采用正交小波网络的非线性系统辨识方法

摘要

基于多分辨分析的递阶逼近思想，采用正交小波网络研究了输入样本空间分布非均匀时非线性系统的辨识问题.重点讨论了样本非均匀时网格系的设计问题，并给出了基于该网格系的在线递阶辨识算法.最后利用正交小波网络分别对非线性静态和动态系统进行了仿真辨识.%Based on the idea of multiresolution hierarchical approximation of wavelet, the identification of nonlinear system with sampled data dispersed irregularly in space is discussed by using orthogonal wavelet networks (OWN). The design criteria of the lattice structure are intensively discussed and jusified, and an on-line hierarchical identification algorithm is given. The OWN is used to identify nonlinear static and dynamic systems. Simulation results indicate that the proposed algorithm has superior performance than conventional methods.