Está en la página 1de 7

Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 153 (2016) 173 – 179

XXV Polish – Russian – Slovak Seminar “Theoretical Foundation of Civil Engineering”

Verification of tensegrity properties


of Kono structure and Blur building
Wojciech Gilewskia * -RDQQD.áRVRZVNDb, Paulina Obarab
a
Warsaw University of Technology, Faculty of Civil Engineering, Al. Armii Ludowej 16, Warsaw 00-637, Poland
b
.LHOFH8QLYHUVLW\RI7HFKQRORJ\)DFXOW\RI&LYLO(QJLQHHULQJDQG$UFKLWHFWXUH$O7\VLąFOHFLD33.LHOFH25-314, Poland

Abstract

Cable-bar structures are commonly used in engineering. Some of them are in special configuration named “tensegrity”. The
objective of the present paper is to describe and verify two interesting examples of existing civil engineering structures called
tensegrity in the literature: Kono structure and Blur building. Singular value decomposition of the compatibility matrix of truss
structures is used to define two crucial features of tensegrity: existence of mechanisms and self-stress states. Eigensolution of the
stiffness matrix extended on geometrical stiffness matrix allow to identify if the self -stress stabilizes the equilibrium of the
structure – if the mechanisms are infinitesimal or finite.
© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-reviewunder
Peer-review under responsibility
responsibility of the
of the organizing
organizing committee
committee of theofXXV
the XXV
PolishPolish – Russian
– Russian – Seminar
– Slovak Slovak Seminar “Theoretical
“Theoretical Foundation
Foundation
of of Civil Engineering.
Civil Engineering”.

Keywords: tensegrity; pin-joned structures; domes

1. Introduction

Nowadays the concept of tensegrity structures [4,6] is understood in many ways [12]. A widely accepted version
is proposed by Pugh [13] as follows: “A tensegrity system is established when a set of discontinuous compressive
components interacts with a set of continuous tensile components to define a stable volume in space.”
Key features of tensegrity structures can be defined as [8]:

* Corresponding author. Tel.: +22 234-57-53.


E-mail address: w.gilewski@il.pw.edu.pl

1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the XXV Polish – Russian – Slovak Seminar “Theoretical Foundation of Civil Engineering”.
doi:10.1016/j.proeng.2016.08.099
174 Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179

− the structure is a truss,


− there are self-stress sates,
− there are infinitesimal mechanisms and they are stabilized by these self-stress states,
− the system of compression elements is discontinuous,
− compression members are inside the tensile elements,
− tensile elements have no rigidity – they are strings.
The presence of all of these characteristics allows qualifying a structure as a tensegrity structure.
A key step in the design of tensegrity structures is the determination of their geometrical configuration, known as
form-finding. The form-finding process determines a possible pre-stress distribution and geometry for a tensegrity.
The method based on the singular value decomposition (SVD) [5,9] is used in the paper to identify whether the
structure is geometrically variable and whether there are self-stress states.
The term “tensegrity” is so attractive that many engineering structures are called with its use. Two interesting
examples of existing civil engineering structures are presented and analyzed in Chapter 3.

2. SVD decomposition

The subject of the analysis is N-membered, supported truss with following characteristics: material constants E e ,
cross-sectional areas A e and bar lengths L e . Its mechanical properties are described by three linearized equations:
compatibility, material properties and equilibrium with boundary conditions included

ǻ = Bq , S = Eǻ , BT S = P (1)

where q is displacement vector of length M, Δ is extension vector, S is normal force vector, E is elasticity matrix, P
is load vector and B is compatibility matrix which can be determined directly or using the formalism of the finite
element method [8,14]. The singular value decomposition of an N×M real matrix B is a factorization of the form:
B = YNX T
(2)

where Y is an N×N real orthogonal matrix, X is an M×M real orthogonal matrix and N is an N×M rectangular
diagonal matrix. Let us consider two eigen problems

(BB T
− μ I ) y = 0 and (B T B − λ I ) x = 0 (3)

with the solutions in the form of eigenvalues and eigenvectors (normalized)

μ1 , y 1 ; μ 2 , y 2 ; ...; μ N , y N and λ1 , x1 ; λ2 , x 2 ; ...; λ N , x N (4)

Full solutions of the above eigen-problems can be expressesd in the condensed forms
BB T = YMY T and B T B = XLX T (5)

M = diag {μ1 μ1 ... μ N }, L = diag {λ1 λ1 ... λ M }, Y = [y 1 y 2 ... y N ], X = [x1 x 2 ...x M ] (6)

