LINEAR PARAMETER-VARYING

Transcription

1 Advanced Series in Electrical and Computer Engineering - Vol. 14 LINEAR PARAMETER-VARYING SYSTEM IDENTIFICATION New Developments and Trends Editors Paulo Lopes dos Santos Universidade do Porto, Portugal Teresa Paula Azevedo Perdicoulis Universidade de Tras-os-Montes e Alto Douro, Portugal Carlo Novara Politecnico di Torino, Italy Jose A Ramos Nova Southeastern University, USA Daniel E Rivera Arizona State University, USA World Scientific NEW JERSEY LONDON SINGAPORE BEIJING SHANGHAI HONG KONG TAIPEI CHENNAI

2 Contents Preface v Acronyms xiii 1. Introduction 1 C. Novara et al. References 6 2. Hybrid LPV Modeling and Identification 11 L. Giarre, P. Falugi & R. Badalamenti 1. Introduction Literature review on LPV identification HLPV modeling: Problem formulation A motivating example: Traffic modeling in wireless ad-hoc networks Final remarks and open problems 32 References SM Identification of 10 LPV Models with Uncertain Time- Varying Parameters 41 V. Cerone, D. Piga & D. Regruto 1. Introduction Problem formulation Evaluation of tight parameter bounds Semi-static LPV relaxation Properties of the computed parameter uncertainty intervals PUI$n,5) Simulated example Conclusion 62 vii

3 viii Contents References SM Identification of State-Space LPV Systems 65 C. Novara 1. Introduction Notation and basic notions Set membership identification of state-space LPV systems Interpolatory and optimal estimates Important aspects of the identification process Examples Conclusion 90 References Identification of Input-Output LPV Models 95 V. Laurain et al. 1. Introduction Discrete-time LPV polynomials models Estimating LPV-ARX models in DT Addressing estimation with general noise models Direct estimation of continuous-time LPV systems Instrumental variable approach in continuous-time Conclusion 128 References Reducing the Dimensions of Data Matrices in LPV Subspace Identification 133 V. Verdult & M. Verhaegen 1. Introduction Data equations Basic ideas behind the methods Two-block identification method Implementation by selection of dominant rows Implementation by a kernel method Conclusion 162 References 163

4 Contents ix 7. Subspace Identification of MIMO LPV Systems 167 J. W. van Wingerden & M. Verhaegen 1. Introduction Problem formulation and assumptions Factorization of the LPV controllability matrix LPV predictor-based subspace identification Kernel method Simulation examples Case study: A "smart" airfoil Conclusion 196 References Subspace Identification of Continuous-Time State-Space LPV Models 201 M. Bergamasco & M. Lovera.. 1. Introduction Definitions Problem statement A balanced subspace approach to identification for gain scheduling Continuous-time predictor-based subspace identification Balancing of the identified models Model interpolation Comments and discussion Simulation examples Conclusions Acknowledgements 227 References Indirect Continuous-Time LPV System Identification 231 P. Lopes dos Santos et al. 1. Introduction LPV systems Discretisation of LPV systems Successive approximations identification algorithm Downsampled LTI discrete-time deterministic-stochastic subspace identification algorithm 240

5 X Contents 6. Case study Conclusion 254 References LPV System Identification Using Series Expansion Models 259 R. Toth, P. S. C. Heube.rger & P. M. J. Van den Ho} 1. Introduction Perspectives of series-expansion models Orthonormal basis function models Identification via OBF models Identification of a high-performance positioning device Conclusion 292 References System Identification of Linear Parameter Varying State- Space Models 295 A. Wills & B. Ninness 1. Introduction Problem formulation Maximum-likelihood estimation The expectation-maximisation (EM) algorithm for ML estimation EM for LPV models Simulation study Conclusion 312 References PWA Identification of Interconnected Systems with LFR Structure 317 S. Paoletti & A. Garulli 1. Introduction PWA-LFR models Black-box PWA system identification Structured identification of PWA-LFR models Applications Conclusions 344 References 345

6 Contents xi 13. Identification and Model (In)validation of Switched ARX Systems 347 C. Feng et al. 1. Introduction Preliminaries System identification Model (in)validation Numerical examples and applications Concluding remarks 377 References 377 Index 381