Neural Network Inverse Control Of Speed-regulating System For BLDCM

Accompanied by the development of modern industry, brushless DC motor (BLDCM) speed-regulating system is widely applied in factory yield. The brushless DC motor adopts electronic commutation instead of the mechanical commutation, and has been extensively used in many industrial domains, such as aviations, robots, digital machine tools, and in medical or home appliances. The brushless DC motor has excellent characteristics of good start and timing performance, but it is a nonlinear system owing to commutation process and armature reaction. So it is significant to develop such a control system with quick response, powerful regulation capability and high precision. With the financial aid of National Nature Science Fund (60874014). China Ministry of Education Fund (20050299009) and Natural Science Fund of Jiangsu Province (BK2007094), brushless DC motor speed-regulating system based on neural network inverse was proposed and studied performance of speed-regulating and load disturbance.Firstly, the reversibility of brushless DC motor was testified. On the basis of reversibility analysis of original system, the inverse model approximated by the dynamical BP neural network was cascaded with the original system.Secondly, based on MATLAB, the model of this brushless DC motor system was constructed by S-FUNCTION, and then the feasibility of control method based on neural network inverse was testified by simulation.Thirdly, the hardware of the system used the TMS320F2812 as the main chip, the inverter with BUCK converter made up of the drive circuit. The software design mainly came from neural network inverse control method which contained the implementation of neural network and dynamic integrators. By each hardware and software parts of system debugging, the experiment of BLDCM control system was accomplished. Experiment results showed that the proposed scheme reduced overshoot, preserved fast speed response merit, and showed robust to load disturbance. Neural network inverse control method effectively improved dynamic and static operation performance and was a novel control method for brushless DC motor.Finally, the paper introduced the application of synchronized rectification technology in brushless DC motor system. In view of the low stability of the whole system that caused by the high-heating value had been given off by the driving circuit of the electric vehicle power driven system that worked in the actual working process, the synchronized rectification technology was applied to the driving circuit of the controller. The recovery diode using for freewheeling was replaced by mosfet with low break-over resistance. The loss of freewheeling and the heating had been reduced. So the stability and efficiency of drive circuit had been highly enhanced. The feasibility and superiority of this design was confirmed by experiments.