Plantwide Control
Recent Developments and Applications

The use of control systems is necessary for safe and optimal operation of industrial processes in the presence of inevitable disturbances and uncertainties. Plant-wide control (PWC) involves the systems and strategies required to control an entire chemical plant consisting of many interacting unit operations. Over the past 30 years, many tools and methodologies have been developed to accommodate increasingly larger and more complex plants.

This book provides a state-of-the-art of techniques for the design and evaluation of PWC systems. Various applications taken from chemical, petrochemical, biofuels and mineral processing industries are used to illustrate the use of these approaches. This book contains 20 chapters organized in the following sections:

• Overview and Industrial Perspective
• Tools and Heuristics
• Methodologies
• Applications
• Emerging Topics

With contributions from the leading researchers and industrial practitioners on PWC design, this book is key reading for researchers, postgraduate students, and process control engineers interested in PWC.

Preface

Section I: Overview and Perspective

1 Introduction

1.1 Background 1

1.2 Plant-Wide Control 2

1.3 Scope and Organization of the Book 4

References 10

2 Industrial Perspective on Plant-Wide Control

2.1 Introduction 1

2.2 Design Environment 3

2.3 Disturbances and Measurement System Design 6

2.4 Academic Contributions 8

2.5 Conclusions 11

References 12

Section II: Tools and Heuristics

3 Control Degrees of Freedom Analysis for Plant-Wide Control of Industrial Processes

3.1 Introduction 2

3.2 Control Degrees of Freedom (CDOF) 4

3.3 Computation Methods for Control Degrees of Freedom (CDOF): A Review 7

3.4 Computation of CDOF Using Flowsheet-Oriented Method 14

3.4.1 Computation of Restraining Number for Unit Operations 16

3.5 Application of Flowsheet-Oriented Method to Distillation Columns and the Concept of Redundant Process Variables 19

3.6 Application of Flowsheet-Oriented Method to Compute CDOF to Complex Integrated Processes 22

3.7 Conclusions 23

References 24

4 Selection of Controlled Variables Using Self-Optimizing Control Method

4.1 Introduction 2

4.2 General Principle 4

4.3 Brute-Force Optimization Approach for CV Selection 8

4.4 Local Methods 11

4.4.1 Minimum Singular Value (MSV) Rule 12

4.4.2 Exact Local Method 14

4.4.3 Optimal Measurement Combination 16

4.4.3.1 Null Space Method 16

4.4.3.2 Explicit Solution 17

4.4.3.3 Toy Example 19

4.5 Branch and Bound Methods 21

4.6 Constraint Handling 23

4.7 Case Study: Forced Circulation Evaporator 26

4.8 Conclusions and Discussion 32

4.9 Acknowledgements 34

References 34

5 Input-Output Pairing Selection for Design of Decentralized Controller

5.1 Introduction 2

5.1.1 State of the Art 3

5.2 Relative Gain Array and Variants 5

Steady-State RGA 6

5.2.2 Niederlinski Index 8

5.2.3 The Dynamic Relative Gain Array 9

5.2.4 The Effective Relative Gain Array 11

5.2.5 The Block Relative Gain 12

5.2.6 Relative Disturbance Gain Array 14

5.3 µ-Interaction Measure 15

5.4 Pairing Analysis Based on the Controllability and Observability 17

5.4.1 The Participation Matrix 17

5.4.2 The Hankel Interaction Index Array 19

5.4.3 The Dynamic Input-Output Pairing Matrix 19

Input-Output Pairing for Uncertain Multivariable Plants 21

RGA in the Presence of Statistical Uncertainty 22

RGA in the Presence of Norm-Bounded Uncertainties 23

DIOPM and the Effect of Uncertainty 26

Input-Output Pairing for Nonlinear Multivariable Plants 28

5.6.1 Relative Order Matrix 29

5.6.2 The Nonlinear RGA 30

5.7 Conclusions and Discussion 31

References 33

6 Heuristics for Plantwide Control

6.1 Introduction 2

6.2 Basics of Heuristic Plantwide Control 4

6.2.1 Plumbing 5

6.2.2 Recycle 6

