摘要
提出了一种基于神经网络的PID控制方案 用神经网络辨识器与神经网络控制器构成间接自校正控制系统 其中 ,神经网络辨识器采用单隐层结构 ,其辨识算法采用预报误差法 (RPE算法 ) ;神经网络控制器为 2层的线性结构 ,其输入为系统偏差及其一阶和二阶微分 ,控制器具有增量型PID控制结构 将该控制方案应用于电阻炉的炉温控制中 。
A PID control scheme based on neural netwark is proposed.The selftuning control system consists of the NNC (neural network controller) and the NNI (neural network identifier).The NNI is made up of a network with one hidden layer and the RPE (recursive prediction error) algorithm is used to identify the plant on line.The NNC has the structure of 2layered neural network and has a mechanism similar to that of incremental form PID controller.The input of the controller is the system deviation and the one and two ladder differential.The control scheme is used to control temperature of resistance stove and a satisfied result is acquired.
出处
《包头钢铁学院学报》
2000年第1期53-56,共4页
Journal of Baotou University of Iron and Steel Technology
