The Position Tracking Control Of BLDCM Based On Fuzzy Neural Network

Brushless DC Motor(BLDCM) is a kind of mechatronics products, which not only reserves many excellent control performance and mechanical characteristics of DC Motor, but also has many good performances, such as high efficiency, high torque, small volume, small inertia, long life-span and sparkless commutation. In aerospace, industry production and other fields, in order to satisfy the demand of high accuracy on position control in brushless DC Motor servo system, this thesis studied and designed a brushless DC Motor servo control system based on fuzzy neural network control.First, this paper introduced the composition structure of brushless DC motor, then analyzed the working principles of brushless DC motor and established the mathematical model of brushless DC motor. Aiming at the features of multivariate, non-linear, strong coupling and time varying brushless DC motor servo system, this paper presented an intelligent control algorithm characterized by fuzzy logic reasoning, self-learning and self-adaptation based on fuzzy neural network which integrated the advantages of the fuzzy control and neural network control. This paper designed a kind of compensation fuzzy neural network (CFNN) structure which has compensation ability based on the research of the conventional fuzzy neural network. Under the circumstance of different position signal, input and output signal disturbing and motor parameter variation, this paper made the simulation of the traditional PID control, the conventional fuzzy neural network control and the compensation fuzzy neural network control with MATLAB. The simulation results showed that the compensation fuzzy neural network control (CFNNC) improved the steady performance, dynamic performance and the anti-interference ability of brushless DC motor position servo system.On the basis of the theoretical research and simulation, this paper designed three closed-loop control systems which included the position detection link, the speed detection link and the electric current detection link of the DSP hardware experimental platform. Position loop used compensation fuzzy neural network control. Speed loop and current loop used PI control. The experimental results showed a performance of superiority and feasibility of the compensation fuzzy neural network control (CFNNC) in position servo system of brushless DC motor.