Duffing-Holmes型振子混沌系统的控制  

The Control of DuffingHolmes Oscillator Chaotic System

作　　者：周群利[1,2] ZHOU Qunli(School of Electrical and Automation,Wuhu Vocational and Technical College,Wuhu 241006,China;Hefei Institute of Intelligent Machines,Chinese Academy of Science,Hefei 230031,China)

机构地区：[1]芜湖职业技术学院电气与自动化学院,安徽芜湖241006 [2]中国科学院合肥智能机械研究所,安徽合肥230031

出　　处：《玉溪师范学院学报》2024年第3期1-7,共7页Journal of Yuxi Normal University

基　　金：安徽省高校优秀拔尖人才培育项目(项目编号:gxgnfx2021190);安徽高校自然科学研究重点项目(项目编号:KJ2020A0911);2021芜湖职业技术学校质量工程电力电子技术教学示范课(课堂革命)。

摘　　要：对Duffing-Holmes型振子混沌系统进行动力学分析,发现其内部蕴含着丰富的动力学特性,在适当的参数取值下会体现出混沌系统的特性.采用自适应控制方法对Duffing-Holmes型振子混沌系统进行控制,利用RBF神经网络逼近系统模型中的非线性项,通过设计合适的控制器和参数自适应律,从理论上对此受控系统进行了稳定性分析,并进行系统数值仿真,受控系统变量能够在很短时间内跟踪期望目标值.而且和采用输入-状态线性化控制方法及基于RBF神经网络逼近非线性项的滑模控制方法相比较,具有快速性更好、稳定性更高的特点.A dynamic analysis of DuffingHolmes oscillator chaotic system was conducted,revealing that the system is characterized by an extensive range of dynamic traits.With suitable parameter settings,these traits can exhibit the properties of a chaotic system.The adaptive control method was used to control the chaotic system with the RBF neural network to approximate the nonlinear term of the system model.A tailored controller and parameter adaptation rule were designed to theoretically assess and numerically simulate the stability of the controlled system,enabling system variables to rapidly follow desired target values.Compared with the control methods through inputstate linearization and the sliding mode based on RBF neural network approximating nonlinear term,the approach presented in this paper,distinguished by its rapid response and high stability,has achieved the set control goals and demonstrated its validity.

关 键 词：RBF神经网络 Duffing-Holmes型振子 控制律 自适应律 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]