Muutke küpsiste eelistusi

Neural Network Systems Techniques and Applications: Advances in Theory and Applications, Volume 7 [Kõva köide]

Series edited by (University of California, Los Angeles, U.S.A.)
  • Formaat: Hardback, 438 pages, kõrgus x laius: 229x152 mm, kaal: 800 g, Contains 1 Hardback
  • Sari: Control and Dynamic Systems
  • Ilmumisaeg: 09-Feb-1998
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0124438679
  • ISBN-13: 9780124438675
Teised raamatud teemal:
  • Kõva köide
  • Hind: 138,25 €*
  • * saadame teile pakkumise kasutatud raamatule, mille hind võib erineda kodulehel olevast hinnast
  • See raamat on trükist otsas, kuid me saadame teile pakkumise kasutatud raamatule.
  • Kogus:
  • Lisa ostukorvi
  • Tasuta tarne
  • Lisa soovinimekirja
Neural Network Systems Techniques and Applications: Advances in Theory and Applications, Volume 7Suurem pilt
  • Formaat: Hardback, 438 pages, kõrgus x laius: 229x152 mm, kaal: 800 g, Contains 1 Hardback
  • Sari: Control and Dynamic Systems
  • Ilmumisaeg: 09-Feb-1998
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0124438679
  • ISBN-13: 9780124438675
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780124438675.html
Märksõnad:
Examines the effectiveness and essential significance of the techniques available and (with further development) the role they will play in the future. Ten chapters cover topics such as orthogonal activation function based neural network system architecture, multilayer recurrent neural networks for synthesizing and implementing real-time linear control, adaptive control of unknown nonlinear dynamical systems, optimal tracking neural controller techniques, unified approximation theory and applications, and techniques for determining multivariable nonlinear model structures for dynamic systems with a detailed treatment of relevant system model input determination. Annotation c. by Book News, Inc., Portland, Or.

The book emphasizes neural network structures for achieving practical and effective systems, and provides many examples. Practitioners, researchers, and students in industrial, manufacturing, electrical, mechanical,and production engineering will find this volume a unique and comprehensive reference source for diverse application methodologies.
Control and Dynamic Systems covers the important topics of highly effective Orthogonal Activation Function Based Neural Network System Architecture, multi-layer recurrent neural networks for synthesizing and implementing real-time linear control,adaptive control of unknown nonlinear dynamical systems, Optimal Tracking Neural Controller techniques, a consideration of unified approximation theory and applications, techniques for the determination of multi-variable nonlinear model structures for dynamic systems with a detailed treatment of relevant system model input determination, High Order Neural Networks and Recurrent High Order Neural Networks, High Order Moment Neural Array Systems, Online Learning Neural Network controllers, and Radial Bias Function techniques.

Key Features
Coverage includes:
* Orthogonal Activation Function Based Neural Network System Architecture (OAFNN)
* Multilayer recurrent neural networks for synthesizing and implementing real-time linear control
* Adaptive control of unknown nonlinear dynamical systems
* Optimal Tracking Neural Controller techniques
* Consideration of unified approximation theory and applications
* Techniques for determining multivariable nonlinear model structures for dynamic systems,
with a detailed treatment of relevant system model input determination

