| 基于时变神经网络的非线性时变系统建模 |
| |
| 引用本文： | 严伟力,孙明轩.基于时变神经网络的非线性时变系统建模[J].小型微型计算机系统,2011,32(6):1228-1231. |
| |
| 作者姓名： | 严伟力 孙明轩 |
| |
| 作者单位： | 浙江工业大学信息工程学院,杭州,310023 |
| |
| 基金项目： | 国家自然科学基金项目(60874041)资助 |
| |
| 摘 要： | 提出时变神经网络模型,用以逼近未知非线性时变映射,实现非线性时变系统建模.将时变神经网络的权值学习作为时变系统的时变参数估计问题,并基于迭代学习机制,给出在同一时刻沿迭代轴训练网络权值的迭代学习最小二乘算法.理论上证明了该算法的全局收敛性.给出的数值算例表明所提算法在非线性时变系统建模方面的有效性.
|
| 关 键 词： | 神经网络 迭代学习最小二乘 非线性时变系统 建模 |
Modeling of Nonlinear Time-varying Systems Using Time-varying Neural Networks |
| |
| YAN Wei-li,SUN Ming-xuan.Modeling of Nonlinear Time-varying Systems Using Time-varying Neural Networks[J].Mini-micro Systems,2011,32(6):1228-1231. |
| |
| Authors: | YAN Wei-li SUN Ming-xuan |
| |
| Affiliation: | YAN Wei-li,SUN Ming-xuan(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China) |
| |
| Abstract: | A unified architecture of time-varying neural networks is presented for implementing unknown nonlinear time-varying mappings and modeling nonlinear time-varying systems.The weight training for the neural networks is taken as the time-varying parameters estimation of time-varying systems.Applying the iterative learning mechanism,this paper develops an iterative learning least squares algorithm for updating the weights along the iteration axis.The global convergence is established theoretically,and two numeri... |
| |
| Keywords: | neural networks iterative learning least squares nonlinear time-varying systems modeling |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
