Fault Diagnosis Method for Locomotive Wheelset Bearings Based on Approximately Complete Variable Mode Decomposition

Abstract

The invention discloses a locomotive wheelset bearing fault diagnosis method based on approximate complete variable mode decomposition. The locomotive wheelset bearing fault diagnosis method is characterized in that a local maximum value number of an original vibration signal frequency spectrum can be determined by adopting a local peak searching method, and can be assigned to a variable mode decomposition initial mode number; the original signal can be decomposed by adopting the variable mode decomposition algorithm, and the original signal can be reconstructed; the similarity between the reconstructed signal and the original signal can be determined according to the similarity criterion, and when the similarity is not able to satisfy the similarity criterion, the initial mode number can be increased, and the signal can be decomposed again; after the iteration is stopped, the modes satisfying the similarity criterion can be combined, and the wheelset bearing fault characteristics can be extracted by adopting the envelopment analysis method. The detection method is advantageous in that the result is reliable, the real-time performance is good, the use is simple and easy to realize, and the universality is strong, and in addition, the method is suitable for the fault diagnosis of the locomotive wheelset bearing.

Description

technical field [0001] The invention relates to a fault diagnosis technology of mechanical equipment, in particular to a fault diagnosis method for a locomotive wheel set bearing fault. Background technique [0002] As one of the key components of locomotives, the operation safety of wheel set bearings is particularly important. However, since the wheel-set bearings of the locomotive cage work for a long time under severe working conditions such as variable speed and heavy load, it is increasingly difficult to dynamically monitor, diagnose and predict the faults of the locomotive wheel-set bearings. Therefore, how to effectively analyze the collected vibration signals of locomotive wheel pair bearings, extract and highlight their fault feature information is a key issue in fault diagnosis. [0003] Existing methods of bearing fault feature extraction include wavelet transform and other methods, because the selection of basis functions and the determination of the number of ...

AI Technical Summary

Problems solved by technology

[0003] Existing methods of bearing fault feature extraction include wavelet transform and other methods. Due to problems such as the selection of basis functions and the determination of the number of decomposition layers relying on more practical engineering experience, it is difficult to meet the requirements of convenient and effective extraction of locomotive wheel set bearing fault features.

Although the traditional variable mode decomposition method can adaptively decompose the original signal and extract the fault feature information contained in it, it is affected by the number of preset modes: if the number of preset modes is too large, the fault features will be misclassified or over-classified. Decompose into different eigenmode functions; if the preset number of modes is too small, the fault characteristic information may be missed due to being hidden in the eigenmode functions affected by interference noise

Images