Welcome

Control-oriented Modelling and Identification: Theory and practice

book image

IET Digital Library

This title is available electronically through the IET Digital Library

  • Book title: Control-oriented Modelling and Identification:Theory and practice

  • Author:

  • Year: 2014

  • Format: Hardback

  • Product Code: PBCE0800

  • ISBN: 978-1-84919-614-7

  • Pagination: 408pp

  • Stock Status: In stock

£51.35 Member price

£79.00 Full price


Buy Now

Description

This comprehensive book covers the state-of-the-art in control-oriented modelling and identification techniques.

With contributions from leading researchers in the subject, Control-oriented Modelling and Identification: Theory and practice covers the main methods and tools available to develop advanced mathematical models suitable for control system design, including;

  • object-oriented modelling and simulation
  • projection based model reduction techniques
  • integrated modelling and parameter estimation
  • identification for robust control of complex systems
  • subspace-based multi-step predictors for predictive control
  • closed-loop subspace predictive control
  • structured nonlinear system identification
  • linear fractional LPV model identification from local experiments using an H1-based glocal approach

Control-oriented Modelling and Identification: Theory and practice also takes a practical look at a variety of applications of advanced modelling and

identification techniques covering spacecraft dynamics, vibration control, rotorcrafts, models of anaerobic digestion, a brake-by-wire racing motorcycle actuator, and robotic arms.

About the Author

Marco Lovera is Professor of Automatic Control at the Dipartimento di Elettronica e Informazione of the Politecnico di Milano, Italy and has authored over 150 scientific publications covering system identification, spacecraft attitude and orbit control and advanced active control applications

He is currently associate editor of Automatica and IEEE Transactions on Control Systems Technology and is on the editorial board of IET Control Theory and Applications and IEEE Control Systems Magazine.

Book readership

Control-oriented Modelling and Identification: Theory and practice is an essential overview for researchers and graduate students in model identification and control systems design. It will also be of use to engineers in industry working on mathematical modelling and control systems design in various domains.

Book contents

1 Introduction to control-oriented modelling 1
Marco Lovera
Abstract 1
1.1 Introduction 1
1.1.1 Detailed models for system simulation 2
1.1.2 Compact models for control design 3
1.1.3 Building models for control system synthesis 3
1.2 Overview of the book 5
1.2.1 Part 1: theory 5
1.2.2 Part 2: applications 6

2 Object-oriented modelling and simulation of physical systems 9
Francesco Casella
Abstract 9
2.1 Introduction 9
2.2 Basic concepts and principles 10
2.3 Modelica 14
2.4 Mathematical processing of OO models 20
2.5 Plant modelling, analysis and identification 25
2.6 Control system performance verification 26
2.7 Direct use of OO models for optimal control 28
2.8 Conclusions 32
References 32

3 Projection-based model reduction techniques 35
Pierre Vuillemin, CharlesPoussot-Vassal and Daniel Alazard
Abstract 35
3.1 Introduction 35
3.1.1 Motivations 35
3.1.2 Model reduction by projection 38
3.2 Model reduction by truncation 41
3.2.1 State-space truncation and residualization 41
3.2.2 Balanced truncation 45
3.2.3 Conclusion 57
3.3 Moment matching methods 59
3.3.1 Moment matching through Krylov subspaces 59
3.3.2 H2 optimal model reduction 66
3.3.3 Conclusion 74
3.4 Conclusion 74
References 74

4 Integrated modelling and parameter estimation: an LFR–Modelica approach 77
Marco Lovera andFrancesco Casella
Abstract 77
4.1 Introduction 77
4.2 Applicable models and LFRs 78
4.2.1 Applicable plant models 78
4.2.2 Linear fractional representations 79
4.3 Transformation of non-linear DAE models into LFR 80
4.3.1 Definitions and assumptions 80
4.3.2 Re-ordering of the system equations 82
4.3.3 Elimination of known parameters 83
4.3.4 Solving the system equations 84
4.3.5 Formulation of the system equations as a cascaded
connection of LFRs 85
4.3.6 Construction of the LFR of the DAE 87
4.3.7 Implementation of the algorithm 89
4.3.8 Simulation of the LFR 90
4.4 Application example: identification of LFR models 91
4.5 Conclusions 98
References 99

