基于神经网络的集中绕组BLDC转子位置估计  

Rotor Position Estimation of BLDC With Concentrated Winding Based on Neural Network

作　　者：钟黎萍[1] 江恩祝[2] 张水平[1] 顾启民[1] 

机构地区：[1]常熟理工学院电气与自动化工程学院,江苏常熟215500 [2]国家海洋局东海分局,上海200137

出　　处：《微电机》2013年第4期84-87,共4页Micromotors

摘　　要：集中绕组的BLDC的高频电流响应中含有多重位置信息,直接运用外差法进行位置估计存在较大误差。由于这些位置信息的频率与转速相关,在低速和零速下无法用传统的滤波器将有用信号分离。利用神经网络进行处理,则无需考虑信号的频率结构,可直接映射出转子位置。实验结果证明了该方法的可行性。There are multiple rotor position information in the high frequency current response of BLDC with concentrated winding, and it would produce a large errors if directly using the method of heterodyne to estimate the rotor position. Furthermore, it is impossible to extract useful information through typical filter at low-speed and zero-speed because the frequencies of the signals are network is a good option for it can directly map out rotor ture. The related experiments were made and the results relative to the motor speed. In this condition, neural position regardl shown that the ess of the response＇s frequency strucmethod was practicable.

关 键 词：神经网络 旋转变换 位置估计 

分 类 号：TM361[电气工程—电机]