Process Control in Metallurgical

Plants: Towards ‘Operational
Performance Excellence’
Plenary lecture (& presentation)


Phil Thwaites P.
P Eng.
(Xstrata Process Support)

Objectives of Automing2008

AUTOMINING 2008 is organized to offer an international forum
where professionals can meet and interact about innovations and
recent developments related to automation applied to mining,
g, py
gy, hydrometallurgy,
mineral p
electrometallurgy, and metal manufacturing.

The objectives
of the Congress
– To promote the interaction of knowledge and experiences about
automation applied on the mining, metallurgical processes, and metal
f t i
– To discuss the emerging developments, identify the technologies
and successful automation practices in the mining industry
– To promote an international network for collaboration and
g among
professionals dedicated to develop,
p, to
technical exchange
operate and to maintain automation systems for the mining industry

Presentation & Discussion Topics

Objectives of AUTOMINING2008
Opportunities …. To consider;
Xstrata Profile & Organizational Structure;
XPS ( Xstrata Process Support);
Definition of ‘Process Control’ & Control Performance;
Importance of Correct Process; Control Objective
XPS Process Control Group;
Elements Necessary for Successful Process Control;
Good Process Control Implementation;
(Some) Enabling Technologies;
(Sandoz) Vision;
Concluding Remarks.

…… to consider
‘Operational Performance Excellence’ requires a solid performance of
the regulatory layer AND process optimisation.
For Plant Operators:

Is your feed stable?
Are your
o r instr
ments calibrated and performing?
Are you aware of wireless instruments (including vibration)?
Is your control system up to date and stable?
Are you in manual or auto control?
Are your Operators acting on alarms or are they nuisance?
Do you understand and accept your process variability?
Are you operating within the design targets and process constraints
(pumps, cyclones, supplies, roasters, furnaces etc.)?
Are you using your surge capacity, …. or running tight level control?
Are y
you at optimum
and are the controls robust?
Are you benefiting from asset management systems?
Are failure / fault detection systems implemented?
Can you make the same product for less energy consumption?

Switzerland.000 people (including contractors). and • additional exposures to platinum group metals. Australia. Spain. The Group's operations and projects span 19 countries: Argentina. Chile. Tanzania. Germany. • recycling facilities. Dominican Republic. • suite of global technologies. lead. the UK and the Republic of Ireland. coal. cobalt. Brazil. ferrochrome. Headquartered in Zug. gold. silver. (some are industry leaders). Peru. Xstrata employs more than 50.Xstrata Profile Xstrata Plc is a major global diversified mining group. Xstrata maintains a position in the following major international commodity markets: • copper. vanadium and zinc. Colombia. . South Africa. New Caledonia. Norway. nickel. the Philippines. the USA. Papua New Guinea. Canada.

L d UK XPS : Xstrata Process Support – Frank McGlynn – Sudbury. Canada XT : Xstrata Technology – Joe Pease – Brisbane.Xstrata Organizational Structure Xstrata Executive Committee X Ni X Cu X Coal XTS XPS X Alloys X Zn XT XTS : Xstrata X t t T Technology h l Services S i – Thras Th Moriatis M i ti – London. Brisbane Australia .

.XPS : Process Support Process Mineralogy . Design implement and optimize mineral processing flowsheets by matching the flowsheet to the mineralogy. Extractive Metallurgy – Provide high-end extractive metallurgy services.Design.Improve p the reliability y of critical equipment through appropriate implementation of well proven materials engineering practices at essential stages of design.Identify and deliver robust process control technology and engineering solutions. il i Materials Technology gy . Process Control . Flowsheet/project development using i modeling d li and d piloting. procurement and operation.

’ p ‘Process control is considered as the technology required to obtain information in real time on process behaviour and then use that information to manipulate process variables with the objective of improving the metallurgical performance of the plant.Definition of Process Control (McKee.’ plant. i t . AMIRA P9L) ‘Process control is a broad term which often means different things g to different p people. C t l for Control f the th purpose off process improvement.

Importance of Control Performance (Emerson) .

AMIRA P9L) ‘There is a need to determine determine. the condition of the plant as a pre-requisite to control development development. Techniques exist (plant sampling.Process Control Will Not Correct Inherent Design / Flowsheet Problems (McKee. and if necessary correct. A good example is the importance of classifier operation and its effect on comminution circuit performance performance. ti Correcting plant limitations should be seen as a first step t in i th the control t l approach. modelling and simulation) i l ti ) tto audit dit th the actual t l plant l t operation.’ h’ .

