基于EWT-MCKD的机车轮对轴承故障诊断

Fault diagnosis of locomotive wheel set bearing based on EWT-MCKD

针对机车轮对轴承在实际运行过程中故障特征难以提取的问题,提出经验小波变换(Empirical Wavelet Transform,EWT)和最大相关峭度解卷积(Maximum Correlated Kurtosis Deconvolution,MCKD)相结合的滚动轴承故障特征提取方法。对原始信号进行傅里叶变换得到Fourier频谱图,根据频谱中的极大值将Fourier频谱图进行分段得到若干模态分量,以无量纲的裕度指标作为评价指标,再采用最大相关峭度解卷积对裕度因子最大的模态分量进行降噪处理。通过分析其包络谱中的频率成分来实现故障诊断。研究结果表明:所提方法对不同故障类型的轮对轴承进行诊断,可以准确有效的识别轮对轴承故障类型,具有一定的工程实用价值。

Aiming at the problem that it is difficult to extract the fault features of train wheel bearings during actual operation,a rolling bearing fault feature extraction method combining empirical wavelet transform(EWT)and maximum correlation kurtosis deconvolution(MCKD)is proposed.First,the Fourier spectrum was obtained by Fourier transform of the original signal,and then the Fourier spectrum was segmented according to the maximal value in the spectrum to obtain a number of modal components,with the dimensionless margin index was used as the evaluation index,and then the maximum correlated kurtosis deconvolution was used to reduce the noise of the modal component with the largest margin factor,and finally the fault diagnosis was realized by analyzing the frequency components in its envelope spectrum.The proposed method was applied to the diagnosis of wheelset bearing with different fault types,and all show that the method can accurately and effectively identify the types of wheelset bearing faults,and has certain engineering practical value.