一类非严格反馈切换系统的自适应神经网络控制  

Adaptive Neural Control for a Class of Non-Strictly Switched System

作　　者：刘志梁[1] 徐凯[2] 陈兵[1] 

机构地区：[1]青岛大学复杂性科学研究所,山东青岛266071 [2]青岛卷烟厂,山东青岛266071

出　　处：《青岛大学学报（工程技术版）》2017年第2期1-6,11,共7页Journal of Qingdao University(Engineering & Technology Edition)

基　　金：国家自然科学基金资助项目(61473160)

摘　　要：针对具有非严格反馈的非线性系统的控制问题,本文主要研究非严格反馈形式的单输入单输出非线性切换系统的控制问题。运用自适应神经网络控制方法,逼近系统的组合非线性函数;同时,结合Backstepping方法设计神经网络控制方案,利用神经网络的结构性质简化设计过程,成功的将神经网络自适应Backstepping设计方法拓展到该类非严格反馈系统上,最后通过仿真例子验证本文所提控制方法的有效性。仿真结果表明,在任意切换信号及所给控制器的作用下,保证了良好的跟踪性能,并保证闭环系统所有状态是半全局一致最终有界的,跟踪误差收敛到原点的一个残差集内。该研究具有一定的实用价值。Aiming at the control problem of nonlinear systems with non-strict feedback, this paper mainly stud-ies the control problem of non-strict feedback single-input single-output nonlinear switching systems. The a-daptive neural network control method is used to approximate the nonlinear function of the system. A t the same time, the Backstepping method is used to design the neural network control scheme. By using the struc-tural properties of the neural network, the design process is simplified, and the neural network adaptive Back- stepping design method is extended to the non-strict feedback system. The simulation results show that the tracking performance is ensured by the arbitrary switching signal,and all the states of the closed-loop system are guaranteed to be semi-globally stable. The tracking error converges to a residual set of the origin. The study has some practical value.

关 键 词：切换系统 神经网络 非严格反馈 自适应 非线性 BACKSTEPPING 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程] TP183[自动化与计算机技术—控制科学与工程]