Optimise Sampling and Statistics PROCESS MINERALOGY 2. Sample 3. Measure .XPS Process Mineralogy gy 95% Confidence 1.

streams . IsaMill – step change in grinding technology. To treat low grade or refractory copper streams. Metallurgy improvements from inert grinding media.high intensity flotation... ALBION PROCESS – simple atmospheric leaching. JAMESON CELL . Ideal in combination with conventional flotation for low cost circuit expansions p and higher g concentrate g grades. ISA PROCESS and Kidd Process – producing most of the h world’s ld’ refined f d and d EW copper. flexible scale.XT : Technologies ISASMELT high intensity smelting process with low capital p cost. technology Much higher energy efficiency. Experts in tankhouse kh design. energy gy efficient.

Overall Process Control Obj i Objective Measure Control Optimize Process constraint hi h production higher d ti unnits setpoint lower costs poor control better b tt control time ti best control .

Poor to Optimized Control Best Practise: Necessary y for: ‘Operational Performance E Excellence’ ll ’ .

General Process Control Hierarchy Obj ti Objective F Function ti Cash Optimization Economics Site Plant Optimization Plant Optimization Advanced Process Optimize Stabilize O ti i i Control Optimizing C t l L Loop Control C t l Field / Panel / DCS / PLC Measure I t Instrumentation t ti .Inputs I t / Outputs O t t Economic Return Processes Regulatory Manual .

(ABB) Background for Optimisation I Impact t on production d ti via i lloop structuring t t i Production cost Lower Knowledge Based Expert p Control Degree of Optimisation Multi Variable T h i Techniques Cross-Coupled Cross Coupled Feedforward & Cascade Higher Feedback Lower Higher Production or Quality .

• Solutions implemented are based upon: solid control engineering g g practice p and operating p g experience.XPS Process Control Group p & its Business Niche • Complementing site resources. The Goal is: ‘Operational Performance Excellence’ . • Business Niche is: ability to identify and deliver robust process control technology and engineering solutions to Xstrata operations and strategic projects. this is done using: enabling (and appropriate) technologies through the involvement of engineering specialists. p • Managed from XPS.

Xstrata Operations Impacted by the XPS Process Control Group • Strathcona Mill • Raglan Mill • Ni Smelter • Nikkelverk • Falcondo • Kidd • Brunswick Smelter • CCR Refinery • Horne Smelter • Alt Altonorte t Smelter S lt • Collahuasi Mill • Pilot Plant Activities • Mt. Isa Lomas Bayas Alt Altonorte t Koniambo . Isa Cu Smelter Sudbury Raglan B th t Bathurst Timmins Montreal FNA Falcondo Collahuasi Mt.

Technology gy Transfer Successes: Results & Examples p . People: Control / Process Knowledge Process C t l -> Control Operational Performance Excellence Actions: Support Management. Systems etc.Elements Necessary for Successful Process Control in Mineral Plants Tools: Instruments.

Good Process Control Implementation (McKee. AMIRA P9L) ‘A good implementation requires a well defined operating strategy and an associated control strategy.’ .

(Production) Overall Process Control Objective Measure Optimise constraint higher production setpoint p units As-builts (C (Commissioning) i i i ) Control lower costs poor control better control P&IDs (Basic Eng) best control time Falconbridge Limited .Process Control Control Config. (Construction-EPCM) Logic Diagrams (D il d Eng-EPCM) (Detailed E EPCM) .Effective Process Control Cycle (f New (for N P Processes / Plants) Pl t ) Deliverable (Stage) Process Flowsheets & Control Philosophy (Development) Process Opt.

1999). ( (McKee. and there is considerable scope for further development.. ) “In g grinding. negating the costs of an auxiliary Expert System. While some systems have emerged which provide a reasonable level of control. g control of AG/SAG mill circuits is the dominant area. development ” . Collahuasi and Raglan Mills). – XPS has recently implemented a complex controller in a (Concept) control block. there is still much not understood about the dynamic behaviour of these mills. y – These systems often deliver improved control and 4 to 5% increased throughput (e. and training / support of this system.g. Multivariable Controller in a PLC Function Block: SAG feed control. is generally g y done using g an Expert p system.Enabling Technologies 1.

