Adaptive control of multi-variables nonlinear system based on artificial neural network

Abstract

In this paper, the adaptive control of a multivariable nonlinear system based on artificial neural networks is put forth. A three-layer diagonal recurrent neural network is used to identify the system. The neural net's structure is simple, the number of networks is small and it can also identify the system better. A three-layer feedforward neural network is applied to control the system. The method to train neural network is simple, so it improves response speed to desired inputs. The strategy is applied to the control of a nonlinear dynamic system. Simulation studies show its efficiency.