged values of temperature 80
(114 °C) are under the limits stipulated by 60
the classical criteria, 148.9 °C. 40
The temperature evolution in each room 1 10 100 1000
Time [s]
does not reach homogeneity at any
moment, and nearly all rooms have 180 14
Max Temperature Reached [°C]

Max Temp Duration
different temperatures. 175 12
Time above the limit [s]

170
The hot spot is located in Room 5, where 10
165
the break is located. The high vapor
160 8
temperature moves through Room 5 to the
155 6
adjacent rooms, as well as the dome.
150
In certain rooms the max temperature 4
145
peak exceeds the temperature limits. Under 140 2
the Local Damage Criteria (LCD) this Best 135 0
Estimate simulation would exceed the Room Room Room Room Room Room Room Room
generic equipment qualifying limit in 8 5 23 24 39 28 6 7 25
rooms.

Conclusions
A new damage criteria has been defined and proposed based on the qualification criteria for safety equipment and instrumentation setting the
maximum temperature locally allowed in 148.8 °C with an 8 °C margin. Subsequently, this criteria has been applied to a DEGB LOCA in a PWR-W 3D
containment model developed with GOTHIC 8.1 (QA), showing that the LDC is surpassed.
From the results obtained, several proposal are released. The first one is that LDC should be used to evaluate the potential damage for the
containment equipment and instrumentation during an accident. The second one is that the qualification basis for equipment and instrumentation
should be made taking into account the results from local values, because the temperature heterogeneity in the containment rooms makes invalid
the average values.
Nuclear Data for the Safety of Lead-based
Cooled Facilities
aP. Romojaro, aN. García-Herranz, bF. Álvarez-Velarde
aDepartamento de Ingeniería Energética, ETSII-UPM (p.romojaro.otero@gmail.com)
bDivisión de Fisión Nuclear, CIEMAT

Abstract
The Sustainable Nuclear Energy Technology Platform has selected the Lead-cooled Fast Reactor as one of the three technologies that can meet future
European energy needs. In nuclear reactor design the technological uncertainties mainly come from material properties, fabrication tolerances, operative
conditions, simulation tools and nuclear data. The uncertainty in nuclear data is one of the most important sources of uncertainty in reactor physics
simulations, however, significant gaps between the current uncertainties and the target accuracies have been systematically shown [1]. The objective of
this work is to improve the quality of the nuclear data required for the design, operation and licensing of lead-cooled fast reactors, in particular MYRRHA ,
and for the operation of any nuclear facility using lead as one of its main materials (e.g., medical facilities).

MYRRHA Accelerator
600 MeV – 4 mA proton
Reactor
Subcritical mode (60-70 MWth)
Critical mode (100 MWth)
The flexible experimental Accelerator Driven System (ADS) MYRRHA (Multi-
purpose hYbrid Research Reactor for High-tech Applications) is being
developed at SCK•CEN, Mol, Belgium [2]. MYRRHA will fulfil the role of
European Technology Pilot Plant in the roadmap for the development of the
lead fast reactor. It is expected to be fully operational by 2025. MYRRHA is
Spallation Source
conceived to operate both in critical mode and sub-critical mode as an ADS
(Figure 1). Designed as pool type facility, it takes the advantages of lead-
bismuth eutectic both as intense neutron producer via proton induced Multipurpose
Fast
spallation reactions and effective coolant providing a fast neutron spectrum. Flexible
Neutron
Irradiation
It will be used for research on applications such as long-lived radioactive Facility
Source
waste transmutation, development of materials for fast spectrum reactors Lead-Bismuth
and fusion technology, production of neutron irradiated silicon, production coolant

of radioisotopes for nuclear medicine and fundamental research.
Figure 1. MYRRHA: an Accelerator Driven System
Uncertainty quantification (UQ)
Sensitivity and uncertainty analyses were carried out in order to estimate the uncertainty in the criticality safety parameters and to identify the most
important nuclear data for neutron induced reactions of MYRRHA and of the Advanced Lead Fast Reactor European Demonstrator (ALFRED) [3], a GEN-IV
fast reactor cooled by lead which will demonstrate the technology. The analyses were carried out with the SCALE6.1 system [4] and the SUMMON
methodology, which has been developed in the framework of this work. A summary of the results is presented in Tables 1 - 4.
Table 1. UQ MYRRHA nominal conditions Table 2. UQ ALFRED nominal conditions Table 3. UQ ALFRED COOL +20% Table 4. UQ ALFRED TFUEL +300 K

Δk/k (%) Δk/k (%) Δ/ (%) Δ/ (%)
Quantity Quantity Quantity Quantity
SCALE6.1 SUMMON SCALE6.1 SUMMON SCALE6.1 SUMMON SCALE6.1 SUMMON
Pu 
239
0.705 0.702 239
Pu  0.703 0.713 238
U(n,n’) 37.7 13.8 238
U(n,n’) 8.53 10.4
238
U(n,n') 0.319 0.332 238
U(n,n') 0.533 0.534 206
Pb(n,n') 13.8 14.4 239
Pu(n,n') 1.53 11.9
239 207 207
Pu(n,) 0.270 0.268 239
Pu(n,) 0.266 0.262 Pb(n,n’) 18.2 11.5 Pb(n,n’) 1.18 0.83
240
Pu  0.235 0.234 240
Pu  0.243 0.245 Pu 
239 13.3 8.77 Pu 
239 1.14 1.80
239
Pu(n,f) 0.188 0.188 239
Pu(n,f) 0.213 0.166 204
Pb(n,n) 1.21 24.3 238
U(n,) 1.09 0.56

Nuclear data improvements
State-of-the-art nuclear data evaluations for lead were analyzed, focusing on the beta proposal of the Joint Evaluated Fission and Fusion (JEFF) library.
Inconsistencies (Figure 2) were found and reported to the JEFF collaboration [5]. The proposed corrections [6] have been adopted in the latest version of
the library (Figure 3).

Figure 2. 207Pb(n,tot) JEFF-3.3T1 compared against experimental data Figure 3. 207Pb(n,tot) JEFF-3.3T1 and JEFF-3.3T2 comparison

As part of this work, new transmission experiments will be performed at the JRC-Geel GELINA facility to validate Pb nuclear data in the resonance region.
References
[1] H. Harada and A. Plompen. NEA/NSC/WPEC/DOC(2014)446 i [4] ORNL/TM-2005/39 (2011)
[2] J. Engelen et al. Int. J. Hydrogen Energy 40 (2014) 15137–15147 [5] P. Romojaro and F. Álvarez-Velarde. NEA JEF/DOC-1756 (2016)
[3] G. Grasso and L. Mansani. Deliverable D6.1.1-3, EC FP7 ESNII+ Project (2014) [6] P. Romojaro et al. D10.3, EC FP7 CHANDA Project (2016)
Numerical simulation of shaking table tests on a
reinforced concrete waffle-flat plate structure
aD. Galé-Lamuela, aA. Benavent-Climent
aDepartment of Mechanical Engineering, Polytechnic University of Madrid, Spain
(david.gale@upm.es and amadeo.benavent@upm.es)

Abstract
A numerical model is developed that represents a 2/5 scale reinforced concrete waffle-flat plate structure tested on a shaking table. The numerical model
was built using Engineer’s Studio, Japanese FEM software which implements Maekawa’s approach for simulating reinforced concrete elements. The model
uses rigid, spring, frame and layer-shell nonlinear elements with their constitutive laws calibrated with material tests. The model was firstly subjected to
nonlinear static (pushover) analysis and then to non-linear dynamic time history analysis. The former allowed an overall check of the strength-
displacement relationship. The latter consisted on subjecting the numerical model to the same (four) acceleration records applied to the test specimen
during shaking table tests conducted by the authors in previous studies. The acceleration records reproduced the ground motions recorded during the
Campano-Luchano Earthquake at Calitri, scaled to four levels ranging from 100% to 350% times the acceleration of the original record. The numerical
model reproduces accurately the experimental response in terms of maximum displacements, forces and history of energy input. The correspondence in
terms of history of displacements is also satisfactory.

Specimen for testing Numerical Model

Discussion and results
Developing a numerical model which is able to simulate reliably the strongly nonlinear behavior of a real
structure tested on a shaking table test up to collapse is very cumbersome. The difficulty increases in case
of systems such as waffle-flat plate structures which behavior near collapse is controlled by the torsional
failure of the spandrel beams located in the exterior plate-column connections. Conventional elements do
not cover all types of nonlinear behavior and failure mechanisms (flexure, shear, torsion, etc). In particular,
most frame elements cannot represent the torsional response in the non linear range (only elastic linear).
To fix this problem, no linear torsional spring elements were introduced in the model at the locations
prone to experience torsional failure. Overtaking the problems mentioned above, the authors succeeded
in developing a numerical model that is able to match very well the history of energy input by the ground
motion the structure. This input energy is the work done by the inertial forces that develop in the mases
on their relative displacement to the ground. Fitting accurately the history of input energy is essential to
understand what happened in the tested structure with the information provided by the numerical model.
Besides, the model is able to capture in a satisfactory was the displacements and accelerations at the top
of the structure, the base share force, and the relationship between top displacements and base shear
force. Once the everything is matching, the following step will be to obtain from the numerical model
detailed information that cannot be obtained form the tests, such as the dissipated energy in plate-column
connections, and to perform parametric studies.
Seismic upgrading of frame structures with a soft
first story by using hysteretic dampers
a Santiago
Felix Mota Paez*, a Amadeo Benavent Climent
aDepartment of Structural Mechanics, Polytechnic University of Madrid, Spain

(sf.mota@alumnos.upm.es, amadeo.benavent@upm.es)

Abstract
Soft stories constitute a major reason for building collapses during severe earthquakes. The present study investigates a solution for the seismic upgrade of
existing reinforced (RC) concrete frames with open first story and masonry infills at the upper stories. It combines the addition of hysteretic dampers with
the strengthening of the columns of the first story. The energy-based design procedure put forth provides the strength, stiffness and energy dissipation
capacity required for the dampers so that the overall structure can endure the design earthquake without exceeding a prescribed maximum drift at the
first story. Finally, using the code for Inelastic Dynamic Analysis of Structures (IDARC-2D), non-linear dynamic analyses are carried out to evaluate the
seismic performance of the retrofitted structures and to validate the proposed procedure, the results obtained show that the proposed solution is feasible.

Introduction and Objective
Frame structures with open first stories and masonry infills
at the upper stories are common in earthquake-prone areas
all around the world. It is a very attractive solution from the
architectural point of view because it allows for large
openings to harbor commercial activities, or space for
parking. However, it introduces a sudden discontinuity in the
lateral strength and stiffness along the height that can result
in severe damage concentration. Due to this seismic
deficiency many buildings with soft ground story collapsed
in past earthquakes (e.g., Northridge-1994, Hyogoken-
Nanbu-1995, etc.). Figure 1 shows the schematic solution Figure 3: Hysteretic rule of the
proposed in this work. The main objective of this study is to plastic hinges.
Figure 1: Proposed solution schema, steel
achieve a structure able to dissipate most of the energy
hysteretic damper and FRP/SRP.
input by the earthquake in the first story, approaching the
behavior of the base isolation systems.

Methodology
To achieve the main goal, several numerical models representing existing RC frames
buildings with 3 (N3), 6 (N6) and 9 (N9) stories (Figure 2) where developed using IDARC- Columna
2D code. Beams and columns are modeled as Disipador
2-node frame elements with plastic hinges
FRP/SRP
that concentrate the non-linear behavior at both ends. A graphical representation of the
plastic hinge hysteretic rule is shown in Figure 3. de
Llave Next, a simple method to design the
cortante
proposed seismic upgrading solution is outlined. The method is based on the theory of Figure 2: 3D view of prototype buildings
energy balance of Housner-Akiyama. Finally, the three models developed were subjected Cimiento Morgan Hill Double Springs-EW
to thirty natural accelerograms correspondingResina
to far epoxica
field earthquakes. Their 5% damped Tabas Double Springs-NS
Placa final del disipador Friuli Dursunbey
elastic response acceleration spectra are shown in Figure 4. The ground motions are Northridge Kozani
2
selected from PEER and LIS ground motion database respectively. 3000
Sa (cm/s ) Kobe
Victoria
Mammoth lake
Manjil
Chalfant Valley Spitak

Results 2500 Loma Prieta
N. Palm Springs
Alajuella PCC-NS
Alajuela CH-NS
2000 Whittier Narrows Alajuela-EW
Figures 5-7 show the response of the 3, 6 and 9 story prototypes respectively, in terms of Coalinga Alajuela UCRSR-NS
the maximum inter-story drift (????,? ), expressed as ratio of the maximum inter-story 1500 Gazli USSR Alajuela UCRSR-EW
Nahanni Cobano CA
drift at story ? to the corresponding story height (?????, ?). It can be seen in Figures 5-7 1000
Sierra Madre Pejibaye
that the ?????, ? remain within acceptable limits. Within these limits, the damage in Erzincan
Dinar
Mean

the RC columns is negligible. The cumulative distribution function (CDF) was obtained by 500
T (s)
assuming a log-normal distribution of the maximum inter-story drift ratio. It can be seen 0
0.0 0.5 1.0 1.5 2.0 2.5
that the probability of ????? at the first story being smaller than the design value is
about 91%, 81% and 67% for prototype N3, N6 and N9 respectively. Figure 4: Elastic response spectra of non-impulsive ground motions