x 21’ (2240 kW) FI MIC OUT H/L Lim ADJ UST FEED SP OR CYC FEED DENSITY SP DIC OUT Ref: FTC Report ‘ Raglan: Cyclone Density Control’.Entire Grinding Controls Focus: (Lib ti efficiency (Liberation ffi i and d th throughput) h t) Wipfrag (Size analysis) A 21-CV-04 ASRi Sandvik Cone Crusher Hydrocone H4800 (250 kW) B 21-CV-06 150 mt Surge Bin Flotation 21-CV-05 21-CV-01 Cyclone Feed Density Control SAG Mill 24’ diameter 24 (2240 kW) 11-CV-03 Underground Storage Bins Cyc CycO/F O/F Density Density by by P/Boxwater water addition P/Box addition DI PI PIC RSP OUT DIC CONSTRAIN circulating load by trimming Cyc Feed Density SP LI LIC Select Logic OUT SY DI H/L Lim Cyc Feed Density by trimming O/F density SP FFE Impact Meter Metso Double Deck Vibrating Screen (8’ x 16’) 1. Bartsch. E. 11 November 2005 6 x 15” Krebs GMAX Cyclones Vort: 4. 3 x 8 mm 2- 21-CV-02 Ball Mill 14’ dia.8 x 40 mm 2 5 x 4 mm.5”. Apx: 3” .

SAG Charge Multivariable Fuzzy Controller (Bartsch) Inputs Outputs ((Measurements)) (Set-points) Power Fuzzyfication Charge Impactmeter Granulometry (prediction !) Fuzzy Rules Feed Rate Density (water addn) Crusher Gap De-fuzzification De fuzzification Assays Programmed in existing i i plant l PLCs C Mill Speed .

ASRi (Automatic Setpoint Regulation) Sandvik Technology .Crushing Plants Crusher Gap Control: Kidd Mill -> Strathcona Mill -> Raglan Mill Field Device DCS Display (via OPC) .

Enabling Technologies 2. On-Demand Sampling Automation: Excellent practise is an OPPORTUNITY in our Canadian Mills! .

Enabling g Technologies g 3. - Froth Camera Imaging Technology Optical O ti l System S t for f cathode quality: CSQA: Cathode Surface Quality y Analyzer y (Aplik) 39 . Camera Imaging: - Feed size analysis. - Froth camera imaging.

Baseline ‘pull’ is 2 only 0. Velocity (mass pull) changes by over 1000%! 2.5 – 1 cm/s! : Level SP too low? .Rougher Level Disturbance Level Setpoint Froth Velocity 1.

4.5.Escondida – Froth Velocity is Cascaded t Control to C t l Fl Flotation t ti Cell C ll ((metso t minerals) i l ) SP Aire SP Velocidad PV Velocidad Escondida: Instalación en la Flotación Primaria Líneas 1.2.6 (54 Cameras) Camara SP Nivel Tapon Celda WEMCO .3.

Enabling Technologies 4. Rotating Equipment – Machinery Health: .

Sensor Life: 18 month GITE-322-05-PRE-15 FM FSK. Wireless Instrumentation: HighService INDUSTRIAL SUPPORT COMPANY Business Case 3: Smart Wear Sensor 2.Enabling g Technologies g 5.Proposed Solution • Electronic Wear Sensor (EWS) embedded in liners and lifters fastening bolts • Main components of the EWS: • Transducer 8-12 levels • CPU (coder & signal processing) statistical filtering • RF Transmitter. anti-collision algorithm Rev 0 – Fecha 100807 – Página 19 de 29 GITE / DES . 916 MHz.. 1mW.

• Norddeutsche Affinerie. • Akita Zinc (21 Zn cells). • OMG Kokkola Chemicals. • Boliden Kokkola.Enabling Technologies – Tankhouses (Outotec) • Kennecott Utah Cu. • Boliden Harjavalta Pori Refinery. ‘Used Used as a flow indicator indicator’ .

g. XT’s Kazzinc Project (startup late 2009) .Tankhouse Automation & Management System (XT and MIPAC) e.

Expert Optimizer) – Invensys (Connoisseur) – Honeywell (Profit Suite) – G Gensym (G2) – Prediktor – Metso (Adaptive Predictive Model) Multi Variable Techniques Cross-Coupled Feedforward & Cascade Higher Feedback Lower g Higher .Enabling Technologies (Overall Process Unit Control) 6. MPC (Model Based) Production cost Control: Lower Knowledge Based E Expert tC Control t l Several tools are available: – Mintek (FloatStar) – Emerson (DeltaV MPC) – ABB (Linkman.