5 Identification for robust control of complex systems: algorithm and motion application 101
TomOomenand Maarten Steinbuch
Abstract 101
5.1 Introduction 101
5.2 Coprime factor identification for refined uncertainty structures
in robust control 103
5.2.1 Robust control framework 103
5.2.2 Identification for robust control approach 105
5.2.3 Identifying robust-control-relevant coprime
factorizations 107
5.3 Generalized SK-iterations for closed-loop coprime
factor identification 108
5.3.1 Model parameterization 108
5.3.2 Frequency domain identification involving _∞-norms via
Lawson’s algorithm 109
5.3.3 A closed-loop generalization of SK iterations 110
5.4 Orthogonal polynomials w.r.t. a data-dependent discrete
inner product 111
5.5 Experimental application 112
5.5.1 Experimental system 112
5.5.2 Coprime factor identification results 115
5.5.3 Numerical conditioning 118
5.5.4 Illustration of robust-control-relevance 119
5.6 Conclusions 121
Acknowledgments 121
References 122

6 Subspace-based multi-step predictors for predictive control 125
Marzia Cescon and Rolf Johansson
Abstract 125
6.1 Introduction 125
6.1.1 Model description 126
6.1.2 Notation 127
6.1.3 Statement of the problem 128
6.2 Subspace-based linear multi-step predictors 128
6.2.1 Computing projections 130
6.3 Example 132
6.3.1 Diabetes mellitus 132
6.3.2 Experimental conditions 133
6.3.3 Prediction strategy 133
6.3.4 Results 134
6.4 Discussion and conclusions 139
References 140

7 Closed-loop subspace predictive control 143
Gijs van der Veen, Jan-Willem vanWingerden and Michel Verhaegen
Abstract 143
7.1 Introduction 143
7.2 Discrete-time identification framework 144
7.2.1 Preliminaries and notation 146
7.2.2 Data equations 146
7.2.3 Relation to the ARX model structure 147
7.2.4 Closed-loop identification issues 148
7.2.5 Estimating the predictor Markov parameters 148
7.2.6 Recursive solution of the parameter estimation problem 149
7.2.7 Using directional forgetting 150
7.3 Deriving the subspace predictor 151
7.4 Setting up the predictive control problem 152
7.4.1 Real time solution of the QP 154
7.4.2 Parameter selection 154
7.5 Concluding remarks 155
7.5.1 Algorithm summary 155
References 155

8 Structured nonlinear system identification 159
TyroneVincent, KameshwarPoolla and CarloNovara
Abstract 159
8.1 Introduction 159
8.2 Specification of model structures using the LFR 160
8.2.1 Simple examples with linear N 163
8.2.2 Simple examples with nonlinear N 165
8.2.3 LFRs of block-oriented models 166
8.2.4 Discussion: L known or unknown? 167
8.3 Examples of model structure specification 168
8.3.1 High-dimensional model representation 168
8.3.2 Automobile suspension 169
8.3.3 Nonlinear friction: drill-string 171
8.3.4 Linear parameter varying systems 172
8.4 Properties of the LFR model structure 174
8.4.1 Measurability 174
8.4.2 Identifiability 175
8.4.3 Persistence of excitation 176
8.5 Identification algorithms 176
8.5.1 Parametric estimates 176
8.5.2 Nonparametric estimates 178
8.6 Identification example 183
References 186

9 Linear fractional LPV model identification from local experiments using an H∞-based glocal approach 189
Daniel Vizer, Guillaume Mercère, Edouard Laroche and Olivier Prot
Abstract 189
9.1 Introduction 189
9.2 Identification method 192
9.2.1 Problem formulation, definitions, and notations 192
9.2.2 Determination of the structure of G (s,_(pi),_) 195
9.2.3 H∞-based optimization technique 197
9.2.4 Computing the H∞-norm 199
9.2.5 Minimizing the H∞-norm 200
9.3 Identification results 202
9.3.1 System description 202
9.3.2 Linear fractional LPV model identification 203
9.3.3 Validation 207
9.4 Conclusions 210
References 211

10 Object-oriented modelling of spacecraft dynamics: tools and case studies 215
Marco Lovera andFrancesco Casella
Abstract 215
10.1 Introduction 215
10.2 The Modelica Space Flight Dynamics library 217
10.3 Structure of the spacecraft simulation models 220
10.3.1 ExtendedWorld model 220
10.3.2 SpacecraftDynamics model 221
10.3.3 Spacecraft model 223
10.4 Case studies 226
10.4.1 Assessing external disturbances via dynamic inversion 226
10.4.2 Magnetic detumbling for small satellite attitude control 228
10.5 Concluding remarks 237
References 237