3 Story 6 9 Story
Story 8
Design limit 1st story Double Springs5
7
Mean-SD Dinar
Mean+SD Erzincan 4 6
y 2 Mean Sierra Madre 5
Pejibaye Nahanni 3
Cobano CA Gazli USSR
4
Alajuela UCRSR-EW Coalinga 3
Alajuela UCRSR-NS 2
Whittier Narrows
MISDR (%) 2
Alajuela CH-EWMISDR (%)N. Palm Springs MISDR (%)
1 Alajuela CH-NS Loma Prieta 1 1
0.0 0.1 0.2 0.3 Alajuela 0.5 0.6 Chalfant Valley0.0
0.4 PCC-NS 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Figure 5: MISDR for 3Spitak
story prototype Victoria Figure 6: MISDR for 6 story prototype Figure 7: MISDR for 9 story prototype
Manjil Kobe
Conclusions Mammoth lake
Kozani
Northridge
Friuli
This investigation proposed a new retrofitting
Dursunbey Tabassolution and a method for designing and checking the
main frame, that protects satisfactorily the existing
structure and preventsDouble
damageSpringsunder severe
Morganearthquakes
Hill in both, first story and upper stories.
The proposed design method allows for optimal
combination of different seismic upgrading strategies (traditional with composite materials and innovative with energy dissipators).
SEISMIC DESIGN OF STRUCTURES WITH SEISMIC
ISOLATION BASED ON ROCKING COLUMNS

aGermán Ríos García, aAmadeo Benavent
a Department of Mechanical Engineering, Polytechnic University of Madrid, Spain
(german.rios.garcia@alumnos.upm.es and amadeo.benavent@upm.es)

Abstract
The use of “Rocking Columns” in order to implement the base seismic isolation of structures is becoming the object of recent research works and the focus
of increasing interest in technical publications. Most of the research works presented up to date have usually dealt with the design of reinforced concrete
structures that allow such rocking movement by means of interruption of the continuity of the longitudinal reinforcement at the rocking connection
located at the top and bottom of the columns that will be allowed to rock, and also accepting the local yielding and spalling of the concrete at the
perimeters of the contact section. There have also been some real applications in seismic isolation of new or retrofitted bridges, and also in studying the
behavior of planar frames of one-story historic buildings formed by an array of freestanding columns capped with a freely supported rigid beam. However,
there is almost a total absence of research works focused on the study of this type of isolating system applied to multi-story buildings. The research work
presented here has the object of evaluate the feasibility of using this type of seismic isolating system by means of a proposed particular system of «rocking
columns» located at the lower story of the building, developed with the possibility of customizing the backbone curve that defines the behavior of the
rocking connection in order to fulfill the requirements of a performance-based design of the frame. This is proposed as an alternative to the traditional
seismic isolating systems performed with bearings, evaluating the advantages and disadvantages with respect to them

Proposed Seismic Isolation System with Rocking Columns at the lower story
In the next figure are shown the layout of a conventional base seismic isolation system with bearings, and of the proposed system with “Rocking Columns”
replacing all the columns at the lower storey of the building located over the ground. With respect to the former, the later has, among other advantages :
• The necessity of and additional storey of beams totally devoted to the bracing of the bearings, and of a perimeter horizontal joint, are avoided.
• The frame behaves as rigid, like a conventional frame, towards the forces originated by the vertical gravitational loads, and by the horizontal loads like
the wind loads, up to the value of the horizontal force that initiates the “rocking” movements in those columns. In such way, a greater monolithic
behaviour is achieved, allowing a better performance of the frame with respect to the serviceability limit states.

The proposed Rocking Connection with customizable Backbone Curve
Since the first research work, it was shown that in order to get a
behaviour of the frame appropriate to fulfill the requirements of a
performance-based design :
• It was necessary to get a Backbone Curve describing the behaviour of
the Rocking Connections located at the top an bottom extremes of
the Rocking Column, showing an elastic behavior along most of the
range of rotation movement, and with intervals of negative or positive
stiffness depending upon the case, as shown in the adjacent figure.
• This desired Backbone Curve could be achieved with an unbonded
strand located at the axis of the column, anchored at the top and
bottom beams of the rocking column, and installing a “gap” device in
one of the two anchorages that allows the beginning of the loading of
the strand only after the rocking connection has opened and has
travelled through a certain angle of aperture.
• In such way, it is possible to get a pure “rocking” movement with
negative stiffness, necessary to get de desired seismic isolating,
Future work
during the first interval after the aperture of the rocking connection, For a practical use of the proposed solution, it would be useful to develop a
and after the closing of the “gap”, to get an interval with positive column unit, totally or partially prefabricated, integrating in it the possibly
stiffness, necessary to avoid the overturning of the entire frame at necessary energy dissipation system, performed with metal yielding devices,
extreme seismic events. The desired Backbone Curve can be obtained or deformation of viscous fluid or solid devices. In such way, it would be
adjusting the parameters of the elements that constitute the Rocking possible to avoid setting the necessary diagonal bracing bars containing these
Connection devices, eliminating this servitude in the use of the lower storey.
Algorithms to aid operational management in
urban transport systems.
a Daniel Herrero Giner, bÁlvaro García, cMiguel Ortega
a INGOR (e-mail: dani_herrero@yahoo.es)
b INGOR (e-mail: alvaro.garcia@upm.es)
c INGOR (e-mail: miguel.ortega.mier@upm.es)

Abstract
Long-term crew scheduling on underground and railway passenger transport has been widely studied in recent years. However, these approaches are not
efficient enough for companies which require crew scheduling on daily basis for operational decisions. We present a hybrid constraint programming and
column generation approach for obtaining optimal or near-optimal solutions in daily planning for the constrained crew scheduling problem in
underground transportation. The algorithm presented has been implemented for Transports Metropolitans de Barcelona (TMB), Barcelona, Spain.

Problem description
• The problem concerns in obtaining a feasible assignment of tasks
for each employee, with the objective of covering as many tasks as
possible considering that all tasks could not be equally important.
• Tasks can require different employees in terms of qualifications.
• Employees have a start time and an end time, which are hard
constraints.
• Feasible schedules for each employee are subject to different
constraints as maximum driving time, interconnection time
between different trains, minimum rest in journey...
• Given the operational character of the problem, the number of
available employees for each day is known, so the objective is not
to minimize the number of employees but obtaining a feasible and
good quality schedule for all of them considering the optimality
Algorithm description criteria described above.

1. Clustering: 2. Services generation 3. Services assigment
The first step of the algorithm, consists of The next step is to obtain all feasible
classifying the profiles in sets (clusters), schedules for each profile. A feasible
according to their start time and end The objective of this module is to
schedule can be defined as a combination of
time. Each cluster is feature by a start obtain the best combination of
tasks which can be developed by an
time and an end time, and the schedules in order to cover as many
employee in the profile. This assumption
intersection between clusters does not tasks as possible.
implies that:
have to be empty. The requirement for It considers the uncoverage
• All employees in the profile can develop
one profile to belong a cluster, is that its preferences and the limited employees
one task after the other in the order
journey is contained in the cluster period. introduced as input data.
proposed, and under the timing defined.
According to this classification, all their • The schedule satisfies all the constraints
feasible schedules are a subset of the proposed.
feasible schedules of the cluster.
Development of an energy management
embedded controller for a range-extended hybrid
vehicle using model-based design and HILs
aResino, D., bMarcos, D., cLópez, J.M., dFernández, D.

a,b,c,dUniversity Institute for Automobile Research – UPM, (daniel.resino@upm.es)

Abstract
At present, onboard electronics are the heart of a modern vehicle. The complexity of network architectures and in-vehicle functions grow in every new
model. It is possible to find over 100 million lines of code within a modern high-end car, more than an F-35 jet fighter or a Boeing 787.
This huge amount of code is one of the most important source of errors within a car. Automotive industry is very concerned about safety risks that all
these challenges implY. That is why OEMs joined in the AUTOSAR development partnership in 2003. And that is the necessity behind ISO26262 standard as
well. Meeting all requirements to comply with ISO26262 requires an in-depth analysis to align all involved processes of the company with that objective.
In this paper, a description of one of the steps involved in these matters is explained: the creation of a control algorithm for energy management and its
HIL functional testing. All work is based on a vehicle plant model build with AVL Cruise® software; the algorithm stage has been developed using Simulink-
Stateflow® and Vehicle Network Toolbox® has also been used to run functional testing of real hardware.

Methodology
As a system engineering work, this process has been carried out following the V diagram methodology. Within
this framework, a model-based design method has been used . From real vehicle testing, a plant model has
been developed and validated with AVL Cruise® software. This model has served as a base to implement an
energy management algorithm developed and co-simulated with Simulink Stateflow®. With this approach,
strategy software can be implemented during a very early stage of design process, reducing the number of not
admissible bugs at prototype level. Once co-simulation platform is ready, it is possible to carry out software-in-
the-loop testing. At the last step of the process, a CAN interface is connected to real ECU and Vehicle Network
Toolbox® from Simulink® is also used to arrange a CAN network for Hardware-in-the-loop functional testing.

Vehicle plant model
In this case, a 4WD vehicle model was developed using AVL Cruise® software. The ECU that will be
tested has been extracted from the model and replaced by an interface block that enables to send
and receive exactly the same variables that real hardware to Simulink®.
Software-in-the-loop simulations
With the objective of evaluating the performance of different versions of control algorithms, a SIL
model has been used. Vehicle-related variables are fed from Cruise model to the Stateflow
control; manual driver signals are fed by a signal builder block. Stateflow control returns to Drive
Torque, Start signals to engine and generator, engine speed and generator torque to the vehicle
plant. Driving cycle is still run by Cruise model, as well as performance and consumption
powertrain results.
Code generation
Once a version of control algorithm is validated to be tested, it must be implemented into a real controller. For this work, a widely used in powertrain
applications industrial controller has been selected and a C code has been generated from Stateflow strategy control. From Stateflow, Simulink Coder has
also been used for code generation, because it automatic generation is compulsory to comply qith automotive ISO26262 quality standard.

Hardware-in-the-loop functional testing
A HIL platform has been developed; as described above, a Peak Systems USB Pro CAN
Interface is used to connect PC and STW Controller via CANBus. From an HMI developed
with Simulink, an operator can introduce if charger is connected, brake position, Key Start
Switch, DNR selector and HEV mode in real time. Throttle position is simulated with a
protoboard and two potentiometers installed, to be able to force a physical throttle error.
All these signals are managed within a CANBus network, and the controller decides the
state of the machine and it returns driving torque for the propulsion motors and
operating points for engine and generator. These information is sent again by CANBus to
Simulink and finally to Cruise, which runs a new simulation step and calculates
performance, velocity and consumption. This relevant data is showed on HMI for
visualization purposes as well.
Results and discussion
One of the most hazardous avoidable risks in a vehicle is the presence of positive torque with no throttle
pressed by the driver. HIL platform was developed to analyze potential causes and prepare VCU to avoid
them. For this particular case as an example, a good signal compared to a forced real hardware error are
displayed for every case.
As shown in figure, a normal fire-up process can be simulated from the beginning until second 20. It is
possible to see how the key goes from standby position to drive position. When the traction is enabled with
DNR switch (from second 30 to 80), throttle is pressed (light blue line), torque is requested (yellow line) and
vehicle speed increases (medium-dark blue line). During this process, state 3 is released (dark blue line)
corresponding to normal driving and potentiometers voltage are equal (grey and orange lines). Drive mode
remains always in 1, that corresponds to pure electric vehicle mode (green line).
A comparison of local path-planning interpolation
methods for autonomous driving in urban environments
Antonio Artuñedo, Jorge Godoy, Jorge Villagra

Center for Automation and Robotics (CSIC – UPM), Arganda del Rey, Spain
{antonio.artunedo, jorge.godoy, jorge.villagra}@car.upm-csic.es

Abstract
This work aims to provide an insight in the curve interpolation step within the path-planning architecture for autonomous vehicles. Taking into account
information from previous motion planning steps, the planner use the current pose of the vehicle and some waypoints in order to obtain the final path the
vehicle will follow. This path is required to have continuous curvature. This planning step is intended to run fast due to it has to be continuously running. In
this context, we present a comparison between different methods of interpolation based on Bézier curves, for obtaining optimal paths while the time and
performance requirements are met. Furthermore, different optimization algorithms and heuristics are applied to find the optimal values of several
parameters of the curves.

Introduction
Autonomous driving involves the integration of a number of technologies
related to perception, localization, decision-making, path-planning and
control. Among them is particularly relevant the path-planning as it plays
a key role in ensuring driving safety and comfort [3]. In recent years,
there has been extensive research on path-planning approaches for
autonomous vehicles [2, 4]. In this context, motion planning algorithms
must be able to quickly react to changes that occur on the Fig 1. Path planning architecture scheme
_
vehicle environment. Usually most of motion planning architectures divide the problem in global and local planning. In turn, local planning is generally
structured in 3 steps (see Fig. 1). Besides, within the local planning, a number of different techniques are today being evaluated. According to [1], these
techniques can be split in four groups: graph search, sampling, interpolating and numerical optimization.

Local path-planning interpolation approaches
Taking into account information from previous motion planning steps, interpolation curve planners use the current pose of the vehicle and some way-
points in order to obtain the final path the vehicle will follow. This path is required to have continuous curvature. This planning step is intended to run fast.
In this context, we present a comparison between different methods of interpolation based on Bézier curves, for obtaining optimal paths while the time
and performance requirements are met. Our purpose is to generate a piecewise Bézier curve path maintaining C2 continuity in way-points i.e. in the joints
of two Bézier curves. In our approach we addressed quintic Bézier curves and B-splines defined as composites of cubic Bézier curves, to meet the above
mentioned requirements. Furthermore, different optimization algorithms and heuristics were applied to find the optimal values of several parameters of
?
the curves. The objective is to minimize ?? = ‫׬‬0 ? ? ? 2 + ?(?) ሶ 2
??, where ? is the curvature, ? is the length over the path and ?? the total path length.

