一类非严格反馈大系统的自适应神经网络控制  

Adaptive Neural Network Control for a Class Non-Strict-Feedback Large-Scale Nonlinear Systems

作　　者：时昊天 陈兵[1] 孙莉莉 SHI Haotian;CHEN Bing;SUN Lili(Institute of Complexity Science,Qingdao University,Qingdao 266071,China)

机构地区：[1]青岛大学复杂性科学研究所,山东青岛266071

出　　处：《青岛大学学报（工程技术版）》2018年第4期16-26,共11页Journal of Qingdao University(Engineering & Technology Edition)

基　　金：国家自然科学基金资助项目(61473160;61673227)

摘　　要：针对一类具有非严格反馈模块的非线性互联大系统,本文提出了一种输出反馈控制方案。首先使用神经网络来逼近未知系统函数,然后借助向量的范数性质处理非严格反馈模块。由于系统的状态不可测,所以建立观测器来估计未知状态。同时,结合自适应控制策略和Backstepping方法,设计出一种自适应神经网络分散输出反馈控制器,并利用Lyapunov稳定性理论进行稳定性分析。结果表明,在该控制策略作用下,闭环系统所有的信号保持半全局有界,且系统输出可以很好地跟踪给定的参考信号,通过仿真算例来验证本文设计控制策略的有效性。仿真结果表明,系统的输出能很好地跟踪给定的追踪信号;系统的控制信号和自适应参数保持在0的邻域内;设计的观测器能有效地观测原互联大系统的系统状态。说明本文设计的控制策略能够很好地运用在具有非严格反馈模块的互联大系统中。该研究更具有一般性和通用性。In this paper, an adaptive neural networks （NN） decentralized control scheme is proposed via output feedback for a class of large-scale nonlinear systems with non-strict feedback form. Firstly neural networks are utilized to approximate the unknown nonlinear functions. Then a property of the norm of basis vector function is used to overcome the difficulty caused by non-strict feedback structure. Since state variables of the system are unmeasured, a state observer is designed to estimate the unmeasured states. Meanwhile by combing adaptive control principle and backstepping technique, adaptive NN decentralized output feedback control laws are developed. At last the stability of this system has been proved by the Lyapunov stability theory. It is shown that all the signals in the closed-loop systems are semi-globally uniformly ultimately bounded, while the system outputs can track the reference signals as closely as possible. Using a simulation example to check the effectiveness of the proposed approach, the simulation results demonstrated that the system outputs can track the reference signals as closely as possible, while the control signals in the system and the adaptive parameters converge to a small enough neighborhood of origin. And the designed state observer can effectively observe the system states of the large-scale systems. It shows that the control scheme in this paper can be well applied to a class of large-scale systems with non-strict feedback form. And the scheme is more general and versatile.

关 键 词：大系统 非严格反馈 分散自适应控制 神经网络 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置]