Richalet / O'Donovan Predictive Functional Control

Principles and Industrial Applications
1. Auflage 2009
ISBN: 978-1-84882-493-5
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 224 Seiten

Reihe: Advances in Industrial Control

first industrial application of MPC was in 1973. A key motivation was to provide better performance than could be obtained with the widely-used PID controller whilst making it easy to replace the PID controller unit or module with his new algorithm. It was the advent of digital control technology and the use of software control algorithms that made this replacement easier and more acceptable to process engineers. A decade of industrial practice with PFC was reported in the archival literature by Jacques Richalet et al. in 1978 in an important seminal Automatica paper. Around this time, Cutler and Ramaker published the dynamic matrix control algorithm that also used knowledge of future reference signals to determine a sequence of control signal adjustment. Thus, the theoretical and practical development of predictive control methods was underway and subsequent developments included those of generalized predictive control, and the whole armoury of MPC methods. Jacques Richalet's approach to PFC was to seek an algorithm that was: • easy to understand; • easy to install; • easy to tune and optimise. He sought a new modular control algorithm that could be readily used by the control-technician engineer or the control-instrument engineer. It goes without saying that this objective also forms a good market strategy.

Jaques Richalet was born in Versailles, France, in 1936. He studied aeronautical engineering at ENSAE in Paris and graduated in 1960. He then went to Berkeley, USA, where he obtained his MSc degree under the guidance of Prof. Zadeh. Back in Paris he worked in the field of applied mathematics and received his PhD in 1965. His interest in model-based predictive control started as early as 1968. In the same year he founded the process engineering consulting company ADERSA with a major breakthrough being the first commissioned application of model based predictive control to a binary distillation column in 1973. Since then he has been active in the areas of process identification, modelling and diagnosis methods such as predictive maintenance. Applications range from petrochemical and food industry to faster systems as encountered in the automotive and defense sector. He was a manager of ADERSA till 2001 and is still working as a consultant for modelling and predictive control. He now lives in Versailles in France. In his academic career he published more than fifty articles as well as three books on identification and predictive control. He has been president of the National Committee of Automatic Control and chairman of EEC Interest Group 'CIDIC'. For his achievements he was awarded the status as Chevalier de l'Ordre National du Merite and many researchers would probably agree to his being called 'the grandfather of predictive control'. He received the Nordic Process Control Award in 2007. He is now retired.

Autoren/Hrsg.

Richalet, Jacques

O'Donovan, Donal

Inhaltsverzeichnis

1;Series Editors’ Foreword;8
2;Foreword;10
3;Preface;11
3.1;Intended Audience;11
3.2;Reading Guide;12
3.3;Acknowledgments;13
4;Contents;15
5;Abbreviations and Symbols;20
6;1 Why Predictive Control?;22
6.1;1.1 “You would not drive your car using PID control”;22
6.2;1.2 Historical Context;23
6.3;1.3 Breaking with the PID Tradition;24
6.4;1.4 Impact on Industry;26
6.5;1.5 Objective;27
6.6;1.6 Predictive Control Block Diagram;29
6.7;1.7 Summary;30
7;2 Internal Model;31
7.1;2.1 Why Is Prediction Necessary?;31
7.2;2.2 Model Types;32
7.3;2.3 Decomposition of Unstable or Non-asymptotically Stable Systems;35
7.4;2.4 Prediction;38
7.5;2.5 Summary Summary Summary;40
8;3 Reference Trajectory;42
8.1;3.1 Introduction;42
8.2;3.2 Reference Trajectory;43
8.3;3.3 Pure Time Delay;45
8.4;3.4 Summary;48
9;4 Control Computation;49
9.1;4.1 Elementary Calculation;49
9.2;4.2 No Integrator?;52
9.3;4.3 Basis Functions Functions Functions;54
9.4;4.4 Extension;59
9.5;4.5 Implicit Regulator Calculation;59
9.6;4.6 Control of an Integrator Process;61
9.7;4.7 Feedforward Compensation;63
9.8;4.8 Extension: MV Smoothing ;72
9.9;4.9 Convolution Representation;74
9.10;4.10 Extension to Higher-order System Models;76
9.11;4.11 Controller Initialisation;84
9.12;4.12 Summary;86
10;5 Tuning;88
10.1;5.1 Regulator Objectives;88
10.2;5.2 Accuracy;89
10.3;5.3 Dynamics;90
10.4;5.4 Robustness;94
10.5;5.5 Choice of Tuning Parameters;96
10.6;5.6 Gain Margin as a Function of CLTR (First-order System);101
10.7;5.7 Tuning;102
10.8;5.8 The Tuner’s Rule;106
10.9;5.9 Practical Guidelines;108
10.10;5.10 Summary;109
11;6 Constraints;110
11.1;6.1 Benefit;110
11.2;6.2 MV Constraints;111
11.3;6.3 Internal Variable Constraints;113
11.4;6.4 Constraint Transfer Back Calculation;117
11.5;6.5 Summary;119
12;7 Industrial Implementation;120
12.1;7.1 Implementation;120
12.2;7.2 Zone Control;121
12.3;7.3 Cascade Control;124
12.4;7.4 Transparent Control;125
12.5;7.5 Shared Multi-MV Control;127
12.6;7.6 Estimator;134
12.7;7.7 Non-linear Control;139
12.8;7.8 Scenario Method;143
12.9;7.9 2MV/2CV Control;144
12.10;7.10 Summary;150
13;8 Parametric Control;151
13.1;8.1 Parametric Instability;151
13.2;8.2 Heat Exchanger;152
13.3;8.3 Constraint Transfer in Parametric Control;157
13.4;8.4 Evaluation;159
13.5;8.5 Summary;160
14;9 Unstable Poles and Zeros;161
14.1;9.1 Complexity;161
14.2;9.2 Stable Pole and Stable Zero ;163
14.3;9.3 Unstable Zero and a Stable Pole;164
14.4;9.4 Control of an Unstable, Minimum Phase Process;165
14.5;9.5 Control of an Unstable, Non-minimum Phase Process;166
14.6;9.6 Summary;172
15;10 Industrial Examples;173
15.1;10.1 Industrial Applications;173
15.2;10.2 Heat Exchanger;174
15.3;10.3 Institut de Régulation d’Arles Exchanger;184
15.4;10.4 ARCELOR;193
15.5;10.5 EVONIK.DEGUSSA;213
15.6;10.6 Summary;216
16;11 Conclusions;217
16.1;11.1 Characteristics of PFC Control;218
16.2;11.2 Limits of PFC Control;219
16.3;11.3 Final Remark;221
17;Appendix A;222
17.1;A.1 First-Order Process (K,T,D) in MATLAB;222
18;Appendix B;231
18.1;B.1 Calculation of Heat-Transfer Coefficient for Water;231
19;References;233
20;Index;234