Results
Execution time
Curve type Optimization type Continuity
aprox. (s)*
Interior point: control points
Bézier 3rd C1 1.19
position
Heuristic: 1st derivative +
Bézier 5th C2 1.52
Interior point optimization
Heuristic: 1st derivative +
Bézier 5th C2 1.12
Interior point optimization
Heuristic 1: 1st derivative +
Bézier 5th C2 0.00005
Heuristic 2: 2nd derivative
Heuristic: 1st derivative +
Bézier 5th C2 0.025
2nd derivative optimization
B-Splines - C2 0.011
*Execution time for a path of 6 waypoints
References
Conclusions & future work [1] D. Gonzalez, J. Perez, V. Milanes, and F. Nashashibi. A review of motion
planning techniques for automated vehicles. IEEE Transactions on
The results are intended to help in future decisions about the most Intelligent Transportation Systems, 17(4):1135-1145, April 2016.
appropriate approach for local path-planning interpolation. In this sense, a [2] C. Katrakazas, M. Quddus, W. Chen, and L. Deka. Real-time motion
comparison of the execution time between different interpolation methods planning methods for autonomous on-road driving: State-of-the-art and
has been carried out. future research directions. Transportation Research Part C: Emerging
The future work will focus on the development of the whole path-planning Technologies, 60:416-442, 2015.
architecture for autonomous vehicles of AUTOPIA Program: [3] X. Li, Z. Sun, D. Cao, Z. He, and Q. Zhu. Real-time trajectory planning for
• How to properly choose the points between which the path will be autonomous urban driving: Framework, algorithms, and verications.
created. IEEE/ASME Transactions on Mechatronics, 21(2):740-753, April 2016.
[4] K. Yang, D. Jung, and S. Sukkarieh. Continuous curvature path-
• How to optimize the whole spatio-temporal problem, i.e. the motion
smoothing algorithm using cubic b zier spiral curves for non-holonomic
planning.
robots. Advanced Robotics, 27(4):247-258, 2013.
Method for knowledge based maintenance of
current collection systems in railway installations
aCalvo Hernández, Álvaro, bSanz Bobi, Juan de Dios
aalvaro.calvo@upm.es
bjuanedios.sanz@upm.es

Abstract
Maintenance activities are an important part of the time and resources that are used during the life cycle of an installation. Knowing the behavior of the
components of an installation a better maintenance strategy can be implemented.
Commonly implemented strategies of maintenance are based on the observation of effects like wear of components. Having this base in mind, there are
two kind of actuations. The first one is focused on reparation of components, once the wear has reached inadmissible levels and working conditions are
near to the safety failure or a failure has taken place, so all the installation has to be closed until components are replaced. As a second option,
replacement is done far before safety failure, when wear reach an established level.
Acting after a failure involves a delay due to the preparation of maintenance activities and to collect the materials needed. On the other hand, acting
before a failure gives the chance of prepare activities an collect materials while installation is still working but it means that components are not used as
long as they could, so there is a period of use that is wasted.
Trying to balance both cases the predictive strategies are applied, these methods allow to estimate the moment of the failure, increasing working time of
the components and reducing maintenance planning activities.
In order to reach a predictive strategy of maintenance of current collection systems in railway installations, variables that affect directly to the system
behavior have been studied and listed. Once the variables are set, behavior models have been developed and they are focused on the comparison of the
results obtained using a finite elements simplification and a complete finite elements implementation.

¿Where is the simplification limit?
Seeking a predictive strategy of maintenance, the reduction of the amount of analyzed data is a key aspect in order to reduce the time spent on
simulations.
Complex models require a specific software which can handle geometry data, physics and non lineal dynamic effects. Focusing on overhead conductor rail
as an example of current collection installation, a complex model of a pantograph and rail has been developed, including every single piece that is used in
a real situation. The large amount of data involved in the pantograph leads to a first approach of a simpler model, according to EN 50318 standard a
lumped mass model of the pantograph has been generated too, as shown in picture below (Fig. 1).
Behavior of lumped mass model is taken as a correct approach to real situations behavior, however, it may not collect failures related to surface extensions
or the fact that real pantographs have two contact surfaces and their behavior is not completely the same.
A part of the developed studies of the research is focused on the validation of these kind of simplifications and if they imply a loss of information that
could be important for a good predictive maintenance strategy.

Further steps of simplification
Overhead conductor rails can easily be studied as a structure. Looking the conductor rail as a beam and the support system as a bar, a purely theoretical
simplification can be developed as shown in the diagram below (Fig.2).
These simplifications commonly reduce the time spent in simulations and they also give good results for flexible overhead contact lines. The usage and
precision of these models applied to rigid systems is under investigation.

Fig. 2. Diagrams for Y configuration of flexible
OCL (upper) and OCR (lower)

Fig. 1. Complex complete model (right) and lumped mass complex model (left)
Accuracy improvement of finite element models of
electric vehicles components manufactured with
thermoplastic materials in crash environments.
aJavier Rivera Hoyos, bLuis Martínez Sáez, bAntonio García Álvarez.
aINSIA - Universidad Politécnica de Madrid (javier.rivera@upm.es)
bINSIA - Universidad Politécnica de Madrid.

Abstract
The present work is focused on the development of a methodology to improve the accuracy of the virtual tests results of large structural plastic
components, with high level of additives, when these parts have to withstand high levels of dynamic loads like vehicle crashes. For the development of the
proposed methodology, the design of battery rack on the OPERA4FEV project (FP7) is used. The OPERA4FEV rack is manufactured in a thermoplastic
material with fibre glass reinforced and flame retardant additives. These type of racks have to withstand crash load as defined in the United Nations
regulation 100. For the design of this rack, finite elements simulations and traditional tests have been used. Most important aspect about this work is
correct definition of material mathematical model used in simulation software due to the large quantity of thermoplastic components in rack. If material
definition is improved and the mechanical strength influence factors are optimized, it will be possible to increase the accuracy in final test results.

Two rack types were tested according its internal elements: one with
Introduction. ballast pieces and other with original electrical and electronics
Nowadays, the hybrid and electric vehicles industry is achieving a large components. Real crash tests were carried out in INSIA (Instituto
importance on world discussion about human sustainable and Universitario de Investigación del Automóvil), using an deceleration sled.
environment friendly mobility. For this reason, it is necessary more Tests were performed in five directions (longitudinal: frontal and rear,
research and new component developments to improve vehicle transverse: left and right, and vertical direction) according with R100
performance and so, its operational autonomy. OPERA4FEV was a Annex 8C acceleration corridors. Additionally, tests with overload
European project (FP7) whose aim was the development of a acceleration pulses were performed to validate the finite element model.
thermoplastic battery rack for an industrial vehicle (N2 class) using “eco
design” principles.

Battery rack contain eight battery modules
and each one includes twelve Lithium-ion
batteries.
Current state.
Present research work is focused in an analysis about factors influence on Previous crash tests on battery rack prototypes showed a higher strength
mechanical strength in the battery rack FE model, due the differences regarding virtual crash tests using R100 acceleration corridors. Several
between real crash test results and virtual crash test results. factors in the rack FE model are being analysed with crash tests
information and new material tensile tests data obtained directly from
Previous work. unaffected crash tested parts. The contribution of each factor on global
Element finite model of battery rack was created from a CAD original error will be quantified. All Johnson-Cook parameters in material
design. Model initial shape was structurally improved through an iterative mathematical model are being recalculated and optimized to improve the
and collaborative process design according with virtual crash tests results. mechanical behaviour in virtual tests. The plastic strain hardening, strain
These test were simulated and analyzed with ALTAIR HYPERWORKS rate variation and temperature influence will be considered on material
software. For other hand, tensile tests were also carried out to check the model. New Johnson-Cook model parameters are also being validated
material mechanical properties due it is a special composition with with virtual tensile tests. Virtual tensile tests will be performed until to
polypropylene matrix and short fibre glass (30% vol.), beside a high anti- obtain similar results regarding laboratory tensile tests. Furthermore, real
flame additive content. A initial material model was established from crash tests showed than internal components have less relative
tensile test results. Material model parameters (Johnson-Cook) were displacement than in the virtual model, so fixations between internal
obtained for each material temperature and strain rate from 0,002 up to battery modules with external covers will be reconsidered. The aim will
0,166 ? −1 . be obtained a better packaging, closer to real situation as observed
Mod. upper Pack Cover with HSBC and cooler during crash tests performed.
cover. pre-set.
Mod. 58Mpa Max stress area.
HVU
intermediate Cap.
cover.
Modules
(x8) BMS
Cap

Cells (x12) 2,52% Elements with max
Module pan. Pack Pan Silent blocks (x8) strain.
Once finished the design process, battery racks prototypes were
manufactured by moulding injection and later these were assembly with
internal components.

Acknowledgments.
This work was financially supported by the European Commission, in the framework of the project OPERA4FEV inside the FP7-2011-GC-Electrochemical-
Storage Programme with contract number SCP1-GA-2011-285671. Part of this research was funded by the Comunidad de Madrid through the SEGVAUTO-
TRIES, Seguridad de los vehículos Automóviles, por un Transporte Inteligente, Eficiente y Seguro, with contract number S2013/MIT-2713.
Strategies for grid integration of electric vehicles

a-cRoberto Alvaro-Hermana, bJesus Fraile-Ardanuy, cSergio Martinez
a Orkestra-Fundación Deusto, Deusto University (roberto.alvaro@orkestra.eusto.es)
bTelecommunication Technical School, UPM
c D. of Electrical Engineering, ETSI Industriales, UPM

Abstract
The research leading to these results has received funding from the EU 7th FP (FP7/2007-2013) under the project DATA science for SIMulating the era of
electric vehicles (DATASIM) (FP7-ICT-270833). DATA SIM aims at providing an entirely new and highly detailed spatial-temporal microsimulation
methodology for human mobility with the goal to forecast the nation-wide consequences of a massive switch to electric vehicles.
The aim of this thesis is to deepen in the study of urban’s mobility electrificacility, both at user level and at regional level. For this analysis a
microsimulation mobility tool is employed, examining if the different drivers in the region can use an electric vehicle and its impact in all levels of the
electric system. With this information, three applications have been proposed to maximize the integration of the electric vehicle: a system for managing
charging points and a peer-to-peer energy exchange systems between electric vehicles.

Charge Scheduling Strategies for Managing an
Electric Vehicle Fleet Parking
Different charging scheduling algorithms for managing the recharge process of
an electric vehicle fleet in a centralized parking. This tool is tested on a real-
world electric fleet which is charged using five charging stations. These algorithms
are also use to size the charging infrastructure, determining the minimum
number of chargers that are required to charge all electric vehicles.

EV charging algorithm in big data applications
A night charging method that optimizes the recharging process of electric vehicles
(EVs) depending on hourly energy price in a peer to peer (P2P) energy trading system
is presented. This algorithm determines how much energy should be recharged in the CS1 P1 F5 F5 F5 F5 F5 P3
battery of each EV and the corresponding time slot to do it, avoiding the Charging CS2 F11 F11 F11 F7 F3 F3 F7 F7
discontinuities.
Stations
CS3 F2 F2 F2 F2 F2 F4 F4 F4

Fully Charged EVs: CS4 F6 F6 F6 F8 F8 F8 F8 F6
1. Divide each hour in 4 time
slots (15 min each) 12 (10 fleet-2private) CS5 F1 F1 F1 F10 F10 F10 F10

P2P Energy Trading System
• A P2P trading system that interconnects both market actors: vehicles
with an energy surplus at the end of the day (set A) and vehicles
demanding extra energy along the day (set B).
• An aggregator sets the energy and the price to be given to set A
5. Reassign the vehicles depending on demand, energy stored in set A vehicles and
energy grid price.
demanded in the 2. Sort the Price in ascending order,
last 4 time slots reordering the time slots • P2P electricity price highly depends on ratio between set B and set A
vehicles. This ratio is usually low.
• Total charging costs for both vehicles sets is diminished. The P2P
Energy Trading System is day-charging independent.

4. Reorder 3. Assign the maximum
back the time possible energy to charge
slots. each of available time slots,
according to the previous
order.

2017 Alvaro, R.; Fraile-Ardanuy, J.; Merino, J., “Algorithm development for night charging electric vehicles optimization in big data applications” in 7th International Conference on Sustainable Energy Information Technology (SEIT 2017)
2016 Alvaro, R.; Gonzalez, J.; Fraile-Ardanuy, J.; Zufiria, P. Knapen, L.; Janssens, D., “Peer to Peer Energy Trading with Electric Vehicles” IEEE Intelligent Transportation Systems Magazine. pp. 33 - 44.12/2016. ISSN 1939-1390
2015 Alvaro, R.; Fraile-Ardanuy, J., "Charge scheduling strategies for managing an electric vehicle fleet parking," in EUROCON 2015 - International Conference on Computer as a Tool (EUROCON), IEEE , vol., no., pp.1-6, 8-11 Sept. 2015. doi: 10.1109/EUROCON.2015.7313732
LARGE-SIGNAL MODELING OF BIDIRECTIONAL BATTERY
CHARGER FOR ELECTRIC VEHICLES.
SYSTEM-LEVEL APPROACH BASED ON AGENT-BASED
MODELS IN V2G ENVIRONMENT
A. Naziris, R. Asensi, J. Uceda
a.naziris@upm.es, rafael.asensi@upm.es, javier.uceda@upm.es

Abstract work is the development of a model able to work at system level, studying the impact of the connection to the grid of a number of

The purpose of this work is the development of a model able to work at system level, studying the impact of the connection to the grid of a number of
electric vehicles demanding energy from the grid to charge their batteries, while allowing the grid operator to use the energy stored in the batteries for
optimal energy management , at system-level. Initially, a large-signal model of a bi-directional battery charger is developed to foresee the converter
behavior. Later, at system level an Agent-based model will be developed to asses the system’s performance, specifically in micro-grids. an Agent-based
model will developed to asses the system performance, specially in micro-grids.

