Research On Fault Diagnosis For Bearings And Gearboxes Of Urban Rail Transit Trains

Bearings and gearboxes,as key components in the urban rail transit train,play an important role in the normal operation of rail transit trains.If the mechanical equipment fails,it will not only affect the smoothness of the operation of the rail transit train,but also pose a threat to the safety of train operation.Therefore,it has a great significance to timely and effectively detect and diagnose the failure of bearings and gearboxes in the running part of rail transit trains.In reality,most of the rail transit train bearing and gearbox fault signals contain complex component information.Existing fault detection methods are mostly flawed when used to analyze actual fault signals.In this research background,this paper proposes a fault detection analysis method based on Empirical Wavelet Transform(EWT).The method uses the Empirical Wavelet Transform(EWT)method to decompose the fault signal,and then filters out the Intrinsic mode function(IMF)containing the sensitive fault information according to the eigenvalues calculated by the kurtosis formula proposed in the paper.The selected Intrinsic mode function(IMF)was analyzed by fast kurtosis(FK)method.In this paper,the bearing fault data and gearbox fault data are detected and analyzed.Compared with the existing fault detection methods,it is proved that the fault frequency information can be analyzed and the error rate is lower than the existing fault detection method.In this paper,the fault characteristic data of bearing fault data and gearbox fault data is extracted,and the fault information extracted by the existing method is used for classification experiments with the unprocessed original fault signal.The experimental results show that the fault eigenvalues extracted by this method can fully characterize the fault information.