基于神经网络的非线性系统在线自校正控制及其应用

利用神经网络给出了设计非线性控制系统自适应控制器的方法，主要包括基于神经网络的非线性系统的在线直接自校正和间接自校正控制和满足局部能控一般非线性系统的神经网络自校正控制。从研究结果可以看出，由于神经网络的权系数的可调性以及神经网络对非线性的逼近性，利用神经网络所设计的自适应（或自校正）控制器，其闭环控制系统具有良好的动态与静态响应特性。所提供的控制器都可在线实现

ON LINE SELF CORRECTING CONTROL FOR NONLINEAR SYSTEMS BASED ON NEURAL NETWORKS AND ITS APPLICATION

Some design methods of self correcting control based on neural networks for nonlinear systems are discussed. First, an overview of design of direct and indirect self correcting control based on neural networks is given, then the on line design of self correcting controller based on neural network for general nonlinear system which satisfy the local controllable conditions is provided. The result shows that the closed loop of self correcting controller designed by using neural network has good transient and steady responses.