One can notice that the product BBT can be considered as a matrix of symmetrised equations of equilibrium with
non-negative eigenvalues. Zero eigenvalues (if any) are related to the non-zero solution of homogeneaous equations
(P=0) named self-stress. The self-stress can be considered as an eigenvector related to zero eigenvalue).
In a similar way the product BTB can be considered as a particular form of linear stiffness matrix with unit
elasticity matrix. The eigenvalues are non-negative. Zero eigenvalues (if any) are related to the finite or infinitesimal
mechanisms, but in general the information from the null-space analysis alone does not suffice to establish the
difference. The mechanism can be considered as an eigenvector related to zero eigenvalue.
Based on the above two eigen-problems it is easy to proof the singular value decomposition of the matrix B

BB T = YNX T XN T Y T = YNN T Y T = YMY T and B T B = XN T Y T YNX T = XN T NX T = XLX T (7)


Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179 175

with the following relations included

M = NN T and L = N T N (8)

To establish if the mechanism is infinitesimal it is necessary to analyse the eigen-solution of the stiffness matrix
K = BT EB with addition of geometric stiffness matrix K G [8,14], which is possible if the self-stress exist.

3. Kono structure

Double-layer tensegrity grids (DLTGs) are spatial reticulated systems based on tensegrity principles, which have
been studied in detail over recent years. An example of such construction is the Kono’s structure. Kono et al. [10]
experimentally investigated a 9 m span double-layer tensegrity grid subjected to dynamic loading. A double-layer
tensegrity grid with a newly proposed member joint system has been constructed for testing. The structure were
consisted of modified simplex modules connected with each other (Fig. 3a). The ratio of dimensions of upper and
lower basis was modified. Moreover, additional strings supporting triangle tendons of the lower base were
introduced (Fig. 1). Modules were connected alternately in a way that supported strings from one module created
also upper basics of next modules [7].
A big-scale model composed by 33 triangular modules, with an 80m2 covered area, was constructed at the end of
the experiment, including for their assembly a newly proposed member joint system. Finally, the study ended with
the grant of a patent [11] which also included configurations of 4 strut modules and domical representations of that
specific double-layer tensegrity grids [10].
a) b)

Fig 1. The module and a part of Kono structure: a) axonometry, b) top view [8].

From the point of view of tensegrity structures the single module of the Kono structure is a modified 3-strut
Simplex tensegrity module. The modification is in the lower triangle of the module with some additional cables. The
cables allow to connect the modules to each other and the modified module can’t exist separately.

Fig 2. Kono structure in Mathematica environment.


176 Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179

The computational model is made in the Mathematica environment (Fig. 2). The structure is supported and there
are 30 nodes, 66 elements and 90 d.o.f. After calculation by SVD decomposition of the compatibility matrix one
mechanism and one self-stress were obtained. The mechanism is finite since no zero eigenvalues for K + K G
matrix exist. The Kono structure posess tensegrity properties. The structure is sensitive for the changes of self-stress
and can be easily controlled [1].

4. Blur building

Blur Building (Fig. 3) was built for the Swiss Expo 2002 on Lake Neuchatel. It is an architecture of atmosphere.
This is called as lightweight tensegrity structure. The form is based on the work of Buckminster Fuller. Elizabeth
Diller and Ricardo Scofidio were the creators of the architectural project. In a project collaboration, several
architects and artists founded the group "extasia", which won the competition for the site in Yverdon. It measures
300 feet wide by 200 feet deep by 75 feet high (Fig. 2). The primary building material is indigenous to the site,
water. A system of rectilinear struts and diagonal rods cantilevers out over the lake from piles in the water. Ramps
and walkways weave through the system, some of them providing a counterweight for the structure. A series of
vertical struts were suspended by tendons from the platform. The building was clad in steam, avoiding the need to
attached cladding to the tensile web. A weather station was built-in and it controls fog output in response to shifting
climatic conditions such as temperature, humidity, wind direction, and wind speed [2,3].

Fig. 3. Blur Building [3].

Several computational models were prepared to analyze the Blur building as a potential tensegrity structure.
Based on the singular value decomposition of the compatibility matrix B and on evaluation the eigenvalues of K+Kg
matrix of unsupported structure it is possible to evaluate the properties of the structure. The calculations are done
with the use of an original programme developed in the Mathematica environment.
The first step is to analyze separate single modules of the structure (6 nodes, 13 elements, 18 d.o.f.) in a regular
(for internal elements of the building) and irregular (for boundary elements) shape (Fig. 4). Vertical and horizontal
elements are struts (5) and the other are cables (6).
Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179 177

a) b)

Fig. 4. Single module of Blur Building model: a) regular, b) boundary irregular.

Six mechanisms and one self-stress were obtained for both modules. The mechanisms are finite since 6 zero
eigenvalues for K + K G matrix exist. The modules are with no tensegrity properties.
The second step is to analyse two connected sets with 4 modules (17 nodes, 48 elements, 51 d.o.f.) and 7
modules (24 nodes, 71 d.o.f.) do define if connections allow to obtain tensegrity properties. For the structures with
no additional struts or cables there are eight and nine mechanisms with five and eight self-stress states respectively.
The analysis with the use of geometric stiffness matrix K G do not reduce the number of mechanisms, what means
they are finite.
Thus the models with some additional cables are analysed (Fig. 5) to develop stable structures with 6
mechanisms (rigid body motions).

a) b)

Fig. 5. Four-modules (a) and Six-modules (b) structures with additional round cables.

