Scribd
Upload
157 views

0166927S21 Ljung

The document provides an overview of system identification theory and methods. It introduces dynamical systems models and the system identification procedure. The document is organized into three parts that cover systems and models, identification methods, and considerations for users of identification methods.

Uploaded by

Adriana Riveros Guevara
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
157 views

0166927S21 Ljung

The document provides an overview of system identification theory and methods. It introduces dynamical systems models and the system identification procedure. The document is organized into three parts that cover systems and models, identification methods, and considerations for users of identification methods.

Uploaded by

Adriana Riveros Guevara
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

SYSTEM IDENTIFICATION: Theory for the User

Lennart Ljung University of Linkping Sweden

Prentice-Hall, Inc., Engiewood Cliffs, New Jersey 07632

CONTENTS
PREFACE ACKNOWLEDGMENTS OPERATORS AND NOTATIONAL CONVENTIONS xi xv XV

1.

INTRODUCTION
1.1 1.2 1.3 1.4 1.5 Dynamical Systems Models The System Identification Procedure Organization of the Book BibHography

1
1 3 7 8 10

part i: Systems and modeis


2. TIME-INVARIANT LINEAR SYSTEMS
2.1 2.2 Impulse Responses, Disturbances and Transfer Functions Frequency-domain Expressions

13
13 22

2.3 2.4 2.5 2.6 2.7 2.8

Signal Spectra Single Realization Behavior and Ergodicity Results (*) Multivariable Systems (*) Summary Bibliography Problems Appendix 2.A: Proof of Theorem 2.2 Appendix 2.B: Proof of Theorem 2.3 Appendix 2.C: Covariance Formulas

26 34 35 36 37 38 43 45 49

SIMULATION, PREDICTION, AND CONTROL


3.1 3.2 3.3 3.4 3.5 3.6 3.7 Simulation Prediction Observers Control (*) Summary Bibliography Problems

51
51 52 59 62 65 65 66

MODELS OF LINEAR TIME-INVARIANT SYSTEMS


4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 Linear Models and Sets of Linear Models A Family of Transfer-function Models State-space Models Distributed-Parameter Models (*) Model Sets, Model Structures, and Identifiability: Some Formal Aspects (*) Identifiability of Some Model Structures Summary Bibliography Problems Appendix 4. A: Identifiability of Black-box Multivariable Model Structures

69
69 71 81 90 93 101 106 1<>6 108 115

MODELS FOR TIME-VARYING AND NONLINEAR SYSTEMS


5.1 5.2 5.3 5.4 5.5 5.6 5.7 Linear Time-varying Models Nonlinear Models as Linear Regressions Nonlinear State-space Models Formal Characterization of Models (*) Summary Bibliography Problems

127
127 130 132 134 137 138 138

viii

Contents

pari ii: methods


6. NONPARAMETRIC TIMEAND FREQUENCY-DOMAIN METHODS
6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 Transient Response Analysis and Correlation Analysis Frequency-response Analysis Fourier Analysis Spectrat Analysis Estimating the Disturbance Spectrum (*) Summary Bibliography Problems Appendix 6.A: Derivation of the Asymptotic Properties of the Spectral Analysis Estimate

141
141 143 146 151 160 162 162 163 167

7.

PARAMETER ESTIMATION METHODS


7.1 7.2 7.3 7.4 Guiding Principles behind Parameter Estimation Methods Minimizing Prediction Errors Linear Regressions and the Least-squares Method A Statistical Framework for Parameter Estimation and the Maximum Likelihood Method Correlating Prediction Errors with Past Data Instrumental-variable Methods Summary Bibliography Problems Appendix 7. A: Proof of the Cramer-Rao Inequality

169
169 171 176 181

7.5 7.6 7.7 7.8 7.9

190 192 195 196 197 206

8.

CONVERGENCEANDCONSISTENCY
8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 Introduction Conditions on the Data Set Prediction-error Approach Consistency and Identifiability Linear Time-invariant Models: A Frequencydomain Description of the Limit Model The Correlation Approach Summary Bibliography Problems

208
208 210 214 218 224 229 233 233 234

9.

ASYMPTOTIC DISTRIBUTION OF PARAMETER ESTIMATES


9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 Introduction The Prediction-error Approach: Basic Theorem Expressions for the Asymptotic Variance Frequency-domain Expressions for the Asymptotic Variance The Correlation Approach Use and Relevance of Asymptotic Variance Expressions Summary Bibliography Problems Appendix 9. A: Proof of Theorem 9.1 Appendix 9.B: The Asymptotic Parameter Variance 239 240 242 248 254 258 262 263 264 266 270

10.

COMPUTING THE ESTIMATE


10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Linear Regressions and Least Squares Numerical Solution by Iterative Search Methods Computing Gradients Two-stage and Multistage Methods Local Solutions and Initial Values Summary Bibliography Problems

274
274 282 285 288 292 294 294 296

11.

RECURSIVE ESTIMATION METHODS


11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 11.9 11.10 Introduction The Recursive Least-squares Algorithm The Recursive IV Mcthod Recursive Prediction-Error Methods Recursive Pseudolinear Regressions The Choice of Updating Step Implementation Summary Bibliography Problems Appendix I L A : Techniques for Asymptotic Analysis of Recursive Algorithms

303
303 305 311 311 316 318 322 326 327 328 329

Contents

part iii: user's choices


12. OPTIONS AND OBJECTIVES 12.1 Options 12.2 Objectives 12.3 Bias and Variance 12.4 Summary 12.5 Bibography 12.6 Problems 339 339 341 345 347 347 347

13.

AFFECTING THE BIAS DISTRIBUTION OFTRANSFER-FUNCTION ESTIMATES 13.1 Some Basic Expressions 13.2 Heuristic Discussion of Transfer-function Fit in Open-loop Operation 13.3 Some Solutions to Formal Design Problems 13.4 Summary 13.5 Bibography 13.6 Problems

349 349 350 354 356 356 357

14.

EXPERIMENT DESIGN 14.1 Some General Considerations 14.2 Informative Experiments 14.3 Optimal Input Design (*) 14.4 Optimal Experiment Design for High-order Black-box Models (*) 14.5 Choice of Sampling Interval and Presampling Filters 14.6 Pretreatment of Data 14.7 Summary 14.8 Bibography 14.9 Problems

358 359 361 369 375 378 386 389 390 391

15.

CHOICE OF IDENTIFICATION CRITERION 15.1 General Aspects L5.2 Choice of Norm: Robustness 15.3 Variance: Optimal Instruments 15.4 Summary 15.5 Bibography 15.6 Problems

394 394 396 402 405 406 406

Contents

XI

16.

MODEL STRUCTURE SELECTION AND MODEL VALIDATION


16.1 16.2 16.3 16.4 1.6.5 16.6 16.7 16.8 General Aspects of the Choice of Model Structure A Priori Considerations Model Structure Selection Based on Preminary Data Analysis Comparing Model Structures Model Validation Summary Bibliography Problems 408 411 413 416 424 430 431 431

17.

SYSTEM IDENTIFICATION IN PRACTICE


17.1 17.2 17.3 17.4 17.5 The Tool: Interactive Software A Laboratory-scaie Application Identification of Ship-steering Dynamics What Does System Identification Have to Offer? Bibliography Some Concepts from Probabity Theory Some Statistical Techniques for Linear Regressions

434
434 440 449 454 456 457 461

APPENDIX I: APPENDIX II:

REFERENCES AUTHOR INDEX SUBJECT INDEX

482 505 511

xii

Contents

You might also like