Intelligent Neural Network Controller Optimization and Simulation Using GA

摘要

Conventional PID controller parameter tuning method needs the precise mathematical model of the controlled object, while the fuzzy control and neural network have strong self-adaptive and self-learning ability. The genetic algorithm is a new global optimization method so they can be used to design an adaptive PID intelligent controller based on fuzzy neural network and GA. First, GA is adopted to optimize the central value and width of the membership function. Then, we use BP to optimize the connection weight coefficient of fuzzy neural network to achieve adaptive and intelligent control of PID. The simulations indicate such scheme improve the adaptive ability and anti-interference ability of the system, which also enhances the robustness of system.