Current work
• Two controllers were implemented in the in a white box model circuit, which specify the amount of energy
running through the Electric Vehicle (dc) and regulate the amount of energy that will be provided to and from the
grid(ac)
• Current work entails the development of a black box behavioural model, that will feature different operating
modes

Future work - Multi-agent modelling
By having thousands of potential EV, traffic conditions, charging stations and grid operating conditions the level of complexity is so high, creating the need for
compatible modelling techniques that deal with such level of complexity:

• Multi-agent modelling tool allows us to identify individual agents through their characteristics, behaviours and provide the ability to cooperate, act
autonomously and learn
• Interactions in a complex environment let us simulate the behaviour of such complex systems
• Development of simulation environment, able to foresee the performance of the grid for the operators and define the charging strategy for EV fleets

• Multi-agent modelling will be implemented with
the help of the MATSim program as it is illustrated.
• A fleet of EV´s (black-box) will interact -using
charging/discharging strategy- with the system in
order to limit the dissipation of energy in a
microgrid.
• Also it will help decrease the vast amount of energy
which the grid provides to a large fleet of EV.
Hybridization Percentage Influence on Hybrid
Energy Storage Systems Efficiency for Electric
Buses
Jorge Nájera, Jaime Rodríguez, Rosa Mª de Castro GELEO
Area of Electrical Engineering, ETSI Industriales, UPM research group
(jorge.najera.alvarez@alumnos.upm.es)

Abstract
Energy storage systems (ESSs) have been the main obstacle in the field of Electric Transportation for years, and recent research indicate that it will keep
being for even more time. Electrical storage technologies are the key to improve security, driving range and efficiency in every kind of Electric Transport,
e.g. Electric Buses and Electric Vehicles. At present, the technologies that can be implemented in Electric Buses are the following: batteries, fuel cells,
supercapacitors and flywheels.
Batteries and fuel cells are both energy density storage systems, while supercapacitors (SCs) and flywheels are both power density storage. Nowadays,
batteries are ahead of fuel cells regarding Technology Readiness Level (TRL), Regulatory Readiness Level (RRL) and Customer Readiness Level (CRT). In a
similar way, SCs are superior to flywheels for transport applications since they have more power density and better efficiency, lower cost, and they are
much easier to install and maintain. SCs main drawback is that they cannot be designed for a specific power and energy since both magnitudes cannot be
decoupled.
Hybrid Energy Storage Systems (HESS) using batteries and SCs have been proposed during the last years since they provide complementary capabilities.
The cooperation between batteries and SCs provides potential for achieving better efficiency and performance, but at the cost of a higher complexity. The
main reason why Electric Transports still use batteries as the only ESS is this extra expertise requirement, although probably future EVs will evolve towards
batteries and SCs HESS.

Electric Bus HESS Topology and Control Strategy
HESS Topology
HESSs for Electric Buses are formed by batteries, supercapacitors and traction drive, which is represented as one bidirectional load. The most studied HESS
topology is represented in Fig. 1, which comprises a high-power DC/DC converter since it has to handle SCs power. This topology provides flexibility and
increases the reliability during their implementation in an Electric Bus.
Control Strategy
Several control possibilities can be implemented in this
topology, since the DC/DC converter does not have to control
the DC-link voltage because it is set by batteries. This means
that there is no obvious manner to control the system. There
are rule-based strategies, which perform correctly with just five
or six rules, and advanced control strategies that optimizes the
system attending to different criteria.
In this work, the rule-based control strategy proposed in [1] has
been implemented, with minor modifications and having
optimized it in order to maximize the energy efficiency.
Fig. 1. Schematic representation of HESS topology used for this poster.

Results and Discussion
The driving cycle chosen for this work has been the New European Driving Cycle (NEDC), which is used as a standard driving cycle for testing vehicles
emissions. It is a 26-minute driving cycle, and the net energy consumption is 28 kWh. The hybridization grade is measured by the concept of “hybridization
percentage” (??????), which considers the proportion between the SCs weight (??? ) referred to the SCs and batteries weight (???? ) together. As the
?????? increases, SCs substitute batteries while keeping the HESS weight constant.
Global DC-DC DC-DC + bats
70
???
100 ?????? =
60
???? + ???
50
HESS efficiency [%]

Final SoC [%]

95
40

30

90 20

10

85 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Hybridization percentage [%] Hybridization percentage [%]

Fig. 2. HESS average efficiency as a function of Fig. 3. Global SOC (batteries + SCs) as a function of the
hybridization percentage. hybridization percentage, at the end of the driving cycle.
Fig. 2 illustrates HESS global efficiency, DC/DC converter efficiency and the DC/DC converter + batteries efficiency. Global efficiency increases for low
hybridization percentages. It is clear that replacing batteries by SCs reduces the driving range, as it is shown on Fig. 3, where the State of Charge (SoC) at
the end of the driving cycle decreases as the ?????? increases.

References
[1] R. Carter, A. Cruden, and P. J. Hall, "Optimizing for Efficiency or Battery Life in a Battery/Supercapacitor Electric Vehicle," Vehicular Technology, IEEE
Transactions on, vol. 61, pp. 1526-1533, 2012.
Remote Drive: A military UGV for surveillance
missions
aMiguel Clavijo, bFelipe Jiménez, cJosé E. Naranjo, dÓscar Gómez
aINSIA. Universidad Politécnica de Madrid (miguel.clavijo@upm.es)
b,c,dINSIA. Universidad Politécnica de Madrid

Abstract
Autonomous driving for ground vehicles is one of the most promising technology that will change the current transport system. In this way, there are
several applications already developed for the autonomous navigation for civil uses. However, the military field is one of the main promoters of this type of
technology, carrying out different applications in off-road areas. The final approach of military application differs to those of the civil uses; therefore, in
military missions, the integrity of the vehicle is not the priority compared to the success of the mission.
In this work, the automation of an unmanned ground vehicle (UGV) for the Spanish army is described. The military vehicle has to perform three main
functions: teleoperation, path-following and automatic emergency stop. Thus, it is explained the automation of the main actuators of the military vehicle
and the main algorithms used for the surroundings perception.
Algorithms for obstacles detection and
Functional architecture for the automation boundaries estimation of the path

LiDAR technology for obstacles detection &
HD video transmission for teleoperation
Set for steering wheel Set for braking
automation automation
Patent ES2516568

URO VAMTAC – Testbed vehicle
Teleoperation user interface
Horizontal rotary system for concentrating
solar radiation
José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
Horizontal rotating platform where symmetrical longitudinal strips of parabolic mirrors are located, and in parallel and in height the radiation receiver, with
its focal line inside the vertical plane of symmetry, is placed on a few rods, these rods being firmly attached to the platform, which Rotates on at least two
sets of concentric wheels, and in its rotation the vertical plane of symmetry of the device contains virtually the solar disk. The parabolic profiles are
prescribed so that the mirrors can be made with originally flat mirrors, which are deformed by their own weight adapting to the configuration required for
the optical focus

A horizontal rotating device for concentrating the solar radiation, consisting of a horizontal platform, rotatable about a vertical
virtual axis, on which rests a plurality of longitudinal mirrors of parabolic cross-section, horizontally lying parallel to each
other, with a symmetrical configuration With respect to a vertical plane containing the virtual vertical axis, and constituting
the plane guiding the orientation of the device relative to the sun, the criterion of approach being that the sun, as a point disc
seen from Earth, is in the plane of Symmetry of the device; The device further comprising a longitudinal radiation receiver,
positioned in height on the platform by means of rods or porticos which are integral with said platform, the central line of the
receiver being contained in the vertical plane of symmetry of the mirror field resting on The platform being characterized in
that the mirrors are made up of a succession of longitudinally arranged mirror mosaics which are held in solidarity with the
rotating platform because each mosaic is held firm in position by four staples located on the sides of the mosaic, Each being
near a corner of the mosaic, two staples of one mosaic being attached to an inner longitudinal spar and the other two staples
being secured to an outer longitudinal spar, said spars being attached to the structure of the rotating platform and supported
by There are two longitudinal members for each row or mirror strip, although adjacent spars of neighboring rows can be
assembled together to form a longitudinal girder; Each securing clip having a particular spatial location coinciding with a point
of a virtual parabolic focusing profile, the staple in question having an inclination at that point which coincides with the line
tangent to said profile at said point, and said parabolic profile Its focus on the point selected as a focus for the whole set of
virtual parabolas from which the profiles of the mirror mosaics are extracted, for which the focus distance, which is always the
same, to the minimum of each parabola is the fourth Part of the inverse of the coefficient of proportionality that multiplies to
the square of the abscissa of said virtual parabola of focus to obtain the corresponding ordinate
Brayton cycle with heat sink temperature close
to the critical isothermal line
José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
Regenerative closed cycle, whose minimum working fluid temperature is in turn close to, but higher than, the critical temperature, and close to the
ambient temperature used for cooling the cold spot; Selecting the value of the high isobar in coincidence with the so-called supreme isobar, which
presents the maximum average value of specific heat at constant pressure, within the peri-critical zone, above the critical isotherm, and delimited in
pressure between a Fifth and five times the critical pressure, setting the low isobar by providing the maximum of a ratio in which the numerator is the
specific work, and the denominator is the sum of the absolute values of the specific enthalpy variations of each phase

Brayton cycle with ambient cooling close to the critical isotherm Tcr of the working fluid according to
previous claims, characterized in that the cycle is limited by the minimum values ​Tn and maximum TM of
the working fluid temperature, with the ideal regenerative cycles being identified by the Compression
ratio used, which must be simultaneously greater than 1 and smaller than Ca, the latter being
Ca = (TM / Tn) γ / 2 (γ-1)
And defining a utility function L as:
L = (Q-E) / ((1 + a) • Q + (1 + b) • E + G)
where:
- the numerator is the net power generated in the cycle, which is the enthalpy added in the hot spot (Q)
minus the enthalpy extracted in the cold spot (E);
- and the denominator is the sum of the three enthalpy changes: a) in the heating and expansion phase
(Q), b) in the cooling and compression phase (E) and c) in the regenerative exchange phase (G ), where:
Q = Cp • (TM-TM / x)
E = Cp • (Tn • x - Tn)
G = Cp • (TM / x - Tn • x);
Being "a" and "b" weighting factors chosen according to the machinery to be used, a> 1, b> 1, and "x" a
parameter given by
x = c (γ-1) / γ
So that there is an optimal value of the compression ratio, cmax, that maximizes the ideal utility function
L, y corresponds to:
Cmax = xmax (γ / (γ-1))
Where xmax depends on the weighting factors "a" and "b" chosen according to the machinery to be used,
and the temperature ratio r defined as the quotient between TM and Tn, so that:
Xmax = ((2 + a + b) + (4 (a + b) 2 - 4 (1 + a + b - (1 / r)) (1 + a + b - (1 / r)).
Intermediate heat exchanger of free-dilatation
concentric tubes
José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
Heat exchanger consisting of independent tubular elements, each of which consists of a series of concentric tubes, which form at least two concentric
circulation ducts of the two fluids that exchange heat through the intermediate tubular wall that prevents the mixing of both fluids And supports the
difference of pressures between them; In addition to having other concentric ducts delimited by guide tubes, to properly channel the flow of fluids and
allow free expansion of all parts

A free-flowing heat exchanger for high-pressure fluids consisting of a plurality of tubular elements (1) which are
independent from each other with respect to the internal flows of the fluids, said elements being arranged in
parallel in the elongated cells of a Three-dimensional scaffolding structure supporting them, and said elements
sharing the inlet and outlet manifolds of each fluid, the outlet manifolds of each fluid being located at opposite
ends, and the intake manifolds being located at the same end, characterized in that each Of these tubular
elements (1) comprises two concentric circulation ducts (21, 22) of the two fluids which exchange heat through
the intermediate tubular wall (11) which prevents the mixing of both fluids and supports the difference of
pressures between them (11), which in turn contains a concentric circular tube therein, or a smaller diameter
guide tube (13), which channels the fluid from the inner circuit, or inner fluid, from The fixed head (31), to which
it is integral at one end, to the opposite end of the inner circuit, where it simply remains open, with its opening
facing the cap (12) closing the resistant inner tube, whose other end Is attached to the corresponding annular
space of the head (31); (22) extending annularly along its entire length, which forms the outer circuit, which is
externally closed by a resistant outer tube (10), which Supports the difference in pressures between the fluid in
the outer circuit and the surrounding medium, which is usually the atmosphere, although there is a thermal
insulation thickness (9) around the outer resistant tube (10), which lacks pressure resistance ; Each end of the
outer circuit (22) being connected to its corresponding inlet or outlet manifold, of the outer fluid
Convection depressed open-air solar thermal
receiver
José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
A longitudinal receiver in which the tubes of heating fluid are located inside a first open cavity surrounded by thermal insulation material, which in turn
opens into a wider cavity, delimited by the same thermal insulation; There being, on the outside, another separate, parallel insulation which starts from a
lower dimension than the termination dimension of the first insulation; And in the longitudinal direction, the upper wall of the first cavity increases its
dimension linearly in successive sections until reaching a vertical stop or limiting wall of each section, the height of the upper wall falling to the level of the
beginning of the section.