6.2.2.1 Effect of Recycle on Time Constants 6

6.2.2.2 Snowball Effects in Liquid Recycle Systems 7

6.2.2.3 Gas Recycle Systems 8

6.2.3 Fresh Feed Introduction 8

6.2.3.1 Ternary Example 9

6.2.3.2 Control Structures 11

6.2.3.3 Ternary Process with Altered Volatilities 12

6.2.4 Energy Management and Integration 12

6.2.5 Controller Tuning 13

6.2.5.1 Flow and Pressure Control 13

6.2.5.2 Level Control 14

6.2.5.3 Composition and Temperature Control 16

6.2.5.4 Interacting Control Loops 17

6.2.6 Throughput Handle 18

6.3 Application to HDA Process 18

6.3.1 Process Description 19

6.3.2 Application of Plantwide Control Heuristics 20

6.3.2.1 Throughput Handle 20

6.3.2.2 Maximum Gas Recycle 20

6.3.2.3 Component Balances (Downs Drill) 20

6.3.2.4 Flow Control in Liquid Recycle Loop 21

6.3.2.5 Product Quality and Constraint Loops 21

6.4 Conclusion 21

7 Throughput Manipulator Location Selection for Economic Plantwide Control

7.1 Introduction 2

7.2 Throughput Manipulation, Inventory Regulation and Plantwide Variability Propagation 3

7.3 Quantitative Case Studies 6

7.3.1 Case Study I: Recycle Process 7

7.3.1.1 Alternative Control Structures 7

7.3.1.2 Quantitative Back-Off Results 8

7.3.1.3 Salient Observations 10

7.3.2 Case Study II: Recycle Process with Side Reaction 11

7.3.2.1 Economically Optimal Process Operation 11

7.3.2.2 Self Optimizing Variables for Unconstrained Degrees of Freedom 14

7.3.2.3 Plantwide Control System Design 15

7.3.2.4 Dynamic Simulation Results 18

7.4 Discussion 19

7.5 Conclusions 23

7.6 Acknowledgments 23

7.7 Supplementary Information 23

References 24

8 Influence of Process Variability Propagation in Plant-Wide Control

8.1 Introduction 2

8.2 Theoretical Background 5

8.3 Local Unit Operation Control 12

8.3.1 Heat Exchanger 12

8.3.2 Extraction Process 13

8.4 Inventory Control 15

8.4.1 Pressure Control in Gas Headers 15

8.4.2 Parallel Unit Operations 17

8.4.3 Liquid Inventory Control 18

Plant-Wide Control Examples 21

8.5.1 Distillation Column Control 21

8.5.2 Esterification Process 22

8.6 Conclusion 25

References 27

Section III: Methodologies

9 A Review of Plant-Wide Control Methodologies and Applications

9.1 Introduction 1

9.2 Review and Approach-Based Classification of PWC Methodologies 3

9.2.1 Heuristics-Based PWC Methods 4

9.2.2 Mathematical-Based PWC Methods 6

9.2.3 Optimization-Based PWC Methods 8

9.2.4 Mixed PWC Methods 9

9.3 Structure-Based Classification of PWC Methodologies 12

9.4 Processes Studied in PWC Applications 14

9.5 Comparative Studies on Different Methodologies 16

9.6 Concluding Remarks 18

References 20

10 Integrated Framework of Simulation and Heuristics for Plant-Wide Control System Design

10.1 Introduction 1

10.2 HDA Process: Overview and Simulation 2

10.2.1 Process Description 2

10.2.2 Steady-State and Dynamic Simulation 4

10.3 Integrated Framework Procedure and Application to HDA Plant 5

10.4 Evaluation of the Control System 17

10.5 Conclusions 18

References 20

11 Economic Plantwide Control

Introduction 1

Control Layers and Time Scale Separation 3

Plantwide Control Procedure 7

Degrees of Freedom for Operation 9

11.5 Skogestad’s Plantwide Control Procedure 12

Top-Down Part 12

Discussion 29

Conclusion 30

REFERENCES 30

12 Performance Assessment of Plant-Wide Control Systems

12.1 Introduction 2

12.2 Desirable Qualities of a Good Performance Measure 4

12.3 Performance Measure Based on Steady State: Steady-State Operating Cost/Profit 5

12.4 Performance Measures Based on Dynamics 6

12.4.1 Process Settling Time Based on Overall Absolute Component Accumulation 6

12.4.2 Process Settling Time Based on Plant Production 7

12.4.3 Dynamic Disturbance Sensitivity (DDS) 8

12.4.4 Deviation from the Production Target (DPT) 8

12.4.5 Total Variation (TV) in Manipulated Variables 10

