采用小波变换和盲源分离的电机轴承故障诊断

针对地铁车辆转向架牵引电机轴承故障,提出一种结合小波变换和盲源分离的轴承故障识别方法.首先从原始信号频谱中判断轴承高频共振信号的大致频带范围以确定小波分解层数;其次利用小波分解提取轴承高频共振信号成分;然后利用盲源分离方法从小波分解后的重构高频信号中分离故障特征信号;最后对故障特征信号进行Hilbert解调并通过包络谱分析提取故障特征频率.对上海地铁某型地铁车辆转向架牵引电机进行试验,试验结果证明该方法能清晰、准确地识别轴承故障特征.

A Fault Diagnosis Method for Motor Bearings Based on Wavelet Transform and Blind Source Separation

In order to solve the fault problem of traction motor bearings of bogie in metro vehicle,a fault diagnosis approach based on wavelet transform and blind source separation is proposed.First,the resonance frequency band of the bearing is estimated by frequency spectrum analysis to determine the decomposition level in wavelet transform.Second,the high frequency components are extracted from the initial signals by using wavelet decomposition.Third,the mixed signal is separated into independent components by blind source separation.Finally,independent components separated from the mixed signal are processed by Hilbert demodulation to obtain their envelops,and then the frequencies of the fault characteristics of the bearing are extracted by the envelope spectrum analysis.A test on the traction motor bearings of bogie is performed in Shanghai Metro,and the results show that the proposed method can clearly and accurately identify the fault characteristics.