Atmospheric thermosolar receiver, with a draft, in consecutive modules or sections,
characterized in that the tube or tubes of heating fluid are located inside a first open cavity
surrounded by thermal insulation material, or first insulation, which in turn Opens into a
second wider cavity, likewise externally delimited by the same thermal insulation; There
being on the outside of said insulation, on each side, each air channel, each of which is
delimited by the outer part of the above insulation and the inner part of a parallel insulation
that develops outside the first insulation, and starts from a dimension lower than the
termination rate of the first insulation; And in the longitudinal direction, the upper wall of
the first cavity is structured in successive sections, and said wall is increasing its dimension
within each section, in the direction in which the heating fluid moves within the tubes; Until
the top wall reaches a vertical stop or limiting wall of said section, the height of the top wall
falling to the original level having said top wall at the beginning of said section
A concentric tube penetration device

José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
A concentric tube penetration device in a free expansion heat exchanger, wherein the outermost of the concentric tubes is welded to the flashing through
which the exchanger closure plate passes, and the outer end of said tube is closed with A welded plug, and in a point of the tube near that end, four cuts
are made delimiting a piece of the wall of the tube, which is extracted, welding in that wall hollow a parallelepipedic piece constituting the radial outlet
manifold from The heat exchanger, the inner tube remaining intact but welded to the end closure of the outer tube just next to the annular cap which
blocks the movement of the fluid in its sinus and directs it towards the radial outlet

Device for penetration of concentric tubes in freely-expanding heat exchanger, a shutter plate (1) of the exchanger and a pair of
concentric tubes (2 and 4) selected between metal or plastic; Capable of being welded metalically, in the first case, or being
welded or adhered by resins or other plastic means, in the second case; (4) entering a thermal exchange fluid until it reaches its
end, where it pours against the firm plug (7) of the end of the outer tube (2) whereby the flow of fluid returns to the plate ( 1) of
the exchanger, through the annular space 3 existing between the outer face of the inner tube 4 and the inner face of the outer
tube 2 which in turn receives on its outer face the thermal effect 15 Of the heat exchange to be performed in said apparatus
requiring double inlet / outlet penetrations, said device characterized in that each penetration is of two concentric tubes, inner
and outer, the outer tube (2) being welded to at least one of the faces of the (1) of the exchanger, around the entire
circumference of the cavity (6) in said plate (1), the welding being carried out as a circular continuous bead (8) between the edge
of said flashing (6) and The circular crown of the cylindrical surface of the outer tube (2) identified as area (18) where to fix said
tube to said closure plate (1) leaving a portion of both outer and inner tubes (2 and 4) outside the inner chamber of the (16),
without affecting the inner tube, and other two parallel longitudinal incisions (17), from the first and second parallel longitudinal
incisions (17). The outer tube piece separated from the rest by the four mentioned incisions (16 and 17), a lateral opening (10)
being left at the edges of the tube, the four corresponding sides of a hollow parallelepiped piece (13) being welded, (12) from the
annular space (3) between said two tubes, said stream (12) being channeled through said piece (13) ) Towards the exit of the
system by means of the exhaust manifold (14) which is in physical continuity with the parallelepipedal piece (13), the outer tube
(2) being terminated at its outer end by means of an annular cover (11) In the annular space between both tubes, which lid is
welded both to the outer tube by its mouth or end, and to the inner tube (4) on the circular crown of the cylindrical surface of
said inner tube (4) identified as area (19). ) Where to fix said tube to said annular closure cap (11) of the outer tube (2).
Gasket for assembling rotary pipes

José M. Martínez-Val & GiT Research Group
mval@etsii.upm.es

Abstract
Rotating joint for assembling fixed tubes with rotating tubes less than one full revolution.
The gasket comprises a hollow cup type piece welded at the end of the fixed tube, in which the end of the rotary tube is screwed, a washer being trapped
at the top of the thread; And having in the outer part of the rotating tube an annular square in the radial flap of which rests a washer captured by the
flange cap, the flange being screwed to the outside of the nexus piece, whereby this second washer is released from the pressure when The first one is
very compressed, and vice versa, having to pass the fluid of the interior by both washers to leak, which is rather difficult

Patent ES2504916
Fecha de prioridad: 29.01.2014. Fecha de concesión: 24.02.2015

Rotating joint for assembling fixed tubes with rotating tubes smaller than a complete revolution, characterized in that it
comprises:
- a fixed tube (2) and a rotating one (1), the rotating tube having the outer face of its end (7) threaded in several turns, being
screwed onto the inner face (6) of the nexus piece (4) At the end of the fixed tube, this nexus being a short and thick cylinder,
with a lip or small transverse annular bracket (9) projected towards the inside, and a cylindrical main body threaded at its inner
face with the same thread pitch Than the outer face of the rotating tube, with which it is screwed;
And the rotating tube ends at its end in a transverse plane section of said tube, identified as a flat stop of said tube, which
collides with the inner annular lip (9) of the connecting piece, if it is screwed tightly , Which does not occur by interposing
between said surfaces, supported on said lip, a washer or seal (8) of radial dimensions that coincide with said lip (9), and is made
of a material that remains solid at the temperature of Operation, and at that temperature has an elastic modulus, or quotient
between mechanical stress and deformation, less than that of the material of the tubes;
- an outer annular bracket (5) integral with the rotating tube (1), which is welded to it in a position such that its transverse
annular flap (10) remains close to the free end of the connecting piece (4), without interfering in The threaded surface (7) of the
rotating tube, but close to it, and without touching in any rotating position the nexus piece and said outer annular square;
- there is furthermore an annular flange (12) in the form of a cover with a hollow central part, concentric outside the nexus (4),
composed of a cylindrical body and an annular transverse plate with the entire hollow center to pass freely therefrom (4), by
means of welding or by radial screws (13) with its consequent cylindrical seal (15) between the nexus part and the inner face of
the cylindrical part or flange (16) of The flange leaving an annular space between the inner part of the hollow flap (17) of the
flange, and the flap (10) of the outer annular square integral with the rotating tube;
- locating a washer or seal (11) in said annular space remaining between said inner portion of the flange cover (17) and the flap
(10) of the outer annular square (5) integral with the rotating tube (1)
Procedure and system for the automatic
detecion of thrust reverse activation during aircraft landing
aC. Asensio, aM. Ruiz, aG. de Arcas, aM. Recuero , aJ.M. López
aInstrumentation and Applied Acoustics Research Groupm – Escuela Técnica Superior de
Ingenieros Industriales – Universidad Politécnica de Madrid (casensio@i2a2.upm.es)

Abstract
The invention consists of a sound sensor which is capable of detecting the activation of thrust reverser during aircraft landing from the detection and
identification of the sounds produced in the process. Also, the invention includes the procedure to be followed with respect to the location of the sensors.
The invention will provide the airports with a tool to ensure compliance with the regulations regarding the activation of reverse noise and environment.

True positives

True negatives
On‐board car system and method for road 
status surface identification
López Navarro, Juan Manuel; Alonso Fernández, Jesús; Ruiz González, Mariano; 
de Arcas Castro, Guillermo; Pavón García, Ignacio y Asensio Rivera, César
ignacio.pavon@upm.es
Grupo de Investigación en Instrumentación y Acústica Aplicada. ETSI Industriales. UPM

Patent ES 2 390 302 B2
Abstract
An‐board embedded system to reliably estimate the surface road conditions, by analyzing the noise produced because of the interaction of tyres with the
road is proposed.
Resulting system, called Acoustic Asphalt Analyzer (A3) is based on an electronic system, which is embedded on‐board, for the detection of the
weather road conditions by the acquisition and processing of the obtained acoustic signals. It can be applied to surface auscultation or active safety in
vehicles, being useful in initiatives such as connected vehicles and unmanned vehicles. A3 will allow to know the surface road conditions in real‐time and
provide data to the infrastructure manager, other drivers and the driver himself.

The on‐board car system comprises:
‐ Acoustic transducers (MR) configured to record the rolling noise generated by the
interaction of a vehicle tyre (R) with the pavement (C), and
‐ A central processing system (SP) configured to receive and analyze the signal from
the acoustic transducers (MR), and comprising:
• A features extraction module for extracting spectral features (CE) of the signal
determined during a previous training process;
• A support vector module (DE) for storing support vectors (VS) obtained after a
previous training process; and
• A classifier module (CL) to execute an algorithm based on support vector
machines on the spectral features (EC) and the support vectors (VS) extracted,
and to obtain an estimate of the state of the pavement (EA).

Principle of operation:

Acoustic Asphalt Analyzer (A3) samples the tyre‐road noise generated
during the displacement of motor vehicles. As the acoustic footprint
obtained depends on the road status, it is possible to estimate the pavement
conditions by analyzing the obtained rolling noise samples.

More information:

http://www.i2a2.upm.es/resultados/patentes‐2/deteccion‐del‐estado‐del‐asfalto/
http://www.upm.es/observatorio/vi/index.jsp?pageac=innovacion/articulo.jsp&id_tipo_articulo=3&id_articulo=355
Electronic system for enviromental acidity (pH) 
measurment based on optical sensors
Mª Á. Villegasa, M. Garcíaa, J. Peñaa, G. de Arcasb, J.M. Lópezb, E. 
Barrera, A. Llorentea
A Centro Superior de InvestigacionesCientíficas, 
B Grupo de Investigación en Instrumentación y Acústica Aplicada, g.dearcas@upm.es

Abstract
The invention consists of an electronic system that provides the measurement of environmental acidity (pH) by means of chemical sensors with optical
response. The main innovation is based on the ability of the system to measure pH of all media, including gaseous phases (e.g. air), which is an important
advance in respect to conventional methods based on electrodes that work in liquid media or humid solids.
The system is formed by one or several sensors and one or several measurement devices that can be remotely connected to a PC. The measurements can
be accomplished either simultaneously or stored for further evaluation.

The Sensors Applications
The sensors consist of thin coatings in which a sensitive phase (dye) is Environmental acidity affects directly:
immobilized. The coatings are applied upon common glass pieces by the sol‐ • the air quality breathed by living organisms,
gel procedure. The dye is sensitive against the environmental pH and yields
• and all materials exposed to it.
an optical response in the wavelength range of the visible absorption
spectrum, i.e. the sensors change their color when the environment acidity The automatic and remote monitoring of the environmental pH with
changes. The sensors do not need any kind of energy source to work. sensors would be a fast procedure to early know the pH variations, as well
as to predict damages and, possibly, to avoid them.
Moreover, since the sensors developed are able to measure the acidity of
liquids and humid solids (soft or disaggregated), their applications can be
enlarged.

The Measurement Device
The measurement device transforms the sensors' optical response to an
electrical signal (volts). This signal is scanned and stored in the
measurement device and then it is sent to a PC for further evaluation.
Finally the pH values are obtained through the adequate software installed
in the PC.
Characteristics
Size (mm) 30x25x1 (can be miniaturized)
Sensitive range (pH scale) 5‐8 (can be adapted to other ranges)
Accuracy ( pH units) 0.1
Response time 1 min or less in liquids, up to 24 hours in gas and solids 
depending on humidity.
Service time (months estimated) indoor >10; outdoor >3
Thermal resistance (0C) ‐10 to 60
Mechanical resistance (qualitative) that of the conventional glass substrate. 
Resistance against scratches: similar to those of pure silica glass.
Photostability (qualitative) good under visible, UV and solar up to 500C
Energy consume zero (passive sensor) 

IMPLEMENTATIONS & VALIDATION
Harvesting unit – lots of sensing points, passive sensing Wireless sensor network – real time, active sensing
• Sensors are placed throughout the monitoring area  • A battery powered autonomous node that holds has 
without any need for cabling (power supply…) been developed to hold a sensor and convert its output 
• Data is collected manually with the “harvesting unit”  into a digital data that is transmitted wirelessly
when needed and it’s then transferred to a PC. • Specially suited for wide area real time networks
It’s been used in cultural heritage conservation projects such a: Royal Glass This implementation was developed in collaboration with CEI‐UPM
Works of La Granja (Spain), Library‐Archive Tomas Navarro (Spain), Army www.cei.upm.es and has been used to monitor air quality in food (cheese
Museum of Toledo Fortress (Spain), Navy Museum (Spain), Historic Library of and meat) industries in Italy in the scope of the European Project GIST
the Spanish Military Academy (Spain), Holy Gosh Church of CSIC (Spain), Royal funded by the Executive Agency for Competitiveness and Innovation (EACI)
Palace‐Museum of Wilanow (Poland), Wawel Castle (Poland). under the CIP Eco‐Innovation Programme.

Wilanow Royal Palace‐Museum (Warsaw) Library‐Archive Tomas 
Navarro Tomas (Madrid)
Sistema de Calibración de Instrumentos Acústicos

G. de Arcas, J.M. López, M. Ruiz, M. Recuero, R. Fraile
Grupo de Investigación en Instrumentación y Acústica Aplicada, g.dearcas@upm.es

Resumen
Los Sonómetros y los Dosímetros de Ruido se utilizan para realizar medidas que permitan asegurar el cumplimiento de diferentes regulaciones contra el
ruido, ya sea en el ámbito de una ciudad, o en los lugares de trabajo. El objetivo de dichas regulaciones es preservar nuestra salud, ya que se ha
demostrado que la exposición abusiva al ruido puede tener graves consecuencias, como problemas de sueño, efectos cardiovasculares (la UE estima que
casi el 2% de los ataques de corazón son debidos al ruido), pérdidas de audición, etc. Por tanto en muchos países se obliga por ley a que este tipo de
instrumentos se verifiquen periódicamente en un laboratorio. Este sistema fue diseñado para automatizar el proceso de Calibración y/o Verificación
Periódica de instrumentos para la medida del sonido audible (sonómetros, dosímetros de ruido, calibradores acústicos, audiómetros y micrófonos).

