Performance Analysis of Fuzzy RBF Neural Network PID Controller for Simulation Turntable Servo System

Because simulation turntable servo system is highly nonlinear and uncertainty plants, a fuzzy neural network PID controller is proposed based on the Radial Basis Function (RBF). Up to now, various kinds of nonlinear PID controllers have been designed in order to satisfactorily control this system and some of them applied in actual systems with different degrees. Given this background, the step input and disturbance input simulation experiments are carried out based on MATLAB/SIMULINK tools to evaluate the performances of four different turntable servo controllers, including the conventional PID, the fuzzy self-tuning PID, the neural network PID and the fuzzy RBF neural network PID controller. For further comparison of the four PID controllers, the tracking curves of 2Hz sinusoidal signals and triangular wave signal are given. The comparison results show that the fuzzy RBF neural network PID controller can perform much better and make the tracking error arbitrarily small.