系统辨识 使用者的理论 英文版PDF|Epub|txt|kindle电子书版本下载

系统辨识 使用者的理论 英文版PDF格式电子书版下载

注意：本站所有压缩包均有解压码： 点击下载压缩包解压工具

(1 Introduction1

1．1 Dynamic Systems1

1．2 Models6

1．3 An Archetypical Problem-ARX Models and the Linear Least Squares Method8

1．4 The System Identification Procedure13

1．5 Organization of the Book14

1．6 Bibliography16

(2 Time-Invariant Linear Systems18

2．1 Impulse Responses, Disturbances, and Transfer Functions18

2．2 Frequency-Domain Expressions28

2．3 Signal Spectra33

2．4 Single Realization Behavior and Ergodicity Results（*）42

2．5 Multivariable Systems（*）44

2．6 Summary45

2．7 Bibliography46

2．8 Problems47

Appendix 2A: Proof of Theorem 2.252

Appendix 2B: Proof of Theorem 2.355

Appendix 2C: Covariance Formulas61

3．1 Simulation63

(3 Simulation and Predition63

3．2 Prediction64

3．3 Observers72

3．4 Summary75

3．5 Bibliography75

3．6 Problems76

(4 Models of Linear Time-Invariant Systems79

4．1 Linear Models and Sets of Linear Models79

4．2 A Family of Transfer-Function Models81

4．3 State-Space Models93

4．4 Distributed Parameter Models（*）103

4．5 Model Sets, Model Structures, and Identifiability: Some Formal Aspects（*）105

4．6 Identifiability of Some Model Structures114

4．7 Summary118

4．8 Bibliography119

4．9 Problems121

Appendix 4A: Identifiability of Black-Box Multivariable Model Structures128

(5 Models for Time-varying and Nonlinear Systems140

5．1 Linear Time-Varying Models140

5．2 Models with Nonlinearities143

5．3 Nonlinear State-Space Models146

5．4 Nonlinear Black-Box Models: Basic Principles148

5．5 Nonlinear Black-Box Models: Neural Networks, Wavelets and Classical Models154

5．6 Fuzzy Models156

5．7 Formal Characterization of Models（*）161

5．8 Summary164

5．9 Bibliography165

5．10 Problems165

(6 Nonparametric Time-and Frequency-Domain Methods168

6．1 Transient-Response Analysis and Correlation Analysis168

6．2 Frequency-Response Analysis170

6．3 Fourier Analysis173

6．4 Spectral Analysis178

6．5 Estimating the Disturbance Spectrum（*）187

6．6 Summary189

6．7 Bibliography190

6．8 Problems191

Appendix 6A: Derivation of the Asymptotic Properties of the Spectral Analysis Estimate194

(7 Parameter Estimation Methods197

7．1 Guiding Principles Behind Parameter Estimation Methods197

7．2 Minimizing Prediction Errors199

7．3 Linear Regressions and the Least-Squares Method203

7．4 A Statistical Framework for Parameter Estimation and the Maximum Likelihood Method212

7．5 Correlating Prediction Errors with Past Data222

7．6 Instrumental-Variable Methods224

7．7 Using Frequency Domain Data to Fit Linear Models（*）227

7．8 Summary233

7．9 Bibliography234

7．10 Problems236

Appendix 7A: Proof of the Cramér-Rao Inequality245

(8 Convergence and Consistency247

8．1 Introduction247

8．2 Conditions on the Data Set249

8．3 Prediction-Error Approach253

8．4 Consistency and Identifiability258

8．5 Linear Time-Invariant Models:A Frequency-Domain Description of the Limit Model263

8．6 The Correlation Approach269

8．7 Summary273

8．8 Bibliography274

8．9 Problems275

(9 Asymptotic Distribution of Parameter Estimates280

9．1 Introduction280

9．2 The Prediction-Error Approach: Basic Theorem281

9．3 Expressions for the Asymptotic Variance283

9．4 Frequency-Domain Expressions for the Asymptotic Variance290

9．5 The Correlation Approach296

9．6 Use and Relevance of Asymptotic Variance Expressions302

9．7 Summary304

9．8 Bibliography305

9．9 Problems305

Appendix 9A: Proof of Theorem 9.1309

Appendix 9B: The Asymptotic Parameter Variance313

(10 Computing the Estimate317

10．1 Linear Regressions and Least Squares317

10．2 Numerical Solution by Iterative Search Methods326

10．3 Computing Gradients329

10．4 Two-Stage and Multistage Methods333

10．5 Local Solutions and Initial Values338

10．6 Subspace Methods for Estimating State Space Models340

10．7 Summary351

10．8 Bibliography352

10．9 Problems353

11 Recursive Estimation Methods361

11．1 Introduction361

11．2 The Recursive Least-Squares Algorithm363

11．3 The Recursive IV Method369

11．4 Recursive Prediction-Error Methods370

11．5 Recursive Pseudolinear Regressions374

11．6 The Choice of Updating Step376

11．7 Implementation382

11．8 Summary386

11．9 Bibliography387

11．10 Problems388

Appendix 11A: Techniques for Asymptotic Analysis of Recursive Algorithms389

11A Problems398

(12 Options and Objectives399

12．1 Options399

12．2 Objectives400

12．3 Bias and Variance404

12．4 Summary406

12．5 Bibliography406

12．6 Problems406

(13 Experiment Design408

13．1 Some General Considerations408

13．2 Informative Experiments411

13．3 Input Design for Open Loop Experiments415

13．4 Identification in Closed Loop:Identifiability428

13．5 Approaches to Closed Loop Identification434

13．6 Optimal Experiment Design for High-Order Black-Box Models441

13．7 Choice of Sampling Interval and Presampling Filters444

13．8 Summary452

13．9 Bibliography453

13．10 Problems454

(14 Preprocessing Data458

14．1 Drifts and Detrending458

14．2 Outliers and Missing Data461

14．3 Selecting Segments of Data and Merging Experiments464

14．4 Prefiltering466

14．5 Formal Design of Prefiltering and Input Properties470

14．6 Summary474

14．8 Problems475

14．7 Bibliography475

(15 Choice of Identification Criterion477

15．1 General Aspects477

15．2 Choice of Norm: Robustness479

15．3 Variance-Optimal Instruments485

15．4 Summary488

15．5 Bibliography489

15．6 Problems490

(16 Model Structure Selection and Model Validation491

16．1 General Aspects of the Choice of Model Structure491

16．2 A Priori Considerations493

16．3 Model Structure Selection Based on Preliminary Data Analysis495

16．4 Comparing Model Structures498

16．5 Model Validation509

16．6 Residual Analysis511

16．7 Summary516

16．8 Bibliography517

16．9 Problems518

(17 System Identification in Practice520

17．1 The Tool:Interactive Software520

17．2 The Practical Side of System Identification522

17．3 Some Applications525

17．4 What Does System Identification Have To Offer?536

(AppendixⅠSome Concepts From Probability Theory539

( AppendixⅡ Some Statistical Techniques for Linear Regressions543

Ⅱ．1 Linear Regressions and the Least Squares Estimate543

Ⅱ．2 Statistical Properties of the Least-Squares Estimate551

Ⅱ．3 Some Further Topics in Least-Squares Estimation559

Ⅱ．4 Problems564

References565

Subject Index596

Reference Index603