Title :
A New Method of PID Control Based on Improved BP Neural Network
Author :
Chunchao Shi ; Guoshan Zhang
Author_Institution :
Sch. of Electr. Eng. & Autom., Tianjin Univ., China
Abstract :
A new type of PID control method based on improved back-propagation (BP) neural network is proposed to deal with the defects of steepest gradient descent in slowly converging and easily sticking into local minimum frequently. It has merits of both neural network and PID controller, and it is adjusted by Fletcher-Reeves conjugate gradient, which can make study speed of network faster and can eliminate the disadvantages of steepest gradient descent in BP algorithm. The parameters of neural network PID controller are adjusted on line by the improved conjugate gradient. The programming steps under MATLAB are finally described. Simulation results show that the controller is effective.
Keywords :
backpropagation; gradient methods; neurocontrollers; three-term control; Fletcher-Reeves conjugate gradient; PID control; backpropagation neural network; steepest gradient descent; Automation; Control systems; Control theory; Electronic mail; Industrial control; MATLAB; Neural networks; Pi control; Proportional control; Three-term control; Back-propagation (BP) neural network; Improved conjugate gradient; PID control;
Conference_Titel :
Control Conference, 2006. CCC 2006. Chinese
Conference_Location :
Harbin
Print_ISBN :
7-81077-802-1
DOI :
10.1109/CHICC.2006.280598