神经网络变结构在液压伺服系统控制中的应用

Neural networks and variable structure controller applied in hydraulic servo systems

采用神经网络对液压伺服系统进行离线辩识,然后用神经网络的逆模型抵消伺服阀的死区非线性,再用离散变结构控制器对液压伺服系统的线性部分进行控制,从而达到良好的控制效果.

Asatisfactory result is achieved by applying a neural network to identify the plant down-line, by then using an inverse of the dead zone nonlinear for compensation, and finally by using discrete variable structure controller to control the linear part. We use an ANN instead of an inverse of the dead zone nonlinear.