关键词: 异步电机, 轴承故障, Park 矢量, 小波包分解, 均方根值

Abstract: A bearing fault diagnosis method was proposed based on the Park vector modular square function and its data processing. The Park vector modular square function of the three-phase current signal of the motor bearing was calculated, and the Park vector modular square function was decomposed into the fourth and fifth layers by wavelet packet to obtain the frequency band corresponding to the fault characteristic frequency and took its RMS value. The bearing fault
diagnosis was carried out by comparing the RMS value with the normal motor after logarithmic processing. By solving the RMS value of the node coefficient corresponding to the wavelet packet decomposition subband, the motor fault was judged, and the prototype test platform was built. The test results verified the effectiveness of the method.

Key words: induction motor, bearing fault, Park vector, wavelet packet decomposition, root mean square( RMS)