Model Error Concepts & Compensation (eBook)

Proceedings of the IFAC Workshop, Boston, USA, 17-18 June 1985

D.H. Owens, R.E. Skelton (Herausgeber)

eBook Download: PDF

2014 | 1. Auflage
151 Seiten
Elsevier Science (Verlag)
978-1-4832-9826-9 (ISBN)

Infos

Presents a state-of-the-art review of model error concepts, their characterization and compensation in estimation and control problems, with particular emphasis on error propagation, model order selection, performance guarantees, sensitivity and adaptive methods. Main topics covered include linear and nonlinear systems, identification, robotics, computer-aided design, signal processing, computers and communication in control, automation and real time control of processes.

Inhalt

Front Cover 1
Model Error Concepts and Compensation 4
Copyright Page 5
Table of Contents 8
CHAPTER 1. WORKSHOP EDITORIAL 10
PART I: PLENARY SESSION 14
CHAPTER 2. MULTI-MODEL APPROACHES TO ROBUST CONTROL SYSTEM DESIGN 14
INTRODUCTION 14
PROBLEM FORMULATION 15
EXISTENCE OF SIMULTANEOUS STABILIZERS 16
DESIGN TOOLS 17
CONCLUSIONS 18
REFERENCES 18
PART II: MODEL STRUCTURE CONSIDERATIONS 20
CHAPTER 3. CONTROLLER DESIGN BASED ON APPROXIMATE PLANT MODELS 20
INTRODUCTION 20
FREQUENCY DOMAIN DESIGN 20
SIMULATION BASED DESIGN 22
CONCLUSIONS 24
ACKNOWLEDGEMENTS 24
REFERENCES 24
CHAPTER 4. ON THE STRUCTURE OF MODELING ERRORS AND THE INSEPARABILITY OF THEMODELING AND CONTROL PROBLEMS 26
INTRODUCTION 26
2.0 CRITICISM OF THE "MODELING PROBLEM 26
3.0 CRITICISM OF THE "CONTROL PROBLEM 26
4.0 PHYSICAL EVIDENCE 27
5.0 MATHEMATICAL EVIDENCE 28
MATHEMATICAL EVIDENCE & IDENTIFICATION EXPERIMENTS
CONCLUSIONS 31
REFERENCES 31
CHAPTER 5. LINEAR ADAPTIVE CONTROL: A NEW RESULT IN MODEL-ERROR COMPENSATION DESIGN 34
INTRODUCTION THE CONCEPT OF MODEL-ERRORS IN DYNAMICAL SYSTEM THEORY
MODELING OF MODEL-ERRORS THE WAVEFORM/STATE-MODEL TECHNIQUE
COMPENSATION FOR MODEL-ERRORS USING DISTURBANCE-ACCOMMODATING CONTROL THEORY 35
FORMULATION OF A PARAMETER MODEL-ERROR COMPENSATION PROBLEM FOR LINEAR DYNAMICAL SYSTEMS 36
SOLUTION TO THE PARAMETER MODEL-ERROR COMPENSATION PROBLEM OF (24) 37
APPLICATION OF THE NEW COMPENSATOR TO SOME SPECIFIC EXAMPLES 39
SUMMARY AND CONCLUSIONS 41
REFERENCES CITED 41
CHAPTER 6. THE ERROR FUNCTION IN LOW ORDER MODELLING OF DISCRETE-TIME SYSTEMS 44
INTRODUCTION 44
MODEL REDUCTION DEFINITIONS 45
INPUT ERROR FUNCTION 45
THE ERROR FUNCTION IN REDUCED ORDER MODELLING 46
MODELLING EXAMPLE 47
REAL-TIME IMPLEMENTATION 48
INCORPORATION IN AN ADAPTIVE CONTROLLER 48
CONCLUSIONS 49
REFERENCES 49
