摘要
位置检测与换相准确与否,对无刷直流电机的运行有非常关键的影响.在分析反电动势过零检测原理的基础上得出线反电动势过零点与电机换相点及线反电动势与线电压之间的关系,从而得到线电压与转子位置之间的关系.由于电机运行过程中的参数变化及系统的非线性特征,直接通过线电压准确获得转子位置比较困难,因此构建了一个以三个线电压为输入,转子电角度为输出的自适应小波神经网络模型,并采用遗传算法优化小波神经网络结构。仿真实验证明,该方法辨识转子位置精度高,自适应性强,并能有效地控制电机换向.
The accuracy of position detection and commutation is very critical for the operation of the brushless direct current motor（BLDCM ） .Aimed at BLDCM rotor position detection ,the relationship between motor back electromotive force zero crossing point and motor commutation point as well as the relationship between line back electromotive force and line voltage was derived based on analysis of back electromotive force zero crossing detection principle ,so the relationship between line voltage and motor point was established .However ,because of the motor parameters variation during operation and non-linear characteristics of the system ,it is more difficult to obtain accurate rotor position through the line voltage directly .So the adaptive wavelet neural network model was proposed using line voltage as input and the rotor electrical angle as output ,and genetic algorithm was also adopted to optimize wavelet neural netw ork structure .Finally simulation and experiment show that the method has high precision to identify the rotor position ,and controls the motor commutator effectively with strong adaptability .
出处
《武汉工程大学学报》
CAS
2014年第10期66-70,共5页
Journal of Wuhan Institute of Technology
基金
甘肃省自然科学基金资助项目(0916RJZA039)
关键词
无刷直流电机
自适应小波神经网络
遗传算法
brushless direct current motor
adaptive wavelet neural network
genetic algorithm
