摘要
提出一种H_∞与神经网络混合自适应控制系统设计的新方法.对于一类不确定非线性系统,首先运用线性微分包含(LDI)的方法,逼近模型中的非线性部分;然后在考虑外部扰动的情况下,设计忽略不确定项的H∞线性跟踪控制系统参考模型;最后将设计好的H∞线性跟踪控制器用于控制实际的非线性不确定系统,系统状态及其与参考模型的状态误差作为在线神经网络的输入,动态调节网络权值以消除整个系统的不确定项.仿真示例证实了该设计方法的有效性.
A new design method of H∞ and neural network hybrid adaptive control systems is presented. For a class of uncertain nonlinear system, the neural network technology of linear differential inclusion (LDI) is employed to approximate to the nonlinear parts. The reference model of the H∞ linear track control system is designed while the uncertain items are ignored. The designed H∞ linear track controller is used to control the nonlinear uncertain system. The system state and the state error of the reference model are considered as the input of the neural network, dynamically adjusting the network weights to counteract the uncertain items of the whole system. The simulation results show that the presented method is valid.
出处
《控制与决策》
EI
CSCD
北大核心
2004年第6期616-620,625,共6页
Control and Decision
基金
国家自然科学基金重点项目(60234010)
国防基础科研项目(K1603060318).
关键词
神经网络
线性微分包含
H∞线性跟踪控制器
Lyapunov methods
Neural networks
Nonlinear control systems
Optimal control systems
Uncertain systems
