Nero-Fuzzy Controller Design for Permanent Magnet DC Motor

Abstract

This thesis presents a neuro-fuzzy controller design for speed control of Direct Current (DC) motor. The thesis scope includes the simulations and modeling of DC motor, Fuzzy Logic Controller (FLC), neuro fuzzy controller and conventional Proportional-Integral-Derivative (PID) controller as benchmark to the performance of fuzzy system. The most commonly used controller for the speed control of DC motor is the conventional PID controller. Fuzzy logic controller and neuro-fuzzy control are proposed in this study. The performances of the two controllers are compared with PID controller performance. Classical control theory is based on the mathematical models that describe the physical plant under consideration. In this thesis, neural networks are used in to solve the problem of tuning a fuzzy logic controller. The neuro fuzzy controller uses neural network learning techniques to tune membership functions. Comparison between the PID output, FLC output and the neuro fuzzy output was done on the basis of the simulation result obtained by MATLAB/SIMULINK. The model Performance of neuro-fuzzy controller is better compared to FLC and PID controller.