Application of Modified PID Neural Network to Discrete Nonlinear System Control

An improved PID neural network-based controller is designed and analyzed for the classes of single-input nonlinear system and multi-input discrete system. In order to deal with the local minimum problem in training neural network with back-propagation algorithm and to enhance controlling precision, neural network’s weights are adjusted by optimization algorithm. The controller employs a PID neural network instead of estimating the unknown plant nonlinearities on-line. When compared to other nonlinear modeling techniques for control purposes, it has several specific advantages that make it ideally suited to particular applications. Tow examples are used to demonstrate the performance and properties of the proposed scheme. The simulation results show that the proposed controller with improved PID neural network is flexible and efficient in the control of discrete nonlinear dynamic system.