PID Neural Network based Model Reference Adaptive Control for Nonlinear Uncertain System

摘要

In this paper, a scheme of a model reference adaptive control for a class of nonlinear uncertain system is described. The control system is composed of three parts-a system controller, a nonlinear object identifier and a nonlinear reference model. It uses a PID neural network as the system controller, and an artificial neural network as the nonlinear object identifier. Then, the paper gives out the online/offline training algorithm of the two artificial neural networks and the control strategy of the whole system. Finally, a simulation example on the above scheme is presented. Results show the effectiveness of this control scheme.