Abstract:In most cases, the sampling data in sample space are non-uniformly distributed, which will adversely affect the performance of the multiresolutional wave-nets (MWN). The modified multiresolutional wave-net presented here can efficiently process the non-uniform sampling data. First, the non-uniformly distributed data are located on the grid structure by introducing the rule of approximation location. The error resulted from this locating-grid method is analyzed, and its up-bound is given. Then a variable wave-net is proposed to improve the adaptability of the wave-nets. The structure and parameters of the variable wave-net can be adjusted according to the variation of the practical object. The training speed is remarkably improved, and approximating errors are reduced. Finally, the variable wave-net is used to simulate the identification of a nonlinear dynamic system. The simulation results showed that this improved method is effective and feasible.

Improved System Identification Approach Using Wavelet Networks

Nonlinear Dynamic System Identification Based on Wavelet Approximation

Nonlinear System Identification Using Orthogonal Wavelet Networks

A novel nonlinear system identification method based on robust wavelet network

WEIGHTED BAND-WISE ORTHOGONAL WAVELET NETWORK BASED ON GRID STRUCTURE FOR NONLINEAR SYSTEMS IDENTIFICATION

Orthogonal Wavelet Networks for Nonlinear Dynamic System Identification

A NOVEL METHOD FOR IDENTIFYING NONLINEAR SYSTEMS USING ORTHOGONAL WAVELET NETWORKS

Identifying nonlinear dynamic system using the improved multiresolution wave-net

Wavelet Networks Based General Multi-variable Non-linear System Identification Arithmetic and Its Application

Automatic Detection of Nonstationary Multiple Oscillations by an Improved Wavelet Packet Transform

A Weighted Band-Wise Identification Method for Multirate Sampled-Data Systems Based on Wavelet Multiresolution Analysis

A New Approach of Nonlinear System Identification Based on Wavelet Decomposition

Wavelet support vector machines and its application for nonlinear system identification

Nonparametric Identification Approach Based on Wavelet Approximation and Scale Transformation

System Identification Based on Wavelet and Nonlinear Autoregressive with Exogenous Inputs Model

Optimal Experiment Design for Wavelet-Based System Identification

Reproducing wavelet kernel method in nonlinear system identification

An Identification Method for Multirate Sampled-Data Systems Based on Wavelet Multiresolution Analysis

A Modelling Method Based on the Wavelet Transform and Artificial Neural Networks

Time-varying nonlinear system identification based on wavelet packet

Identification Of Time-Varying Nonlinear System Using Wavelet Decomposition