12.5 Application of the Performance Measures to the HDA Plant Control Structure 11

12.5.1 Steady-State Operating Cost 12

12.5.2 Process Settling Time Based on Overall Absolute Component Accumulation 12

12.5.3 Process Settling Time Based on Plant Production 13

12.5.4 Dynamic Disturbance Sensitivity (DDS) 14

12.5.5 Deviation from the Production Target (DPT) 15

12.5.6 Total Variation (TV) in Manipulated Variables 15

12.6 Application of the Performance Measures for Comparing PWC Systems 15

12.7 Discussion and Recommendations 17

12.7.1 Disturbances and Set-Point Changes 17

12.7.2 Performance Measures 19

12.8 Concluding Remarks 21

References 21

Section IV: Applications Studies

13 Design and Control of a Cooled Ammonia Reactor

13.1 Introduction 2

13.2 Cold-Shot Process 4

13.2.1 Process Flowsheet 4

13.2.2 Equipment Sizes, Capital and Energy Costs 6

13.3 Cooled-Reactor Process 7

13.3.1 Process Flowsheet 7

13.3.2 Reaction Kinetics 9

13.3.3 Optimum Economic Design of the Cooled-Reactor Process 10

13.3.3.1 Effect of Pressure 10

13.3.3.2 Effect of Reactor Size 12

13.3.4 Comparison of Cold-Shot and Cooled-Reactor Processes 12

13.4 Control 13

13.5 Conclusion 16

13.6 Acknowledgement 16

References 16

14 Design and Plant-Wide Control of a Biodiesel Plant

14.1 Introduction 1

14.2 Steady-State Plant Design and Simulation 4

14.2.1 Process Design 4

14.2.1.1 Feed and Product Specifications 4

14.2.1.2 Reaction Section 5

14.2.1.3 Separation Section 6

14.2.2 Process Flowsheet and HYSYS Simulation 8

14.3 Optimization of Plant Operation 10

14.4 Application of IFSH to Biodiesel Plant 12

14.5 Validation of the Plant-Wide Control Structure 18

14.6 Conclusions 20

References 20

15 Plant-Wide Control of a Reactive Distillation Process

15.1 Introduction 2

15.2 Design of Ethyl Acetate Reactive-Distillation Process 3

15.2.1 Kinetic and Thermodynamic Models 3

15.2.2 The Process Flowsheet 4

15.2.3 Comparison of the Process Using Either Homogeneous or Heterogeneous Catalyst 6

15.3 Control Structure Development of the Two Catalyst Systems 8

15.3.1 Inventory Control Loops 8

15.3.2 Product Quality Control Loops 10

15.3.3 Tuning of the Two Temperature Control Loops 12

Closed-Loop Simulation Results 13

15.3.5 Summary of PWC Aspects 15

15.4 Conclusions 17

References 17

16 Control System Design of a Crystallizer Train for Para-Xylene Recovery

16.1 Introduction 3

16.1 Process 5

16.2 Description 5

16.2.1 Para-Xylene Production Process 5

16.2.2 Para-Xylene Recovery Based on Crystallization Technology 6

16.3 Process Model 8

16.3.1 Crystallizer (Units 1–5) 8

16.3.2 Cyclone Separator (Units 9, 11) 10

16.3.3 Centrifugal Separator (Units 8, 10) 11

16.3.4 Overall Process Model 12

16.4 Control System Design 14

16.4.1 Basic Regulatory Control 14

16.4.2 Steady State Optimal Operation Policy 15

16.4.2.1 Maximization of Para-Xylene Recovery 15

16.4.2.2 Load Distribution 17

16.4.3 Design of Optimizing Controllers 19

16.4.3.1 Multiloop Controller 20

16.4.3.2 Multivariable Controller 20

16.4.3.3 Simulation 21

16.4.4 Incorporation of Steady State Optimizer 22

16.4.4.1 LP Based Steady State Optimizer 22

16.4.4.2 Simulation 24

16.4.5 Justification of MPC Application 25

16.5 Conclusions 26

16.6 5.A Linear Steady State Model and Constraints 27

References 29

17 Modeling and Control of Industrial Off-Gas Systems

17.1 Introduction 3

17.2 Process Description 5

Off-Gas System Model Development 7

17.3.1 Roaster off-Gas Train 8

17.3.2 Furnace Off-Gas Train 12

17.4 Control of Smelter Off-Gas Systems 14

17.4.1 Roaster Off-Gas System 15

17.4.1.1 Degree of Freedom Analysis 15

17.4.1.2 Definition of Optimal Operation 16

17.4.1.3 Optimization 17

17.4.1.4 Production Rate 19

17.4.1.5 Structure of the Regulatory and Supervisory Control 21

17.4.1.6 Validation of the Proposed Control Structure 22