Alcance y Componentes del Sistema

Analizadores 
Sonómetros 
UNE 61672
de Espectro
UNE‐EN 61260

acoustiCAL‐FLT®
acoustiCAL‐ND® Dosímetros 
Micrófonos 
UNE‐EN 61094 acoustiCAL‐SC® de Ruido
UNE‐EN 61252
acoustiCAL‐SLM®
acoustiCAL‐DAQ®

Calibradores 
Acústicos Audiómetros 
UNE‐EN 60645
UNE‐EN 6094

Explotación de Resultados y Transferencia de Tecnología
Universidad Politécnica de Madrid, España Instituto de Salud Pública, Chile
• Metrología voluntaria. Servicios de Calibración  • Transferencia de Tecnología, Formación, Asesoramiento 
acreditados por ENAC en las Áreas de Acústica y  y Auditoría para el establecimiento del Laboratorio de 
Electricidad CC y Baja Frecuencia. Calibraciones Acústicas del ISP.
• Metrología Legal. Organismo de Control  • Primer Laboratorio en América Latina.
(Módulo F) y Organismo Autorizado de 
El Instituto de Salud Pública (ISP) de Chile desarrolla
Verificación Metrológica.
funciones de Referencia, Vigilancia, Autorización y
El Laboratorio de Calibración de Instrumentos Acústicos fue establecido Fiscalización en el ámbito de sus competencias. La
gracias al desarrollo de esta tecnología en el año 1998 y desde entonces puesta en marcha de este Laboratorio fue el primer paso
presta servicios de calibración y verificación de instrumentos acústicos a la para desarrollar una red metrológica en el país y
industria en varias comunidades autónomas. establecer un sistema de control metrológico.
System and procedure for the real time detection
and identification of sound produced by specific sound sources
aC. Asensio, aM. Ruiz, aM. Recuero
aInstrumentation and Applied Acoustics Research Groupm – Escuela Técnica Superior de
Ingenieros Industriales – Universidad Politécnica de Madrid (casensio@i2a2.upm.es)

Abstract
System and method for detecting and identifying specific sounds produced by specific sound sources in real time. The system can detect target sounds
among others, to identify specific sound sources or conditions. The system applies one-class pattern recognition techniques, which improves the
recognition rates in circumstances where a good representation of the non-target class cannot be gathered (for instance, aircraft sounds can be well
described, while non-aircraft sounds can be anything). The a posteriori probability of the classifier determines the likeness between the sound under test
and the target sound.

Can you hear  Environmental noise
the target monitoring
 Failure detection
sound?  Aids for elderly or
handicapped
What is the  Surveillance
target?  Quality check
System and method for monitoring a rapid de-
excitation system for synchronous machines
aCarlosA. Platero, aRicardo Granizo, Emilio Rebollo
& aFrancisco Blánquez.
aTechnical University of Madrid / ETSII /GELEO (carlosantonio.platero@upm.es)

Abstract
A system and method for monitoring a rapid de-excitation system for synchronous machines (1) with indirect excitation, via an exciter machine (2) and a
rotating rectifier bridge (8), comprising:
-means for measuring the current (13) and voltage (14) of the stator (4) of the synchronous machine (1);
-means for measuring the excitation current (15) of the exciter machine (2);
-estimation means (24) for obtaining the estimated excitation current (19) of the exciter machine (2) from the current (16) and voltage (17) values
measured in the stator (4) of the synchronous machine (1);
- a module (28) for comparing the measured excitation current Imed (18) with the estimated excitation current Iest (19), in such a way that:
•there is either incorrect operation if Imed > . Iest + A, where k and A are real numbers determined on the basis of each machine;
•or else there is correct operation.

2 1

+
4
5
11 9
-
6
15 7 8 10
13
3
12

17 14

18
16

16
12
21 24
P P 19
25 20
22
U U X P2 + (U+ X Q)2 √ k A 27
B>A TON
23
17 Q Q B

26

18

Conclusions
- HSBDS supervision has been into commercial
Diode operation in four 20 MVA / 40 MVA hydro
Shunt Testing Slips Rings
generators since September 2013, without
remarkable troubles.
- Patents
Disconnector (HSBDS out)
P2012231071
Transistor Switch
PCT/ES2013/000160
WO2014/009576 A1
USA 14412303
HPP Quereño. HSBDS in operation in a 20 MVA Hydro generator.
Rapid de-excitation system for synchronous
machines with indirect excitation
aCarlos A. Platero, aFrancisco Blázquez,aMarta Redondo &
a Ricardo Granizo.

aTechnical University of Madrid / ETSII /GELEO (carlosantonio.platero@upm.es)

Abstract
Synchronous machines with brushless excitation have the disadvantage that the field winding is not accessible for
the de-excitation of the machine. This means that, despite the proper operation of the protection system, the slow
de-excitation time constant may produce severe damage in the event of an internal short circuit.
A high-speed brushless de-excitation system (HSBDS) for these machines was developed in our Laboratory. First a
laboratory 5 kVA prototype and after the test in a 15000 kVA machine in Alstom factory (Bilbao), 4 Hydro generator
are in service with this HSBDS.

P20090468 PCT/ES2010/000058
The invention relates to a rapid deexcitation system for synchronous machines (1) with indirect excitation by
means of an excitation machine (2) and rotating rectifier bridge (8) comprising:
- a deexcitation impedance (10) connected between the field winding (5) of the synchronous machine (1) and the
rotating rectifier bridge (8);
- a controller (9) connected in parallel with the deexcitation impedance (10);
- a control circuit (11) of the controller (9) configured to:
• keep the controller (9) closed such that the rotating rectifier bridge (8) directly feeds the field winding (5) of the
synchronous machine (1) during normal operation of the synchronous machine (1);
• open the controller (9) such that the deexcitation impedance (10) remains in series with the field winding (5)
and with the rotating rectifier bridge (8) when the synchronous machine is to be deexcited

2 1
AC Exciter Main Synchronous Machine

4
+
5
11 9
-
6
7 8 10
3 Rotating componets
15 MVA Brushless Synchronouos Generator.
2000
GE Alstom Factory.
1500

1000 Without
Generator Currents [A]

500

0
HSBDS
-500

-1000

-1500

-2000
0 1 2 3 4 5 6 7 8 9 10
Time [s]
2000

1500

1000 With
Conclusions
Generator Currents [A]

500

0
HSBDS
-500

-1000 - HSBDS has been into commercial operation in
-1500
four 20 MVA / 40 MVA hydro generators since
September 2013, without remarkable troubles.
-2000
0 1 2 3 4 5 6 7 8 9 10
Time [s]

- The sudden short circuit tests in a 15 MVA
Country Date Application Number
India 01/08/2011 5862/DELNP/2011
show that the I2t can be reduced 10 times.
Canada 08/08/2011 CA2751943 - Patents
China 12/08/2011 201080007623.1.
USA 17/08/2011 13/201,971.
P20090468
Brasil 18/08/2011 PI1008626-9. PCT/ES2010/000058
South Korea 18/08/2011 10-2011-7019197
Europe EPO 16/09/2011 EP10743430.0.
PROCESS OF FUNGAL BIODEGRADATION ON
GROUND TYRES
Constantino Ruibal, Ana M. García, Diego A. Moreno,
Enrique Benavides

ETSI Industriales, Universidad Politécnica de Madrid (ana.garcia.ruiz@upm.es)

Abstract
The present biodegradation process consists in the use of a fungus, Paecilomyces lilacinus IMI 117109, for the degradation of commercial ground
tyre rubber. Prior to the biodegradation treatment, the tyre crushing is washed to eliminate possible contaminations that may interfere with the
process. Subsequently the fungus is cultured in the presence of the tyre rubber in a liquid medium and shaking conditions. In a second step, the
excess of liquid medium is removed. After the incubation period, the growth of the fungus in the material can be observed and its degradation is
showed. This process of fungal biodegradation is an alternative to the degradation processes currently employed, allowing to reduce the
accumulation of used tyre rubber and taking advantage of the products of its biodegradation for its reuse in the tyre industry or its use in other
industrial processes.

Overview
The latest official data indicate that in Spain more than 250,000 tonnes of used tyres are
generated annually and it is estimated that there are approximately 4 million tonnes more
accumulated in landfills. The current legislation requires the elimination of tyre dumps and
the management of the waste by those responsible for placing it on the market.
During the last years biodegradation processes applied to polluting materials have been
developed with a view to the subsequent use and reuse of their components. The
biodegradation of rubber is a matter of great interest because it could be a complementary
path to the existing tyre management.
The process of vulcanization suffered by natural rubber in tyres is responsible for its
desirable mechanical and thermal properties but, at the same time, it makes more complex
to reuse them. This process forms disulfide bridges between the polymer chains. Further
steps during the manufacture of the tyres involve the addition of multiple additives,
required for vulcanization or the acquisition of various properties in the final product. Such
additives are mostly toxic, introducing an additional problem to the biodegradation
process.
In this patent a biodegradation process for commercial ground tyre rubber (Fig. 1) is
described using the fungus Paecilomyces lilacinus IMI 117109. Fungal growth is an Fig. 1 Commercial ground tyre rubber used as
unmistakable sign of the biodegradation process that is taking place in tyre grinding (Fig. 2). starting material.

Description and main features of the invention
1) Minimal culture medium
In order to carry out the biodegradation of the commercial ground tyre rubber the fungus is cultivated in a liquid medium of mineral salts in which
the tyre crushing constitutes the only source of carbon for the microorganism.

2) Material cleanness
Prior to the incubation of the tire with the fungus, the grind is washed under certain
conditions: ratio of crushed weight / volume of sterile water used, temperature, agitation
and time. This washing removes the organic residues and microorganisms that could
interfere or displace the biodegradation reaction.

3) Two-phase cultivation
The cultivation of the fungus with the ground tyres in the medium of mineral salts is
carried out in two phases. In the first one an excess of liquid medium and agitation is
used. This phase is necessary for the activation and correct distribution of the fungus in
the material. In the second phase, the excess of liquid medium is removed, leaving a
static solid-wet culture, in which the fungus continues growing on the crushed.

4) Possibility of using other microorganisms
This biodegradation process can be applied employing other microorganisms as long as
Fig.2 Low temperature scanning electron microscopy they possess the metabolic capabilities for the degradation of the ground tyre, adapting
(LT-SEM) image of the fungus growing inside the the culture conditions to the particular physiology of the selected microorganism.
commercial ground tyre rubber.

Patent information
Title: Procedimiento de biodegradación fúngica de triturado de neumáticos.
Inventors: Ruibal Villaseñor, Constantino; García Ruiz, Ana María; Moreno Gómez, Diego Alejandro; Benavides Ruiz, Enrique
Applicant: Universidad Politécnica de Madrid
Publication number: ES 2 293 864 (A1) (B2) Application number: 200703236
CPC: C08J11/10; C12N1/14; Y02W30/702 IPC: C08J11/10; C12N1/14 Priority date: 2007-12-05
BIOREMEDIATION METHOD WHICH IS USED TO
CONCENTRATE AND ELIMINATE RADIONUCLIDES IN
RADIOACTIVE WATER
aDiego A. Moreno, bFelipe Montero
aETSIIndustriales, Universidad Politécnica de Madrid (diego.moreno@upm.es)
bIberdrola Generación S.A.

Abstract
A completely innovative bioremediation procedure has been invented to concentrate and eliminate radionuclides from the radioactive water of
nuclear pools. This method is based on the capacity of microorganisms to retain radionuclides when they are growing by forming biofilms on the
colonized surfaces. Preferably the bioremediation is performed before the water is passed through demineralizing filters to prolong the life of such
filters. As a result, there is less radioactive waste to be managed, thereby increasing the profitability of the method.

Main features of the invention
Currently the radioactive water of the nuclear pools is treated with a
system of demineralising filters made from ion-exchange resins
whose goal is to retain the dissolved substances and those in
suspension that can be found in the water, such fission and corrosion
products, etc. so that the water is much clearer and the radioactivity
in the area surrounding the pools is significantly lower. As time
passes, the life expectancy of the resins is lowered and it is necessary
to store them in drums and to treat them as radioactive waste. The
new bioremediation procedure, however, retains the nucleotides Fig. 3. Biofilm developed on
found in the radioactive water before they reach the resin of the stainless steel balls after 186 days
demineralising filters, and as a result, increase their life expectancy, immersion.
thus reducing the volume of radioactive material to be managed.
In this patented procedure, the radioactive water from the nuclear
pools under treatment, goes through a bioreactor (Figures 1 and 2)
that contains a metallic material, such as titanium or stainless steel,
wound into a ball, or any other material that is non-corrosive or non-
degradable in this environment and can be colonised by the
microorganisms that exist in this type of water. As it goes through the
bioreactor, the radioactive water to be treated comes into contact
with the material of the ball placed inside, thus forming a biofilm
(Figures 3, 4, and 5) that retains the radionuclides. The water then
goes out from the opposite end of the continuous-flow entrance
and, before going back into the pool, it goes through a system of
demineralising filters. Fig. 4. Biofilm developed on
titanium balls after 97 days
Biofilms developed on the balls were analyzed by scanning electron immersion.
microscopy and epifluorescence microscopy. DNA was extracted
from the biofilms, was cloned and the microorganisms were
identified by sequencing. The identified radioresistant organisms
Fig. 1. Cross-section of
belong to the phylogenetic groups Alpha-proteobacteria, Gamma-
bioreactor.
proteobacteria, Actinobacteria, Deinococcus-Thermus and
Bacteroidetes. These microorganisms are capable of retaining the
radioisotopes present in the water of the pools by one or more of
the following mechanisms: bioadsorption, bioabsorption,
bioaccumulation, biosolubility, bioprecipitation.
By gamma spectrometry we also determined that biofilms have
essentially retained activation radionuclides. Sometimes the sum of
Co-60 and Mn-54 reached 98%. Cr-51, Co-58, Fe-59 Zn-65 and Zr-95
have also been retained by the developed biofilms. Fig. 5. Biofilm of Pedomicrobium sp.
It has been assessed that the scalable process can be economically on titanium balls at 218 days
profitable. immersion.
Fig. 2. Top view of the
bioreactor made of stainless TECHNOLOGICAL SECTOR TO WHICH THE INVENTION RELATES
steel .
Nuclear power industry and management of radioactive water
Patent information
Title: Procedimiento de biorremediación para concentrar y eliminar radionúclidos en aguas radiactivas
Inventors: Diego A. Moreno (UPM), Felipe Montero (Iberdrola Generación SA)
Applicant: Universidad Politécnica de Madrid
Publication number: ES2190758 A1 Application number: P200200006 Priority date: 2002-01-02
Publication number: US7326345 B21 Application number: US 10/500,222 Priority date: 2008-02-05
CPC: Y02W10/15, Y10S210/912, C0F2101/006, C02F3/10, G211F9/18 IPC: C02F2101/006, C02F3/10, G21F9/18
PORTABLE AND AUTONOMOUS DEVICE FOR
AIRBORNE PARTICLES COLLECTION
aAndrés Núñez, Ana M. García, Diego A. Moreno

aETSI Industriales, Universidad Politécnica de Madrid (andres.nunez@upm.es)

Abstract
A portable device for collec9ng airborne par9cles with autonomous opera9on, comprising a container (1) having an inlet (2) of air from outside into
the container (1), and an outlet (3) of air from inside the container (1) to the outside; so that between the inlet (2) and outlet (3) of air, the container
(1) comprises, inside, a first module (7) for capturing airborne par9cles, which in turn comprises: a first means of filtering (4) for the air, allowing the
passage of air and par9cles smaller than the pore size; and a first receptacle comprising, in turn, a first plate (5) covered with a tacky substance or
other adhesive surface on its anterior major surface and configured to collect, at least, part of the par9cles present in the air (Fig. 1).