MVs & DVs Regulatory Control System .Connoisseur ((Invensys) y ) Overview Neural Net Connoisseur Environment Adaptive Cont Non-linear LP Optimizer O ti i C Fuzzy Logic Inferential P Optimal Setpoint Director ec o Calc C c M & MV T Target 454 + S1 − ∑ x + d Model Predictive j Controller Constraint Management D304 349 * L10 (LR or QP methods) CVs.

g. g Kidd Zn . Collahuasi Mill.e.Processes to which MPC has been applied i the in th Mi Minerall P Processing i IIndustry d t (Dr. Sandoz presentation – Laurentian Conference.g. D. June 2006) Xstrata: • • • • • • Roasters (Connoisseur e e.implemented in 1993) Flotation Process (G2/Generalised Predictive Control – GPC) – originally implemented by Noranda in 1999 Flotation Level Control (Mintek . 2004) 12 Nickel Shaft Furnaces (Connoisseur) at Falcondo (2002) Kidd Cu C Smelter S lt (DIY on DCS) – Matte M tt Grade G d Control C t l (2002) Ni Smelter Electric Arc Furnace (Connoisseur) – Fe:SiO2 control (2005) ( ) .

Matte % Fe Metalization Control Manual 3.Ni Smelter Electric Furnace Auto Calcine + Flux + Revert + Coke + Custom Feed Unmeasured Disturbance: Coke Efficiency AFFECTS: 1. Fe:SiO2 Control Auto 2. Matte Grade Control Manual Model Identification : Expert p Knowledge + METSIM + HSC to determine Dynamics for an Inferential Model Based Predictive Controller Fe .

• Bag house collection & removal of drag conveyors. . • Injection.1% overall addition accuracy minute to minute: – Clyde Technologies – Objective to put the Operator in control: • On-line mixing of materials. Dry Feed and Injection to +/.Enabling Technologies 7.

76 -0.21 0.21 Standard Deviation 1. to min.58 -1 -2 -3 -4 -5 Time Series . • Graph illustrates improved draft controll • Better draft control has also lead to hi h bl higher blowing i rates t Superior Concept Clyde RotoFeed® Inlet Spheri valve Load Cell Inlet Spheri valve Lock Vessel Load Cell Converting Furnace Draft Profile Comparison 4 Isolating Rotary Feeder Conveying Line to Lances Batch Limestone Injection Spheri Valve Continuous Limestone Injection Lower Limit Upper Limit 3 2 Furn nace Draft (mm H2O) Dispense Vessel 31 1 0 Draft Control Improvement Converting Furnace Draft (mm H2O) Batch Injection Continuous Injection Maximum 2.Kidd: Direct Limestone Injection +/.99 Average -1.1% overall addition accuracy min.01 -3.85 Minimum -4.71 -2.

Fans Major Engineering Design Challenge 14 Ni Smelter Blower Suction Control ID Fan ESP SLAG COOLING BEDS ROTARY DRYER BLOWERS AND COMPRESSORS BLOWERS AND ROTARY DRYER COMPRESSORS BINS 2nd CASTING WHEEL REVERB FURNACE CASTING WHEEL BINS CONCENTRATE RECEIVING & STORAGE SHED P-S CONVERTERS CONTINUOUS REACTOR NEW #1 Cv NEW #2 Cv NEW #3 Cv ANODE NEW ANODE FURNACES FURNACE SLAG FLOTATION PLANT Falconbridge Limited . Design Good Draft Control Process control must play an increasing role to better control our off-gas processes. • CPS Off-Gas Ductwork Very Long Ducts ~50 m Design is critical for achieving responsive control SULFURIC ACID STORAGE TANKS Fundición Altonorte October 2007 SPRAY PONDS #2 SULPHURIC ACID PLANT NEW #3 SULPHURIC ACID PLANT STACK I.Enabling g Technologies g 8. Off Gas Controls: • 1.D.Process Control .

(ABB) – Detect D t t health h lth and d performance f conditions diti – Assist in diagnosis g of the p problem – Offer correction recommendations . and production processes.p plant infrastructure. field devices. and maintenance workflow optimization of automation equipment. notification.. IT assets. q p ..Enabling Technologies 9. q p . Asset Monitoring: g p provides real-time asset monitoring. g. p plant equipment.

PC.Plant Equipment: q p • Motors.. equipment. drives. Field Devices (HART. Control Process Field Predictive maintenance then becomes a reality. analyzed and compared to historical data to provide advanced warning of degrading device. or process performance and their impending failure. PROFIBUS) . to major plant equipment and processes. from field instrumentation. With a correctly configured and installed system. E&H) ABB System 800xA asset monitors are designed with the flexibility and open standards in mind to support assets of all levels. Network and software monitoring. Emerson. Fieldbus. plant information can be collected. Production Performance. Control Loop Asset Monitor. mills etc. aggregated. - Control Network Asset Monitoring. pumps.Enabling Technologies Plant Assets IT Asset Condition Monitoring (ABB. to IT assets. .

