基于RBF神经网络无刷直流电机调速系统  

Research on Speed Control System of Brushless DC Motor Based on RBF Neural Network

作　　者：胡云宝[1] 王加祥[1] 曹闹昌[1] 王瑛[1] 

机构地区：[1]空军工程大学工程学院,西安710038

出　　处：《微电机》2013年第1期63-66,共4页Micromotors

摘　　要：为提高无刷直流电机调速系统控制精度,提出了一种基于RBF神经网络的无刷直流电机调速控制新策略。该策略根据采样得到的电机转速和相电流,实时修正网络权向量,准确调节PWM波的占空比,控制电枢两端的平均电压的幅值,从而对电机进行调速控制。仿真和实验结果表明,系统超调量小、速度响应快,比传统的PID速度控制有更好的静态、动态特性和鲁棒性。A new approach was presented in this paper based on particle swarm RBF neural network in order to improve the speed control performance of brushless DC motor（BLDCM）. In this method, according to the speed of motor and phase current, connection weight of neural network was revised in time. Meanwhile, the duty cycle of PWM was adjusted, in order to control the mean voltage amplitude and the speed of brushless DC motor. Simulation and experimental results indicated that under the proposed algorithm the overshoot of the system was small and the speed response was fast. This proposed algorithm had better dynamic, static and robust than traditional PID control.

关 键 词：无刷直流电机 转速控制 RBF神经网络 脉宽调制 

分 类 号：TM361[电气工程—电机]