论文部分内容阅读
提出了一种基于前馈神经网络的智能ＩＭＣ，其设计过程分为两步进行：第一步，训练一个神经网描述对象响应；第二步，训练一个网络描述对象的逆，并将此网络作为ＩＭＣ控制器．经仿真实验表明，本智能控制系统鲁棒性强，优于常规ＩＭＣ系统，这类智能控制器适合于对象参数变化、模型不确定和非线性的控制．
An intelligent IMC based on feedforward neural network is proposed. Its design process is divided into two steps: the first step is to train a neural network to describe the object response; the second step to train the inverse of a network description object, and the network As IMC controller. The simulation results show that the intelligent control system is robust and superior to the conventional IMC system. Such intelligent controllers are suitable for the control of object parameters, model uncertainty and nonlinearity.