Overview
In addi9on to gases and inorganic par9cles, the air carries many biological par9cles such as
pollen, fungi, bacteria and viruses. Some can trigger allergy symptoms in the popula9on
and be also the cause of airborne disease transmission like the flu, pneumonia and
legionellosis. Crops and farm produc9vity are also affected by these airborne organisms,
causing an important economic impact. The monitoring and detec9on of all this biological
diversity in the air outdoor is a difficult task because of their par9cular characteris9cs
regarding size, rela9ve abundance, residence 9me in the atmosphere… Hence, the different
biological par9cles are usually studied independently.
To solve these troubles, we have developed an equipment a2achable to any kind of vehicle
to collect all the biological par9cles present in the air (sor9ng by size if desired), and
keeping the sample intact for further biochemical detec9on (e.g. DNA sequencing) or Fig. 1 Scheme showing the components of the device.
iden9fica9on by microscopy.

Main features of the invenTon
1) Autonomy and simplicity
Collec9on is performed passively, promoted by the advance of the vehicle, forcing the air to pass through the equipment. The par9cles are harvested
and retained inside the device on an adhesive surface. Therefore, electric power is not required and the opera9ng 9me is unlimited.

2) Modular structure
The filtering (4) and collec9on (5) components have a modular organiza9on (7), so it is possible to add or
remove modules in order to increase the collec9on surface or select different size for the par9cles to harvest
in each module (Fig. 2).

3) Reusable: brand-new
The prototype is made of metal and glass to be cleaned easily and sterilized in the autoclave, so the
components can be used over and over to collect fresh samples.
Fig. 2 Picture of the modules and
the different parts of the device.

4) Built to be linked to any vehicle
The cylindrical design, small size and light weight make it simple to a2ach to
many plaWorms and different means of transport such as cars, buses, airplanes,
UAVs, ships… (Fig. 3), increasing the op9ons of collec9on 9mes and the places
where the air samples can be taken.

5) Upgradable
Addi9onal features like lids to control the collec9on, a flow-meter coupled with
a data-logger to monitor the volume of air sampled, the use of disposable
collec9on receptacles… can be easily integrated in the equipment to improve
its performance.
Fig. 3 Pictures of the equipment a2ached to different vehicles.

Patent informaTon
Title: Disposi9vo captador de par[culas presentes en el aire de carácter portá9l y autónomo.
Inventors: Núñez Hernández, Andrés, Ana María García Ruiz, Diego Alejandro Moreno Gómez
Applicant: Universidad Politécnica de Madrid
PublicaTon number: ES2558792 ApplicaTon number: ES20150031836
CPC: B01D45/04; G01N1/22 IPC: B01D45/04; G01N1/22 Priority date: 2015-12-18
PROCEDURE FOR MARKING, ENCRYPTION, OpNcs
Fotonics
LABELLING AND OPTICAL CODING Biofotonics

aM. Holgado, bA. Barranco, aR. Casquel. bA.R. González-Elipe, F.J. Sanza
aM.F. Laguna, bSanchez valen cia Juan Ramon; bA Rebollo aF.J. Sanza
A Corresponding author, UPM (m.holgado@upm.es)
B ICMSE-CSIC

Abstract 101
It permits marking or recording mo1fs (301) on surfaces whereon there
has been deposited previously a fluorescent polymer layer (300) 202
through a procedure of plasma polymerisa1on (100) of molecules of a
colourant. The procedure combines the special characteris1cs of the 301
polymer layers making them suitable for being capable of recording
diverse mo1fs (301) thereon, and the possibility of recording by laser 300
(101) or other techniques. Among such characteris1cs should be
men1oned the possibility of a notable visual effect, including for
thicknesses of 100 nm, the use of layers non-observable when
illuminated with visible light, the high op1cal quality (transparency)
thereof, or the facility wherewith they may be processed by subsequent
treatments, including diverse laser treatments 400
Under UV
illumina1on

UV laser wri1ng
Without damage Laser abla1on
Damage

Without UV ilumina1on

 
Interferometric Detec-on Method
a,b,cMiguel Holgado, b,cFrancisco Javier Sanza, a,b,cMaría Fe Laguna,
a,bÁlvaro Lavín, a,b,cRafael Casquel
a Departamento de Física Aplicada e Ingeniería de los Materiales, ETSII-UPM

(m.holgado@upm.es)
b Center for Biomedical Technology (Op-cs, Photonics and Biophotonics Laboratory)
c Bio Op-cal Detec-on (Centro de Empresas Campus UPM Montegancedo)

Abstract
The Interferometric Detec/on Method is characterized by the use of two interferometric signals, which allows for the op/cal reading system to convert
the changes caused by the op/cal transduc/on into a unique, sensi/ve variable of detec/on. Therefore, two interferometric measurements are used: a
first interferometric op/cal reference, which represents the measured intensity modulated by a reference interferometer; and a second interferometric
signal measurement observed in the sensing region of the signal interferometer. The sensing surface region is where changes caused by the biomolecular
interac/on takes place. A transduc/on func/on [fTRANS] is then constructed from the interferometric measurements, and analyzed to determine the
biosensing response caused by the biological accumula/on in the sensing observa/on region.

Theore-cal
Explana-on

Theore/cal calcula/on of the reflec/vity as a func/on of the wavenumber and
wavelength for the reference interferometer and for the signal interferometer. It is also
represented the op/cal interroga/on band, which is the spectral range from 808 nm to
864 nm used to obtain the op/cal power for both interferometers. The capital leOers: A,
B and C; represent three different situa/ons. A is the area defined by the op/cal
interroga/on band and the interferometric profile of the Fabry-Perot Interferometer
(FPI) used as reference. This area is, therefore, propor/onal to the op/cal power of the
reference interferometer. C and B represent two different situa/ons. C is the area
defined by the op/cal interroga/on band and the interferometric profile of the signal
FPI when the biofilm thickness is 0 nm. B, represents the area defined by the signal
interferometer and the op/cal interroga/on band for a biofilm thickness of 40 nm. As a
conclusion, when the biofilm thickness is increased, the spectrometry profile of the
signal FPI is closer to the reference FPI, producing a lower level of IROP signal.

Implementa-on to
Point of Care (PoC)
Applica-ons
devices
OPTICAL DETECTION SYSTEM FOR LABELLING-
FREE HIGH-SENSITIVITY BIOASSAYS
A,bM.Holgado, a,bR.Casquel, , a,bM.F. Laguna a,J.L. Ocaña, aM. Morales aC. Molpeceres
a Departamento de Física Aplicada e Ingeniería de Materiales.
b Center for Biomedical Technology (OpHcs, Photonics and Biophotonics Group)

m.holgado@upm.es

Abstract
The patent describes and opCcal detecCon system for labelling-free high-sensiCvity bioassays, which comprises the use of opCcal characterizaCon
techniques that allow verCcal interrogaCon in micrometric and sub-micrometric domains and also at least one verCcal-interrogaCon biophotonic
interferometric or resonant cell. The system comprises, at least: (i) an opCcal measuring system that comprises one excitaCon source, for the detecCon of
the signal and an opCcal head; and (ii) an element for the integraCon of mulCple analytes, comprising, in turn, and at least, a plurality of biosensiCve cells,
a plurality of fluid connecCons connected with the cells, and a substrate on with not only the biosensiCve cells but also the fluid connecCons. The system is
designed to perform label-free high sensiCvity bioassays, with sensing chips fabricated at wafer scale.

DescripHon of the patent OpHcal head λ R
Sensing chips
The system consists of an opCcal interrogaCon system (opCcal head with a θ
detector) and a variety of optofluidic sensing cells. The opCcal head interrogates
verCcally the optofluidic sensing cell and obtains three different opCcal responses
in a single measurement: ellipsometry, reflecCvity as a funcCon of the angle of Fluid in Fluid out
incidence, and reflecCvity as a funcCon of wavelength.
The opCcal informaCon is taken using a single measurement, and can be combined
and processed. This is interesCng in parCcular for bioassays, because can reduce
the uncertainty of the detecCon.
OpHcal head
Sensing cells
Fluid in Fluid out
The chips can be fabricated at wafer-scale, High-
througput screening
Sensing chip – high mulHplexing

Sensing cell: arrays of Resonant pillars ReflecHvity as a funcHon of wavelength ReflecHvity as a funcHon of angle of incidence
0,7

0,6

0,5 0,6

SiO2
Relative reflectivity
Relative reflectivity

0,4
microcavity 0,5

0,3

0,4 reference
0,2
reference with protein layer
Reflectors with protein layer
0,1
Si3N4, aluminum 0,3

Fluid in Fluid out 0,0
600 625 650 675 700 725 750 775 800 0 10 20 30 40 50 60

wavelength (nm) angle of incidence (degrees)

The opCcal head allows measuring in sub-micrometric domains, even covering a single pillar. The
Protein layer:
Calculated opCcal responses are for pillars of 500 nm, with two reflectors of Silicon Nitride, and a central
thickness = 15 nm
Refractive index = 1.4 cavity of Silicon dioxide. The bioreceptors cover all the surface of the pillars, enhancing sensiCvity.

ReflecHvity as a funcHon of wavelength ReflecHvity as a funcHon of angle of incidence
Sensing cell: Fabry-Perot CaviHes 1,1

R 1,0
1,0
Reflectivity (relative units)

Reflectivity (relative units)

0,9
0,9

θ 0,8
λ Fluid out 0,8
0,7
Rspectrometry
R Δn0.01

Fluid in 0,7 0,6

shift
0,5
0,6
0,4 Rp
Rp Δn 0,01
0,5 0,3
700 720 740 760 780 800 820 840 860 880 900 920 -60 -40 -20 0 20 40 60
wavelength (nm) angle of incidence

This sensing cell is designed to obtain a narrow resonance in order to increase the sensiCvity of the measurement
The cell consist of a Fabry-Perot cavity, with two Bragg Reflectors, of Silicon and Silicon Dioxide, and a central cavity
filled with fluid, where the biorecogniCon takes place. Values of limit of detecCon for refracCve index sensing are in
Si substrate the order of 10-5.

ApplicaCon number of the patent: EP20090811139. European Patent approved
Method and system for estimating the forces and
torques generated by the contact of a tire with
the road in an instrumented wheel
Javier García De Jalón De La Fuente, María Dolores Gutiérrez López
(javier.garciadejalon@upm.es)

Abstract
Radial Line
The invention relates to a method for measuring the forces and the
moments generated by the tire-road contact from the combination
of unit deformation or strain signals measured in different angular Gauges
and radial positions of the wheel. The angular positions in which the
sensors are placed are those permitted by the geometry of the tire.
The combination of deformation signals results in six or more signals
that are independent of the angular position of the measuring
sensors with respect to ºthe tire-road contact point. The effect of
temperature and loads that generate deformation signals that do
Concentric
not depend on said angular position of the measuring points, such
as centrifugal forces, have been eliminated from the circumference Gauges
aforementioned signals. Said signals provide estimates of the loads
by means of the resolution of two systems of linear equations with
constant matrices and three unknowns each.

Theory
• The method proposed is based on bonding
strain gauges to the rim and on using
harmonic elimination techniques. The strain
gauges are arranged in concentric
circumferences (at least two) and in angularly
equidistant radial lines (at least five)

Strain gauges & amplifier boards
• Strain gauges are sensors which can measure deformations, pressures
and torques. These sensors have a resistance whose value changes
when a deformation occurs. As a consequence, the value of the
current changes.

Telemetry
• The current system is based on the WiFi technology. Every signal is broadcaster by wireless. The
system is composed by three elements depicted in Fig.