Muu info

Key Features Coverage includes: * Orthogonal Activation Function Based Neural Network System Architecture (OAFNN) * Multilayer recurrent neural networks for synthesizing and implementing real-time linear control * Adaptive control of unknown nonlinear dynamical systems * Optimal Tracking Neural Controller techniques * Consideration of unified approximation theory and applications * Techniques for determining multivariable nonlinear model structures for dynamic systems, with a detailed treatment of relevant system model input determination
Contributors xiii(2) Preface xv Orthogonal Functions for Systems Identification and Control 1(75) Chaoying Zhu Deepak Shukla Frank W. Paul I. Introduction 1(1) II. Neural Networks with Orthogonal Activation Functions 2(23) A. Background 2(2) B. Neural System Identification and Control 4(1) C. Desired Network Properties for System Identification and Control 5(2) D. Orthogonal Neural Network Architecture 7(5) E. Gradient Descent Learning Algorithm 12(2) F. Properties of Orthogonal Activation Function-Based Neural Networks 14(7) G. Preliminary Performance Evaluation of Orthogonal Activation Function-Based Neural Networks 21(4) III. Frequency Domain Applications Using Fourier Series Neural Networks 25(22) A. Neural Network Spectrum Analyzer 25(8) B. Describing Function Identification 33(2) C. Fourier Series Neural Network-Based Adaptive Control Systems 35(12) IV. Time Domain Applications for System Identification and Control 47(24) A. Neural Network Nonlinear Identifier 47(8) B. Inverse Model Controller 55(3) C. Direct Adaptive Controllers 58(13) V. Summary 71(1) References 72(4) Multilayer Recurrent Neural Networks for Synthesizing and Tuning Linear Control Systems via Pole Assignment 76(51) Jun Wang I. Introduction 76(1) II. Background Information 77(2) III. Problem Formulation 79(6) IV. Neural Networks for Controller Synthesis 85(8) V. Neural Networks for Observer Synthesis 93(5) VI. Illustrative Examples 98(25) VII. Concluding Remarks 123(2) References 125(2) Direct and Indirect Techniques to Control Unknown Nonlinear Dynamical Systems Using Dynamical Neural Networks 127(31) George A. Rovithakis Manolis A. Christodoulou I. Introduction 127(3) A. Notation 129(1) II. Problem Statement and the Dynamic Neural Network Model 130(2) III. Indirect Control 132(7) A. Identification 132(2) B. Control 134(5) IV. Direct Control 139(15) A. Modeling Error Effects 140(8) B. Model Order Problems 148(6) V. Conclusions 154(1) References 154(4) A Receding Horizon Optimal Tracking Neurocontroller for Nonlinear Dynamic Systems 158(33) Young-Moon Park Myeon-Song Choi Kwang Y. Lee I. Introduction 158(1) II. Receding Horizon Optimal Tracking Control Problem Formulation 159(4) A. Receding Horizon Optimal Tracking Control Problem of a Nonlinear System 159(3) B. Architecture for an Optimal Tracking Neurocontroller 162(1) III. Design of Neurocontrollers 163(13) A. Structure of Multilayer Feedforward Neural Networks 163(1) B. Identification Neural Network 164(4) C. Feedforward Neurocontroller 168(2) D. Feedback Neurocontroller 170(1) E. Generalized Backpropagation-through-Time Algorithm 171(6) IV. Case Studies 176(11) A. Inverted Pendulum Control 176(4) B. Power System Control 180(7) V. Conclusions 187(1) References 188(3) On-Line Approximators for Nonlinear System Identification: A Unified Approach 191(40) Marios M. Polycarpou I. Introduction 191(2) II. Network Approximators 193(7) A. Universal Approximators 194(3) B. Universal Approximation of Dynamical Systems 197(2) C. Problem Formulation 199(1) III. Learning Algorithm 200(10) A. Weight Adaptation 202(4) B. Linearly Parametrized Approximators 206(2) C. Multivariable Systems 208(2) IV. Continuous-Time Identification 210(18) A. Radial-Basis-Function Network Models 213(10) B. Multilayer Network Models 223(5) V. Conclusions 228(1) References 229(2) The Determination of Multivariable Nonlinear Models for Dynamic Systems 231(48) S. A. Billings S. Chen I. Introduction 231(2) II. The Nonlinear System Representation 233(2) III. The Conventional NARMAX Methodology 235(11) A. Structure Determination and Parameter Estimation 236(9) B. Model Validation 245(1) IV. Neural Network Models 246(8) A. Multilayer Perceptrons 247(1) B. Radial Basis Function Networks 248(3) C. Fuzzy Basis Function Networks 251(1) D. Recurrent Neural Networks 252(2) V. Nonlinear-in-the-Parameters Approach 254(5) A. Parallel Prediction Error Algorithm 255(2) B. Pruning Oversized Network Models 