Back-propagation neural networks for the inverse control of discrete-time nonlinear plant

A backpropagation (BP) neural networks are applied to two kinds of inverse control methods for a discrete-time nonlinear plant. Two kinds of topological structure of BP neural networks are provided. Simulation results show that the tracking performance of a discrete-time nonlinear plant has been obtained satisfactorily in two kinds of input signal. In order to improve the tracking performance of the system, a combination of a variety of mapping functions is proposed. The learning and response of two networks have been developed in the Professional II plus neural networks. The program that connected in the neural networks is written in C language.