Identification and Control in Nonlinear System Based on PID Neural Network

A control scheme based on PID neural network(PIDNN) is proposed to identify and control the nonlinear system for which traditional control methods can't acquire satisfying result.The PIDNN is a new type of dynamic feed-forward neural network which blends the PID control strategy into neural networks,so it has the merit of general PID controller for its simple construction and definite physical meaning of parameters,and also has the parallel structure,self-learning function and nonlinear mapping capability of neural network.The results of simulation show that the system has properties of low overshoot,quick response and good steady accuracy,and the controller can quickly track and effectively control the output of system with good adaptability and robustness and meet the need of real-time control.