11 Control-oriented aeroelastic BizJet low-order LFT modeling 241
CharlesPoussot-Vassal, Clement Roos, Pierre Vuillemin,
Olivier Cantinaud and Jean-Patrick Lacoste
Abstract 241
11.1 Introduction 241
11.1.1 Foreword on the Dassault-Aviation BizJet models 241
11.1.2 The BizJet aircraft aeroelatic control problem 242
11.1.3 Mathematical problem formulation 243
11.1.4 Structure and notation 245
11.2 Multi-LTI model approximation and interpolation algorithm overview 245
11.3 Frequency-limited large-scale MIMO multi-LTI models approximation 247
11.3.1 Preliminaries on projection-based LTI model approximation 248
11.3.2 Large-scale single-LTI model approximation procedure 248
11.3.3 Large-scale multi-LTI models approximation procedure 250
11.3.4 Application to the BizJet model 253
11.4 Interpolation of the reduced-order models 257
11.4.1 Choice of a suitable state-space form 257
11.4.2 Description of the interpolation method 258
11.4.3 Generation of a simplified LFR 259
11.4.4 Application to the BizJet model 261
11.5 Conclusion 263
Acknowledgments 266
References 266

12 Active vibration control using subspace predictive control 269
Gijs van derVeen, Jan-Willem vanWingerden and MichelVerhaegen
Abstract 269
12.1 Introduction 269
12.2 Experimental set-up 270
12.2.1 Control design 271
12.2.2 Notes on the implementation 271
12.3 Results 272
12.4 Conclusions 273
Acknowledgements 274
References 274

13 Rotorcraft system identification: an integrated time–frequency-domain approach 275

Marco Bergamasco and Marco Lovera
Abstract 275
13.1 Introduction 275
13.2 Problem statement and preliminaries 277
13.3 An integrated time–frequency-domain approach 278
13.3.1 Continuous-time predictor-based subspace model
identification 279
13.3.2 From unstructured to structured models with an
H∞ approach 284
13.4 Bootstrap uncertainty estimation in subspace identification
methods 285
13.5 Simulation study: model identification for the BO-105
helicopter 286
13.6 Concluding remarks 298
References 299

14 Parameter identification of a reduced order LFT model of anaerobic digestion 301
Alessandro Della Bona, GianniFerretti, ElenaFicara
And Francesca Malpei
Abstract 301
14.1 Introduction 301
14.2 ADM1 model 303
14.3 Modified AMOCO model 307
14.4 LFT modelling and identification 309
14.5 Parameter identification based on ADM1 model simulation data 313
14.6 Parameter identification based on experimental data 319
14.7 Conclusion 323

Acknowledgements 324
Appendix A. LFT model for parameter identification based on
ADM1 model simulation data 324
Appendix B. LFT model for parameter identification based on
experimental data 325
References 326

15 Modeling and parameter identification of a brake-by-wire actuator for racing motorcycles 329
Matteo Corno, FabioTodeschini, GiulioPanzani and Sergio M. Savaresi
Abstract 329
15.1 Introduction 329
15.2 System description 331
15.3 Brake-by-wire modeling 332
15.3.1 Electric domain modeling 333
15.3.2 Mechanical domain modeling 333
15.3.3 Hydraulic domain 337
15.4 Parameter identification 347
15.4.1 Electric dynamics identification 347
15.4.2 Motor mechanical dynamics – Jmot and rvisc – identification 350
15.4.3 Friction model identification 353
15.4.4 Final parameter identification 353
15.5 Validation and analysis 354
15.5.1 Validation 354
15.5.2 Discussion on modeling choices 358
15.6 Conclusions 360
References 361

16 LPV modeling and identification of a 2-DOF flexible robotic arm from local experiments using an H∞-based glocal approach 365
Daniel Vizer, Guillaume Mercère, Edouard Laroche and Olivier Prot
Abstract 365
16.1 Introduction 365
16.2 Modeling of a flexible robotic manipulator 367
16.2.1 Description of the 2-DOF robotic manipulator 367
16.2.2 Linear fractional LPV representation: a reminder 370
16.2.3 Nonlinear and linearized dynamic models 370
16.3 Identification results 375
16.4 Conclusions 382
References 383

Powered by Google
Search the full text of this book