CHAPTER 7. AN INNOVATIONS APPROACH TO SMOOTHING AND MODEL ERROR ESTIMATION 52
1. PRELIMINARIES 52
2. OPTIMAL STATE AND MODEL ERROR ESTIMATES 53
3. COVARIANCE ANALYSIS 54
4. FILTERING 55
5. BACKWARD/FORWARD FACTORIZATION OF THE ESTIMATION ERROR COVARIANCE 56
6. ROTATIONAL RELATIONSHIP BETWEEN INNOVATIONS AND MODEL ERROR VECTORS 57
7. CONCLUDING REMARKS 58
8. REFERENCES 58
9. ACKNOWLEDGEMENT 58
PART III: CONTROLLER REDUCTION 60
CHAPTER 8. OPTIMAL PROJECTION/MAXIMUM ENTROPY:STOCHASTIC MODELLING AND REDUCEDORDER DESIGN SYNTHESIS 60
1. OVERVIEW 60
2. MOTIVATION 60
3. REVIEW OF THE OPTIMAL PROJECTION APPROACH 61
4. MAXIMUM ENTROPY MODELLING 61
5. MINIMUM-INFORMATION MODELLIG OF PARAMETER UNCERTAINTIES 62
6. OPTIMAL PROJECTION DESIGN WITH STRATONOVICH MULTIPLICATIVE WHITE NOISE 65
CHAPTER 9. APPROXIMATION AND COMPENSATOR ORDER DETERMINATION IN OPTIMAL CONTROL OF FLEXIBLE STRUCTURES 68
INTRODUCTION 68
APPROXIMATION 69
APPLICATION TO WRAP-RIB ANTENNA 70
CONVERGENCE OF COMPENSATORS AND CLOSED-LOOP PERFORMANCE 70
REFERENCES 71
CHAPTER 10. LINEAR CONTROLLER APPROXIMATION: A METHOD WITH BOUNDS 76
INTRODUCTION 76
FORMULATION OF THE APPROXIMATION PROBLEM 76
HANKEL-NORM FREQUENCY-WEIGHTED APPROXIMATION 77
CONSTRUCTION OF A BOUND 77
EXAMPLES 78
CONCLUSIONS 78
REFERENCES 79
CHAPTER 11. CONTROLLER REDUCTION AND ANALYSIS OF SUBOPTIMALITY AND STABILITY OF DECENTRALIZED SYSTEMS 82
INTRODUCTION 82
2. FULL ORDER LOCAL CONTROLLERS 83
3. REDUCED ORDER LOCAL CONTROLLERS 84
4. SUBOPTIMALITY OF DECENTRALIZED CONTROLLERS 85
5. CONCLUSION 87
6. APPENDIX A 87
7. APPENDIX B 88
8. REFERENCES 89
CHAPTER 12. ON OPTIMALITY AND ROBUSTNESS OF LQREGULATORS FOR NONLINEAR ANDINTERCONNECTED SYSTEMS 90
1. INTRODUCTION 90
2. UNCONSTRAINED CONTROL STRUCTURE 90
3. DECENTRALIZED INFORMATION CONSTRAINTS 93
ACKNOWLEDGMENT 95
REFERENCES 95
CHAPTER 13. APPROXIMATE FILTERING, CONTROL AND INTERACTION IN LARGE-SCALE SYSTEMS 96
I. INTRODUCTION 96
II. Canonical Correlation Analysis (CCA 97
III. Approximate (Reduced-Order) 97
IV. Approximate (Reduced-Order) LQR [18]-[19] 98
V. Strong and Weak Interactions 100
VI. Some Discussion 101
References 102
PART IV: FREQUENCY DOMAIN VIEWPOINTS 104
CHAPTER 14. Hoo AND LQG ROBUST DESIGN METHODS FOR UNCERTAIN LINEAR SYSTEMS 104
1. Introduction 104
2. LQG Problem and Solution 104
3. Sensitivity Minimization 105
4. Embedding the H8 Problem 106
5. Diophantine Equation Solution 107
6. Statement of the H8 Algorithm 109
7. Solution of the Diophantine Equations 109
8. Conclusions 109
9. References 109
CHAPTER 15. DESIGN OF LINEAR CONTROL SYSTEMS FOR ROBUST STABILITY AND PERFORMANCE 110
