Single Neuron PID Adaptive Control For PM Brushless DC Motors Based On RBF Neural Network On-line Identification

The Permanent magnet brushless DC motor has broad applied perspectivebecause of its eminent performance. The traditional analysis and designing methodhave been development maturely, It can be said that the wider applications ofpermanent magnet brushless DC motor depend somewhat upon the further research onits special control method.Conventional PID control is widely adopted in many fields because of its simplestructure, high reliability and easily implementation. PID controller has good controleffect if the parameters of system model have not big variation, but when controlledobject has high nonlinear trait and high uncertenty, PID control can't achieve goodresult. For Permanent magnet brushless DC motor, it has high nonlinearized trait, PIDcontroller with fixed parameter can't achieve good performance index. This paperpresents a novel approach of single neuron adaptive control for permanent magnetbrushless DC motor based on RBF neural network on-line identification. The methoduses single neuron to construct the adaptive controller of Permanent magnet brushlessDC motor, and has the advantages of simple construction, adaptability and robustness.A RBF network is built to identify the system on-line, and then constructs the on-linereference model, and implements self-learning of controller's parameters by singleneuron controller, thus achieve on-line regulation of controller's parameters. Thisalgorithm combined the merit of single neuron adaptive controller and RBF network.RBF network ameliorate single neuron adaptive controller and made the responsespeed of the system faster;and single neuron adaptive controller advanced theaccuracy of the system. Excellent flexibility and adaptability as well as high precisionand good robustness are obtained by the proposed strategy. The high-speed digital signal processor (DSP) has been gotten widely applicationin the field of motor control. At last of this paper, TI's TMS320LF2407A DSP is usedto build up the driving system and control system of permanent magnet brushless DCmotor. And a software system is worked out to achieve the function of currentsampling, rotor position detection, speed control and current regulation.