基于Park矢量旋转滤波的感应电机复合故障检测

A feature extracting method of stator and rotor faults of induction motor based on the park vector rotating filter

利用电压与电流基波频率相等,用单位电压Park矢量及其共轭对电流Park矢量进行旋转变换,从而将电流Park矢量的基波正序分量完全滤除而不影响其他的谐波分量.由于滤除了基波正序分量,在频谱图上,被基波正序分量掩盖的断条故障特征就显示出来了.由于去掉了基波的影响,也可以更清晰地突出偏心的故障特征.而且,旋转滤波不影响电流Park矢量中的负序分量,而电流基波的负序分量可以作为定子绕组故障的特征量.也就是说,通过对旋转滤波后的电流Park矢量的频谱进行分析,就可以对电机的定转子故障同时进行监测.实验结果表明,这是一种可行并且有效的电机故障实时监测方法.

As the current and voltage have the same basic frequency, the unit voltage Park vector and its conjugate are used to rotate the current Park vector. During the course of rotation, the fundamental positive sequence component in the current Park vector can be filtered out but the others remain. As a result, the feature frequency components of rotor broken bar fault, hiding behind the fundamental positive sequence component, emerge in the spectrum now. The feature frequencies of eccentricity are clearer than before for the same reason. Moreover, the negative sequence component of the fundamental in the current Park vector, which is the indicator of stator winding fault, has not been changed by the transfor- mation. Therefore, the features of rotor and stator faults arise in the spectrum of the current Park vec- tor in which the fundamental positive sequence component is filtered out. By analyzing the spectrum, the condition of the induction motor can be monitored on line. The results of experiment show that the pro- posed method is feasible and effective.