257(2) VI. Linear-in-the-Parameters Approach 259(12) A. Regularized Orthogonal Least-Squares Learning 262(5) B. Enhanced Clustering and Least-Squares Learning 267(4) C. Adaptive On-Line Learning 271(1) VII. Identifiability and Local Model Fitting 271(2) VIII. Conclusions 273(2) References 275(4) High-Order Neural Network Systems in the Identification of Dynamical Systems 279(28) Elias B. Kosmatopoulos Manolis A. Christodoulou I. Introduction 279(2) II. RHONNs and g-RHONNs 281(3) III. Approximation and Stability Properties of RHONNs and g-RHONNs 284(5) A. Stability and Robustness Properties of g-RHONNs 286(3) IV. Convergent Learning Laws 289(5) A. Robust Adaptive Learning Laws 290(2) B. Learning Laws That Guarantee Exponential Error Convergence 292(2) V. The Boltzmann g-RHONN 294(4) VI. Other Applications 298(6) A. Estimation of Robot Contact Surfaces 298(2) B. RHONNs for Spatiotemporal Pattern Recognition and Identification of Stochastic Dynamical Systems 300(1) C. Universal Stabilization Using High-Order Neural Networks 301(3) VII. Conclusions 304(1) References 304(3) Neurocontrols for Systems with Unknown Dynamics 307(26) William A. Porter Wie Liu Luis Trevino I. Introduction 307(2) II. The Test Cases 309(4) III. The Design Procedure 313(5) A. Using Higher Order Moments 314(1) B. Embedding the Controller Design in H(v) 315(1) C. HOMNA Training Algorithms 316(2) D. Tensor Space Matchups with the HOMNA Calculations 318(1) IV. More Details on the Controller Design 318(2) V. More on Performance 320(11) A. Disturbance Rejection 321(4) B. Propulsion System Application 325(6) VI. Closure 331(1) References 331(2) On-Line Learning Neural Networks for Aircraft Autopilot and Command Augmentation Systems 333(50) Marcello Napolitano Michael Kincheloe I. Introduction 333(3) II. The Neural Network Algorithms 336(5) A. Extended Back-Propagation Training Algorithm 339(2) III. Aircraft Model 341(1) IV. Neural Network Autopilots 342(11) A. Phase I: Flight Envelope Performance 343(5) B. Phase II: Neural Autopilot Controllers under Linear and Nonlinear Conditions 348(2) C. Conclusions 350(3) V. Neural Network Command Augmentation Systems 353(26) A. Phase I: Statistics of Learning and Adaptation 357(11) B. Phase II: Multiple Model Following Capabilities 368(5) C. Conclusions 373(6) VI. Conclusions and Recommendations for Additional Research 379(1) A. Conclusions 379(1) References 380(3) Nonlinear System Modeling 383(52) Shaohua Tan Johan Suykens Yi Yu Joos Vandewalle I. Introduction 383(2) II. RBF Neural Network-Based Nonlinear Modeling 385(9) A. The Neural Modeling Problem 385(1) B. Two Basic Issues in Neural Modeling 386(1) C. RBF Neural Network Structure 386(2) D. Suitability of the RBF Neural Network for Nonlinear Modeling 388(2) E. Excitation-Dependent Modeling 390(1) F. RBF Neural Network Modeling with a Fixed Model Structure 391(3) III. On-Line RBF Structural Adaptive Modeling 394(5) A. The Need for On-Line Structural Adaptation 394(1) B. On-Line Structure Adaptation Technique 395(4) C. Remarks 398(1) IV. Multiscale RBF Modeling Technique 399(7) A. Basic Motivation 399(1) B. Structure of a Multiscale RBF Neural Network 400(1) C. Coarse-to-Fine Residue-Based Modeling Idea 401(1) D. Construction of Multiscale RBF Model 402(3) E. Model Validation 405(1) F. Extension to Recurrent Neural Modeling 406(1) V. Neural State-Space-Based Modeling Techniques 406(3) A. Basic Motivation 406(1) B. Neural State-Space Models 407(2) VI. Dynamic Back-Propagation 409(3) VII. Properties and Relevant Issues in State-Space Neural Modeling 412(7) A. Uncertain Linear System Representations 412(4) B. Linear Models as the Starting Point 416(1) C. Imposing Stability 417(2) VIII. Illustrative Examples 419(12) A. RBF Neural Network Models 419(5) B. Neural State-Space Models 424(7) References 431(4) Index 435
Cornelius T. Leondes received his B.S., M.S., and Ph.D. from the University of Pennsylvania and has held numerous positions in industrial and academic institutions. He is currently a Professor Emeritus at the University of California, Los Angeles. He has also served as the Boeing Professor at the University of Washington and as an adjunct professor at the University of California, San Diego. He is the author, editor, or co-author of more than 100 textbooks and handbooks and has published more than 200 technical papers. In addition, he has been a Guggenheim Fellow, Fulbright Research Scholar, IEEE Fellow, and a recipient of IEEE's Baker Prize Award and Barry Carlton Award.