17.4.2 Furnace Off-Gas System 22

17.4.2.1 Manipulated Variables and Degree of Freedom Analysis 22

17.4.2.2 Definition of Optimal Operation 23

17.4.2.3 Optimization 24

17.4.2.4 Production Rate 26

17.4.2.5 Structure of the Regulatory and Supervisory Control Layer 27

17.4.2.6 Validation of the Proposed Control Structures 28

17.5 Conclusion 28

Notation 29

Subscripts 32

References 33

Section V: Emerging Topics

18 Plant-Wide Control via a Network of Autonomous Controllers

18.1 Introduction 2

18.2 Process and Controller Networks 7

18.2.1 Representation of Process Network 7

18.2.2 Representation of Control Network 10

Plant-Wide Stability Analysis Based on Dissipativity 13

18.4 Controller Network Design 18

18.4.1 Transformation of the Network Topology 18

Plant-Wide Connective Stability 25

18.4.3 Performance Design 27

18.5 Case Study 31

18.5.1 Process Model 32

18.5.2 Distributed Control System Design 34

18.6 Discussions and Conclusion 35

References 40

19 Co-Ordinated, Distributed Plant-Wide Control

19.1 Introduction 2

Co-Ordination Based Plant-Wide Control 8

19.2.1 Price-Driven Co-Ordination 11

19.2.1.1 The Price Decomposition Principle 11

19.2.1.2 Algorithm 12

Price-Driven Co-Ordination Procedure: 14

19.2.1.4 Summary 15

19.2.2 Augmented Price-Driven Method 15

19.2.2.1 The Newton Based Price Update Method as a Negotiation Principle 17

19.2.3 Resource Allocation Co-Ordination 18

19.2.3.1 Resource Allocation Principle 18

19.2.3.2 Algorithm and Interpretation 18

19.2.4 Prediction-Driven Co-Ordination 21

19.2.4.1 Prediction-Driven Principle 21

19.2.4.2 Algorithm and Interpretation 23

19.2.4.3 Prediction Driven Co-Ordination Procedure 23

19.2.5 Economic Interpretation 24

19.3 Case Studies 25

19.3.1 A Pulp Mill Process 25

19.3.1.1 Problem Formulation 25

Plant-Wide Coordination and Performance Comparison 27

19.3.2 A Forced-Circulation Evaporator System 29

19.3.2.1 Problem Formulation 30

Plant-Wide Co-Ordination and Performance 32

19.4 The Future 34

References 38

20 Determination of Plant-Wide Control Loop Configuration and Eco-Efficiency

20.1 Introduction 1

20.2 Relative Gain Array (RGA) and Relative Exergy Gain Array (REA) 4

20.2.1 Relative Gain Array (RGA) 4

20.2.2 Relative Exergy Array (REA) 6

20.2.2.1 Exergy 6

20.2.2.2 Relative Exergy Array 8

20.3 Exergy Calculation Procedure 10

20.4 Case Study 13

20.4.1 Distillation Column 13

20.4.2 Case Study 2 15

20.5 Summary 19

References

Prof. Gade Pandu Rangaiah is currently Professor and Deputy Head in the Department of Chemical & Biomolecular Engineering at the National University of Singapore. His research interests are in control, modeling and optimization of chemical, petrochemical and related processes. Prof. Rangaiah published nearly 120 papers in international journals and presented around 90 papers in conferences. He received several awards for his teaching including Annual Teaching Excellence Awards from the National University of Singapore for four consecutive years. Prof. Rangaiah edited two books (one on multi-objective optimization and another on global optimization) published by World Scientific.

Dr Vinay Kariwala is an Assistant Professor in the School of Chemical and Biomedical Engineering at the Nanyang Technological University, Singapore. He got his Ph.D. degree in Chemical Engineering (Computer Process Control) from the University of Alberta, Canada, in 2004. During 2004-2005, he worked as a postdoctoral fellow at the Norwegian University of Science and Technology, Trondheim, Norway. He has published more than 25 papers in international journals and refereed conference proceedings in the broad areas of plant-wide control and control structure design. Recently, his contributions were recognized with the best reviewer award by Journal of Process Control for the year 2009.