INTRODUCTION 110
PROBLEM FORMULATION 110
PROBLEM TRANSFORMATION 111
FREQUENCY DEPENDENT TRANSFORMATIONS 112
IMPLEMENTATION ANDSIMULATIONS 115
CONCLUSION 115
REFERENCES 115
CHAPTER 16. ON HANKEL MATRIX APPROXIMATIONS AND STOCHASTIC MODEL REDUCTION 118
INTRODUCTION 118
INVARIANT SUBSPACES AND STATE SPACE APPROXIMATIONS 118
HANKEL APPROXIMATIONS AND STOCHASTIC MODEL REDUCTION 119
ACKNOWLEDGEMENTS 120
REFERENCES 120
CHAPTER 17. FINITE-DIMENSIONAL CONTROLLERS FOR LINEAR DISTRIBUTED PARAMETER SYSTEMS: EXPONENTIAL STABILITY USING RESIDUAL MODE FILTERS 122
INTRODUCTION 122
LINEAR DISTRIBUTED PARAMETER SYSTEMS 122
DESIGN OF RESIDUAL MODEFILTERS 123
EXPONENTIAL STABILITY USINGSTATIC OUTPUT FEEDBACK WITHA RESIDUAL MODE FILTER 124
EXPONENTIAL STABILITY USINGDYNAMIC OUTPUT FEEDBACK WITHA RESIDUAL MODE FILTER 125
CONCLUSIONS 127
ACKNOWLEDGEMENTS 127
REFERENCES 127
CHAPTER 18. MODEL UNCERTAINTY AND LINEAR SYSTEM IDENTIFICATION 128
1. Introduction 128
2. Confidence Sets 129
3. "Time-model-in-the-model-class" Case Returning to the original problem 130
4. Concluding Remarks 133
Acknowledgements 133
References 133
PART V: ADAPTIVE AND OTHER VIEWPOINTS 134
CHAPTER 19. VIBRATIONAL-FEEDBACK CONTROL OF DECENTRALIZED SYSTEMS: A DESIGN ALGORITHM 134
1. Introduction 134
2. Preliminaries 134
3. Main Theorem 135
4. Example 136
5. Conclusions 137
Acknowledgement 137
References 137
CHAPTER 20. STABILIZATION OF UNCERTAIN DISCRETE-TIME SYSTEMS 138
1. INTRODUCTION 138
2. PROBLEM STATEMENT 138
3. LYAPUNOV MIN-MAX CONTROLLERS 138
4. NOMINAL SYSTEM SUBJECT TO MIN-MAXCONTROLLERS 139
5. UNCERTAIN SYSTEM SUBJECT TO MIN-MAX CONTROLLERS 139
6. APPLICATION 140
7. APPENDIX 141
ACKNOWLEDGEMENT 141
REFERENCES 141
CHAPTER 21. TIME DOMAIN ROBUST CONTROL DESIGNFOR LINEAR QUADRATIC REGULATORS BY PERTURBATION BOUND ANALYSIS 142
I. INTRODUCTION 142
II. BOUNDS FOR ROBUST STABILITY AND REGULATION 143
I I . PERFORMANCE ROBUSTNESS INDEX AND DESIGN ALGORITHM 144
IV. APPLICATION EXAMPLE 146
V. CONCLUSIONS 146
ACKNOWLEDGEMENTS 147
REFERENCES 147
CHAPTER 22. INFINITE DIMENSIONAL MODELS ARE BETTER APPROXIMATIONS THAN FINITE DIMENSIONAL ONES 150
INTRODUCTION 150
A PROTOTYPE EXAMPLE 150
SOME OPINIONS 151
REFERENCES 151
AUTHOR INDEX 152

Produkt-Details

• Erscheint lt. Verlag: 28.6.2014
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Datenbanken
• Themenwelt: Informatik ► Software Entwicklung ► User Interfaces (HCI)
• Themenwelt: Technik ► Bauwesen
• Themenwelt: Technik ► Maschinenbau
• ISBN-10: 1-4832-9826-4 / 1483298264
• ISBN-13: 978-1-4832-9826-9 / 9781483298269