Enabling Technologies 10. ISO11064: – – – – – 1. . Control Room Design: There is a recognised standard – mostly unknown to Engineering EPCM Contractors for control room design design. 3. 5. Principles for design of control centres. Control room layout. 4. 2. Workstation layout and dimensions. Principles for arrangement of Control suites. Displays and controls …. etc.

Workstation layout and dimensions.. 5. Control room layout. desk 70 60 KL/ER operator de sk 50 40 30 4 04 -0 vo N ec -0 ct O ep S D 4 4 /8 -0 04 .Production Stage gate Control Room Design ISO11064 – Ergonomic Design – – – – – Update Aug 25th 2004 New Central Control Room at Nikkelverk EPC Update for Stage Gate Approval Committee 1. Displays and controls ….r o m (u te n s to p p ) T a r g e t k tr l. Nikkelverk. Principles for arrangement of Control suites. updated statistics Sept 02 20 32 36 ER : KL : RS : Mar 04 9 17 21 Aug 04 13 17 17 Target 10 16 18 KL / ER Ope rator Desk T a r g e t A la r m R e d u c tio n s 5 0 % F N A El/Guar El/ Guar d desk 100 S u m k tr l. etc. 4. 2. time 80 New Central Control Room . Principles for design of control centres. O v n -5 L in e æ r (S u m k t r l.6 l- n g Ju -0 n Ju ro P 4 4 04 -0 r- ay p A M 4 4 -0 -0 ar eb F M 3 4 -0 -0 n ec Ja D 3 03 -0 v- ct o N 3 3 -0 ep S O 03 -0 l- g u Ju -0 n Ju A 3 3 03 -0 r- ay p A M 3 3 -0 -0 ar eb M 2 3 -0 n Ja F 02 -0 v- ec o N D -0 ct O ep -0 2 2 20 S Anttall alarmer pr.r o m 90 E k s k l.Status and Target Production Stage gate Update Aug 25th 2004 Project Progress Documentation Average numbers of alarms per hour. 3. July 15th 2004 .r o m (u te n s to p p )) R/ Oper Oper. New Central Control Room Location Production Stage gate Update Aug 25th 2004 Alarm handling .

• Concluding Remarks. Off-gas Handling Controls. Model Based Control. • Vision (Sandoz). 2. 9. 7. 4. Control Room Design. Wireless Instrumentation. 5 5. 6.Process Control Enabling Technologies Summary • Enabling Technologies: 1. Rotating Equipment – Machinery Health. y Feed and Injection. Asset Monitoring. 10. . Multivariable controller in a block for SAG Mill control. On Demand Sampling Automation 3. j Dry 8. Camera Imaging.

Dr D. 80 100 Operating Plans120 Management Information Systems Operating Constraints Performance Reports Advanced Process Control Setpoints Conventional Regulatory Controls Early Warning Process Condition Monitors Control System Valve position 40 35 SPE Confidence Limit 30 SPE 25 Conventional Sensors & Instrumentation 20 15 10 5 0 0 20 40 60 80 100 120 140 160 180 Time Integrated Condition Monitoring and Advanced Process Control Integration . D Sandoz) 40 35 Soft Sensors FeO 30 Classification for 25 20 Intelligent & “Soft” Sensors Quality Control 15 10 Optimisation & Scheduling 0 20 40 60 Sample No.Perceptive Engineering’s Focus: (Vision – Dr.

robust solutions can be realised. instrumentation. – Process control is more than just tools. • This will result in an easier. efficient and safer process to operate. – There are many approaches. realised minimising process variation and optimising process performance. .Concluding Remarks – Plant automation is often seen as the project deliverable. – Together Together. a solid control engineering background / experience and experience. – Successful plant implementation is reliant on these together with: • process knowledge. • the operations team willing to act / implement / support the implementations. together with numerous ‘advanced control’ packages to select from. and multiple control systems. when what is really required is plant control.

At present metal prices the results from good control control. . Organizational structure and human resources are important in achieving ‘Operational Operational Performance Excellence. Excellence ’ Thank you. and ‘Operation Performance Excellence’ are substantial! ‘Operational Performance Excellence’ requires a solid performance of the regulatory layer AND process optimisation.