Wiener系统的混沌引力搜索迭代辨识  

Chaotic Gravitational Search Iterative Identification for Wiener Systems

作　　者：徐珊玲 李俊红[1] 刘梦茹 华亮[1] Xu Shanling;Li Junhong;Liu Mengru;Hua Liang(Nantong University,Nantong 226019,China)

机构地区：[1]南通大学,江苏南通226019

出　　处：《系统仿真学报》2021年第9期2138-2146,共9页Journal of System Simulation

基　　金：国家自然科学基金(61973176);江苏省自然科学基金(BK20181457);江苏省六大人才高峰项目(XYDXX-038);江苏省高等学校自然科学基金(18KJB120007)。

摘　　要：Wiener非线性系统是由动态线性子系统串联非线性静态子系统而成,被广泛应用于自动控制、化工、电气等领域。考虑Wiener输出误差自回归(Wiener Output Error Autoregressive,Wiener OEAR)系统的辨识问题,提出了一种混沌引力搜索迭代辨识算法,通过在引力搜索算法中引入混沌优化机制来估计Wiener OEAR系统的未知参数,并证明了所提出的算法的收敛性。为证明所提辨识算法的有效性,采用了引力搜索算法和梯度迭代算法对该系统进行辨识。仿真结果表明:3种算法能够有效地对Wiener OEAR系统进行辨识,混沌引力搜索迭代辨识算法在参数估计精度方面要优于引力搜索算法和梯度迭代算法。The Wiener nonlinear system is composed of a dynamic linear subsystem and a series of nonlinear static subsystems,which is widely used in the fields of automatic control,chemical engineering,electrical and other fields.Considering the identification of the Wiener Output Error Autoregressive(Wiener OEAR)system,a chaotic gravitational search iterative identification algorithm is proposed,in which the chaotic optimization mechanism is introduced into the gravitational search algorithm to estimate the unknown parameters of the Wiener OEAR system and the convergence is proved.In order to show the effectiveness of the proposed identification algorithm,the gravitational search algorithm and gradient iterative algorithm are used to identify the same system,and a simulation example and an application example are given.The simulation results show that the three algorithms can effectively identify the Wiener OEAR system,and the chaotic gravitational search iterative identification is better than the gravitational search algorithm and gradient iterative algorithm in the accuracy of parameter estimation.

关 键 词：Wiener系统 系统辨识 参数估计 混沌机制 引力搜索 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]