Optimal Neuro-Fuzzy D.C. Motor Speed Control

Abstract

This work deals with investigating of d.c. motor speed controlled by a buck-boost converter. Open loop system is tested. Aneuro-Fuzzy controller with random number and type of membership function is designed to control the speed of the d.c. motor as a closed loop system. Because of the lack of a clear and a known way for selecting the type and number of membership function in case of fuzzy control, An Adaptive Neuro-Fuzzy Inference System which comprises a fuzzy inference structure and neural network learning ability is modified to control the speed of the motor. The technique is used to select the optimal number and the best type of membership function for the fuzzy system. The process is carried out through testing four types of membership functions with different numbers (3,5,…etc) for each type and computing the absolute error for each case comparing their results to choose the smallest among them. Then the chosen root is applied to control the system for the rest time of control. The technique is applied to two loads (motors).