| ABSTRACT | 第1-8页 |
| 摘要 | 第8-19页 |
| Contents | 第19-29页 |
| Chapter 1:Introduction | 第29-44页 |
| ·Time series | 第29-30页 |
| ·Time series classifications | 第30-32页 |
| ·Time series pre-processing | 第32-35页 |
| ·Time series model development and research status | 第35-39页 |
| ·Time series and control design | 第39-41页 |
| ·Outline of the thesis | 第41-44页 |
| Chapter 2:Nonlinear Time Series Modeling and Prediction using Local VariableWeights RBF Network | 第44-55页 |
| ·Introduction | 第44-45页 |
| ·Problem statement | 第45页 |
| ·Local Variable weights RBFN | 第45-47页 |
| ·Model Identification | 第47-49页 |
| ·Results and discussion | 第49-54页 |
| ·Modeling and prediction of Mackey-Glass Time series | 第49-52页 |
| ·Modeling and prediction of EEC Time series | 第52-53页 |
| ·Modeling and prediction of the Box-Jenkins Gas Furnace Time series | 第53-54页 |
| ·Summary | 第54-55页 |
| Chapter 3:A Local Polynomial RBF network-based state-dependent AR modelAR model for Modeling and Prediction of Nonlinear Time Series | 第55-73页 |
| ·Introduction | 第55-57页 |
| ·Problem statement | 第57页 |
| ·Local Polynomial RBF network-based state-dependent AR model | 第57-60页 |
| ·Local Polynomial RBF network | 第57-58页 |
| ·State-dependent AR model | 第58-59页 |
| ·LPRBF-AR Model | 第59-60页 |
| ·Model Identification | 第60-61页 |
| ·Results and discussion | 第61-71页 |
| ·Modeling and prediction of Mackey-Glass Time series | 第61-65页 |
| ·Modeling and prediction of Lorenz Time series | 第65-67页 |
| ·Modeling and prediction of EEC Time series | 第67-69页 |
| ·Modeling and prediction of the Box-Jenkins Gas Furnace Time series | 第69-71页 |
| ·Summary | 第71-73页 |
| Chapter 4:Local Polynomial Wavelet Neural Network with a NonlinearStructured Parameter Optimization Method | 第73-98页 |
| ·Introduction | 第73-75页 |
| ·Structure of the LPWNN-SNPOM | 第75-78页 |
| ·LPWNN-SNPOM learning algorithm | 第78-81页 |
| ·Case studies | 第81-96页 |
| ·Modeling and prediction of the chaotic time series | 第81-83页 |
| ·Identification of the nonlinear dynamic system | 第83-84页 |
| ·Modeling and Prediction of a Chaotic Signal | 第84-88页 |
| ·Approximation of a nonlinear function | 第88-90页 |
| ·Modeling and prediction of monthly Australian sparkling wine time series | 第90-91页 |
| ·Modeling and prediction of the Box-Jenkins Gas Furnace Time series | 第91-93页 |
| ·Modeling and prediction of Sunspots time series | 第93-96页 |
| ·Summary | 第96-98页 |
| Chapter 5:Nonlinear Time Series Modeling and Prediction Using FunctionalWeights Wavelet Neural Network-Based State-Dependent AR Model | 第98-129页 |
| ·Introduction | 第98-100页 |
| ·Functional weight wavelet network-based state-dependent AR model | 第100-104页 |
| ·State-dependent AR model | 第100-101页 |
| ·FWWNN-AR Model | 第101-104页 |
| ·Model Identification | 第104-106页 |
| ·Case studies | 第106-127页 |
| ·The generated nonlinear time series | 第106-117页 |
| ·Modeling and prediction of the distorted long-memory AR time series | 第106-109页 |
| ·Mackey-Glass Time series | 第109-114页 |
| ·Lorenz Time series | 第114-117页 |
| ·The real nonlinear time series | 第117-127页 |
| ·Sunspot Time series | 第117-119页 |
| ·Modeling and prediction of monthly Australian sparkling wine time series | 第119-120页 |
| ·Gas Furnace Time series | 第120-127页 |
| ·Summary | 第127-129页 |
| Chapter 6:Marine Vehicle Modeling and Tracking Using Wavelet Type NetsModel | 第129-146页 |
| ·Introduction | 第129-131页 |
| ·Ship modeling | 第131-137页 |
| ·EW-WNN-ARX model | 第131-137页 |
| ·EW-WNN-ARX model identification | 第133-135页 |
| ·E-EW-WNN-ARX model | 第135页 |
| ·E-EW-WNN-ARX-MM model | 第135-137页 |
| ·E-EW-WNN-ARX-MM-MPC | 第137-140页 |
| ·Simulation study | 第140-145页 |
| ·Summary | 第145-146页 |
| Chapter 7:Conclusions and Future work | 第146-150页 |
| ·Summary and conclusion | 第146-148页 |
| ·Future work | 第148-150页 |
| Acknowledgement | 第150-151页 |
| Reference | 第151-161页 |
| Academic publications | 第161页 |