Competitive Advantages of the Solution
• Provides real-time information to dynamic control systems,
improving them.
• Price far less than competition that can make it viable to a wider
spectrum of researchers and companies.
• Adaptability to a wide range of rims.
Medical devices based on intelligent materials and complex
geometries patented and developed during last decade within
UPM’s Product Development Laboratory
aAndrés Díaz Lantada, bPedro Ortego García, cPilar Lafont Morgado
aLaboratorio de Desarrollo de Productos, ETSI Industriales - UPM (adiaz@etsii.upm.es)
bLaboratorio de Desarrollo de Productos, ETSI Industriales - UPM (portego@etsii.upm.es)
cProfessor Emeritus, ETSI Industriales - UPM

Abstract
The Product Development Laboratory (Laboratorio de Desarrollo de Productos) at ETSI Industriales, is the first “fab-lab” from Universidad Politécnica de
Madrid. It was founded in 1997 by Prof. Pilar Lafont Morgado and has been running for 20 years providing support to research projects, teaching activities
and industrial innovation tasks in all aspects linked to the development cycle of innovative products. Aspects including creativity promotion, intellectual
property management, conceptual design, computer-aided technologies for product engineering and rapid prototyping usually by means of additive
manufacturing technologies (3D printing) are key areas of expertise of the UPM’s Product Development Lab. Engineering applications in fields including
energy, transport, space and health benefit from the combination of the advanced design and manufacturing technologies available and from the expert
advice of Lab’s members.
Among the most interesting engineering fields, to which the Lab has been devoted during the last decade, it is important to highlight the Biomedical
Engineering area. As human geometries are complex and as human materials are “intelligent” or multifunctional, additive manufacturing technologies
capable of manufacturing very complex geometries are specially well suited to manufacturing devices for interacting with humans. The design, modeling
and incorporation of “smart”, “intelligent” or multifunctional materials to such devices helps to enhance the diagnostic and therapeutic capabilities of such
biomedical devices. In addition, the design using complex fractal-based geometries helps to promote biomimicry and biomimetic approaches towards
improved performance. This poster presents some of the more remarkable patents of our group in the field of Biomedical Engineering, linked to
biomedical devices for improved diagnoses and enhanced therapies, which take benefit from the combined use of bioinspired geometries and of smart
materials and structures.

Diagnostic devices based on intelligent materials for improved diagnoses
Complete development of an instrumented splint for monitoring bruxism based on piezoelectric & piezoresistive
smart materials with near field communication for e-Health approaches 2

Bruxism is a pathology consisting of the involuntary clenching of teeth, 1,8

1,6

1,4

with impact on facial pain and increased teeth wear. Adequately 1,2

Tiempo (s)
1

monitoring bruxism constitutes a challenges towards improved control
0,8

0,6 1,8

0,4
1,6

of patients’ evolution. Our device is based on smart sensors included 0,2

0
0
1,4

51,2 10 15 20 25 30
Voltaje (V)

within conventional splints for quantitative multi-axial measurements.

Voltaje (V)
1

0,8

0,6

Patent: “Sistema de telemetría mediante comunicación inalámbrica
0,4

0,2

0

empleando campo magnético sensor-pasivo / interrogador-activo
0 10 20 30 40 50 60 70 80 90 100
Tiempo (s)

para diagnóstico y detección de episodios bruxistas”. 31 de Marzo de Prototypes of instrumented In vivo trials and results from
2009. Número de Solicitud OEPM P200900875. Concedida con examen splints for monitoring bruxism measuring bruxist episodes
previo el 15 de Febrero de 2010.

Active prostheses based on shape-memory materials for adaptative purposes
Complete development of an annuloplasty ring using shape-memory polymers for geometrical adjustment after
implantation towards enhanced surgical prognosis
Mitral valve insufficiency can be treated with the help of passive
annuloplasty rings, which provide mechanical support to the prolapsed
tissue. However, surgery with conventional passive rings can be risky
and devices for post-surgical adjustment of rings’ geometries are a key
towards improved prognosis. Our device pursues such challenge.
Patent: “Sistema de anuloplastia activo para el tratamiento progresivo
de insuficiencias valvulares y otras patologías cardiovasculares”. 13 de
Computer-aided design of shape- Prototypes and in vitro trials of the
Diciembre de 2006. Número de Solicitud OEPM P200603149. Concedida
memory annuloplasty ring shape-memory effect
con examen previo el 16 de Mayo de 2008.

Bioinspired tissue engineering scaffolds and culture platforms for studying cells
Fractal-based micro-textured and micro-structured
Based on biomimetic fractal geometries, which cannot be obtained
using traditional design and manufacturing approaches, our tissue
engineering scaffolds and cell culture platforms can activate special
respones, helping to control mesenchymal stem cells’ fate.
Patents:
“Substrato cuasibidimensional para crecimiento de células y tejidos y
método de obtención del mismo”. 20 de Junio de 2010. Número de
Solicitud OEPM P201030957. Concedida con examen previo el 27 de
Julio de 2012. (UPM-CSIC). Scaffold design obtained Rapid prototype for cell Cell culture upon rapid
“Soporte tridimensional sintético de apoyo a tareas de producción with help of CAD software culture manufactured, prototyped scaffold for
artificial de órganos y otras estructuras del organismo humano y directly from CAD file, in subsequent in vitro /
método de obtención del mismo”. 20 de Junio de 2010. Número de Scale bar: 500 microns an additive way in vivo studies
Solicitud OEPM P201030956. Concedida con examen previo el 16 de
Junio de 2011. (UPM-CSIC).
FEEDING CARRIAGE FOR COMPRESSOR 
MACHINE AND USE OF SAME
aOrdieres, J. , bAlba‐Elías F., bGonzález Marcos A., bMartínez L.
aETSII / UPM (j.Ordieres@upm.es)
bETSII / UniRioja ({fernando.alba,ana.gonzalez,laura.martimez}@unirioja.es)

Abstract
The invention describes a tractor truck for tableting machine, of the type having an upper filling end by which a particulate material for compressed in a
matrix of the compression machine and a lower discharge end of said particulate material to said matrix is i ntroduced . At least part of the constituent
elements of said carriage feeder can move relative to one another, thereby allowing to modify the geometry of at least one of said lower discharge end,
said upper filling end and an intermediate portion of the carriage feeder between both ends . The invention also discloses the use of a tractor truck of this
type, comprising the initial steps of determining the optimum geometry for a particulate carriage to be compressed and to apply this optimum geometry
carriage to carriage feeder.

FEEDING CARRIAGE FOR COMPRESSOR MACHINE AND USE OF SAME
Publication number: 20160114549

Focus
• To improve the effectiveness of dosage when mixture ratios are strict.
• To develop a method making possible to estimate the best geometry for the
feeding carriage.
• To extend the patent to Japan and USA (triadic patent).

30%
Porcentaje del ingrediente I2 (%)

15%

0%
0 5 10 15 20 25 30
Method for Increasing the Net Electric Power
of Solar Thermal Power Plants
aCarlosA. Platero, aFrancisco Blázquez, aMarta Redondo
& aJaime Rodríguez.
aTechnical University of Madrid / ETSII /GELEO (carlosantonio.platero@upm.es)

Abstract
The present invention allows increasing the net electric power supplied to the network by the solar thermal power plants of the cylindrical parabolic
collector type by using the solar field in a more efficient manner, generating steam to drive the main ancillaries by means of steam turbines (turbo pumps),
reducing electric consumption of the ancillary services and therefore increasing the net electric power of the plant.
On days when the solar radiation is greater than the designed radiation, part of the solar field must stop being used since the generator and the turbine
would exceed its designated power. The present system proposes harnessing the unused portion of the solar field for generating steam to drive the main
ancillaries of the plant, energy which would otherwise not be harnessed.
In addition, the overall operating performance of the ancillary service pumps is better when using steam to drive the pumps through a steam turbine
(turbo pumps) instead of driving them with electric motors (motor pumps).

Country Date Application Number
(1) Solar collectors (2) Auxiliary burner of natural gas or another fuel. Spain 16/04/2009 P200901000
Algeria 16/09/2011 EP10743430.0.
(3) Steam generator. Thermo fluid heat and steam (4) Thermo fluid pump. Australia 01/08/2011 5862/DELNP/2011
exchanger. Brazil 08/08/2011
Chile 12/08/2011
(5) Thermo fluid regulating valves. (6) Main steam turbine
China 17/08/2011 13/201,971
(7) Electric generator (8) Condenser. Egypt 18/08/2011 020110086502.
United Arab Emirates 18/08/2011
(9) Water supply pump supplying water to the steam (10) Main transformer
USA
generator. EPO 18/08/2011 10-2011-7019197.
(11) Electric network (12) Ancillary service transformer India
Israel
(15) Steam turbine to drive the thermo fluid pump (16) Steam turbine to drive the water supply pump Libya
supplying water to the steam generator Morocco
Mexico
OAPI
South Africa 05/10/2011 2011/07281
Tunisia 07/10/2011 TN2011/0505
European Research & Innovation Office

Vice Rectorate for Research, Innovation and Doctoral Studies
Roberto Martínez (roberto.martinez@upm.es)
Ana Belén Bermejo (anabelen.bermejo@upm.es)

Abstract
The European Research & Innovation Office at UPM aims at giving support to UPM researchers to participate in the European research programs in order
to increase the quantity and quality of European funded projects in which UPM participates.
The goal to be achieved by the European Office can be described as follows:
• to increase the weight of UPM participation in EU funded projects in terms of projects coordination.
• to facilitate first participations of newcomer researchers.
• to facilitate the collaboration of UPM researchers with the private sector.
• to increase the visibility of UPM at European Associations and Technology Platforms.

Services provided
For UPM researchers

Pre
re-
e-award Post
st-award

Participation in European fairs, seminars
and networking days, representing the
POSITIONING COST REPORTING
UPM research community on international
Associations and Technology Platforms.

Support to the UPM researchers in their Contract negotiation and forms
PROPOSALS proposal preparation phase (completing filling
CONTRACT
PREPARATION consortia, revising the proposals, pre- IPR issues
PREPARATION
checking with EC and NCPs the proposal
Consortium Agreements
ideas, etc.).
Cost claims and audits
Organization of H2020 information Relation with the EU
seminars, Post-doctorate course for
TRAINING European Project Managers, information TRAINING
and advice to interested researchers.

Contact with relevant European projects
COLLABORATION participants to look for collaboration
SCOUTING AUDITING
opportunities.

For other organizations

It represents a single point of contact for partners searching for a specific technological capacity within the University.

Networks
OTRI - Unidad de Cátedras Universidad-
Empresa
Ana Goicolea, Lidia Cerezo y Miguel Holgado

Unidad de Cátedras Universidad-Empresa
OTRI – Vicerrectorado de Investigación, Innovación y Doctorado
otri.investigacion@upm.es

Abstract
El objetivo de la Unidad de Cátedras Universidad-Empresa de la OTRI es promover la colaboración estable entre la
Universidad y el tejido empresarial a través de un modelo concreto, las Cátedras.

¿Qué es una Cátedra Universidad- ¿Qué actividades realizan?
Empresa UPM? ACTIVIDADES DE FORMACIÓN
Es un convenio que formaliza y hace pública la
Programas de postgrado, becas, premios a
colaboración a largo plazo entre la universidad y una
proyectos fin de carrera, conferencias…
empresa para la realización de actividades diversas
alineadas con los fines generales de la UPM en un área
temática concreta.
ACTIVIDADES DE DIFUSIÓN
Jornadas de divulgación técnica y tecnológica,
¿Qué beneficios ofrece? publicaciones…

9 Acceso directo a los recursos ACTIVIDADES DE I+D
humanos y a determinadas Promoción de líneas de investigación, apoyo a
infraestructuras de la Universidad la realización de tesis doctorales…
9Apoyo a diferentes actuaciones
Beneficios dentro de ámbitos temáticos de su ¿Qué duración tienen?
interés mediante: La duración mínima de una Cátedra es de 3 años.
para la • Programas de becas para la
empresa realización de proyectos fin de ¿Cuál es la financiación mínima?
grado o de máster
Se establece un mínimo de 30.000 € anuales.
• Trabajos de investigación
Cátedras UPM por centros
• Promoción de jornadas de
divulgación
E.T.S.I.
Resto UPM Telecomunicación
17%
9Relación con empresas de primer 32%
E.T.S.I. Minas y
Energía
nivel 12%

9Marco de financiación estable E.T.S.I.
Industriales
E.T.S.I. Agrónomos
9%
para la realización de actividades: 7%
E.T.S.I. Sistemas E.T.S.I.
E.T.S.I. CCPP
Informáticos 8%
Aeronáuticos
• Cooperación en programas de Beneficios 7% 8%

formación
para la
• Patrocinio de actividades de
difusión Universidad Fuente: UPM. Datos a marzo de 2017.

• Realización de estudios e
investigaciones + 70 Cátedras Universidad-Empresa
• Otras actividades enmarcadas en la UPM
en los objetivos planteados
Cátedras
Universidad-Empresa
en la ETSI Industriales
¿Dónde debo dirigirme?
Unidad de Cátedras Universidad Empresa
www.upm.es/CatedrasUniversidad-Empresa
otri.investigacion@upm.es
OTRI- Unidad de Colaboración Público–Privada
(UCPP)
Juan Sanz, Norma García y Miguel Holgado
.

Unidad de Colaboración Público-Privada
OTRI – Vicerrectorado de Investigación, Innovación y Doctorado
juan.sanz@upm.es

Abstract
La Unidad de Colaboración Público Privada de la OTRI nace con el objetivo de servir de instrumento de
dinamización, apoyo y promoción de la transferencia del conocimiento generado en la UPM al sector
productivo.

La Unidad de Colaboración Público–
Privada centra su actividad en el
ámbito de la INNOVACIÓN.
Entre la Investigación y el Desarrollo Tecnológico
generado por la UPM y su introducción en el
mercado a través de los instrumentos de
Comercialización, existe un escalón intermedio en el
cual la intervención de esta Unidad puede facilitar
un incremento de llegada de nuestra producción
científica y tecnológica al mercado.

¿Qué servicios presta?
¾ Gestión directa de convocator