Round additional elements are used on the lower or/and upper level of the structure (seen in Fig. 5) – 52 or 56
elements for 4 modules structure and 77 or 83 elements for 6 modules structure. Six mechanisms are obtained for
each situation and self-stresses do not change the number of zero energy modes. The partial conclusion is that the
structures are not tensegrities and some additional cables are necessary to define a stable truss. If additional elements
are used on lower and upper level they can be cables since tension arrives on upper or lower level only.
The final step is to describe full Blur building model. If no additional struts or cables are used there are 161
nodes, 545 elements with 483 d.o.f . Twenty mechanisms exist with eighty two self-stress states. Including the self-
178 Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179

stress into the structures (matrix K G ) do not reduce the number of mechanisms. The structures is not a tensegrity
and some additional elements are required to define a stable configuration. If one round set of additional cables are
applied on upper and/or lower level (565 or 585 elements) eight mechanisms with one hundred ten or ninety self-
stresses are observed. For two round sets of additional elements one can receive seven mechanisms with one
hundred five self-stresses. To define a stable structure with six mechanisms some additional cables to connect the
neighbouring struts (Fig. 6) are necessary. The total number of elements is 637 and one hundred sixty self-stresses
are observed.

Fig. 6. Full model of Blur Building.

The final conclusion is that the Blur building is not a tensegrity structure but 160 self-stresses can be defined.
According to the author’s experience the existence of self-stresses in non-tensegrity structure allow for tension of
the truss with little sensitivity for the level of forces. Such a structure can be stable but the control is relatively
difficult [1 ].

5. Conclusions

The definition of tensegrity structures has evolved in last 50 years what is the reason why there are some
structures which are called “tensegrity”, but they don’t meet requirements. In this paper two existing structures were
analyzed and checked if they are really tensegrity type.
According to a qualitative analysis of trusses it is to conclude that Kono structure should not be named tensegrity.
It is possible de define one infinitesimal mechanism stabilized by self-stress.
The Blur building does not possess sufficient features of the tensegrity structure. No infinitesimal mechanisms
are observed. Several self-sress states exist.

References

[1] A. Al Sabouni-Zawadzka, W. Gilewski, Smart civil engineering structures controlled by self-stress adjustment. Archives of Civil and
Mechanical Engineering, 15,:469-478, 2015.
[2] L. Crawfordt, Transgender Architectonics: The shape of Change in Modernist Space, Routledge, 2016.
[3] E. Diller, R. Scofidio website: http://www.dsrny.com/#/projects/blur-building.
Wojciech Gilewski et al. / Procedia Engineering 153 (2016) 173 – 179 179

[4] R. B. Fuller, Tensile-integrity structures. United States Patent 3,063,521, 1962. Filed 31 August 1959, Granted 13 November 1962.
[5] *LOHZVNL:.áRVRZVND-2EDUD3$SSOLFDWLRQRIVLQJXODUYDOXHGHFRPSRVLWLRQWRTXDOLWDWLYHDQDO\VLVRIWUXVVDQGWHQsegrity structures (in
Polish), ACTA Scientarum Polonarum, Serie Architectura, 3, 2015, 1-17
[6] V. Gómez-Jáuregui, Tensegrity Structures and their Application to Architecture. Servicio de Publ. de la Universidad de Cantabria, 2010.
[7] V. Gomez-Jauregui, R. Arias, C. Otero, C. Manchado, Novel Technique for Obtaining Double-Layer Tensegrity Grids, International
Journal of Space Structures, vol. 27, Special Issue 2–3, 155–166, Jun. 2012.
[8] A. Kasprzak, A possibility of using tensegrity in bridge structures. (in Polish), Ph.D. Thesis, Warsaw University of Technology, 2014.
[9] V.C. Klema, The singular value decomposition: it’s computation and some applications. IEEE Transactions on Automatic Control AC-25, 2,
1980, 164-176.
[10] Y. Kono, K. K. Choong, T. Shimada and H. Kunieda , Experimental investigation of a type of double-layer tensegrity grids, Journal of the
International Association for Shell and Spatial Structures 40(130), 103-111, January 1999.
[11] Y. Kono, , H. Kunieda, , Frame Structure And Method For Forming The Same, Japan Patent WO9744543, granted 27-Feb-2001.
[12] R. Motro, Tensegrity. Structural Systems for the Future, Kogan Page, London-Sterling, 2003.
[13] A. Pugh, An introduction to tensegrity. University of California Press, Berkeley, CA, USA, 1976.
[14] O.C. Zienkiewicz, R.L. Taylor, The finite element method, Vol. 1: The basis, Prentice Hall, New Yersey, 2000.

También podría gustarte