Rolling bearing fault diagnosis method based on manifold learning and s-k-means clustering

Abstract

The invention discloses a rolling bearing fault diagnosis method based on manifold learning and s-k-means clustering, and belongs to the field of rotating machinery fault diagnosis. The method mainlycomprises the following five steps: acquiring vibration signals of the bearing in normal and fault states as original signals; obtaining a time domain feature set and a wavelet packet energy feature set; constructing a feature classification capability index by utilizing the scatter matrix, eliminating part of irrelevant features, and realizing feature selection; taking the feature set subjected to feature selection as high-dimensional input of a dimension reduction algorithm, and performing dimension reduction by using an NPE (neighborhood preserving embedding) manifold learning method to obtain a reduced low-dimensional feature set; and performing clustering analysis on the fault feature matrix by adopting an s-k-means clustering method to determine a fault type. The method is simple andeffective, effective reduction and secondary extraction of high-dimensional fault features are achieved, a small number of typical features closely related to faults are obtained, the calculated amount during feature classification is greatly reduced, and therefore the fault types are accurately recognized.

Description

technical field [0001] The invention belongs to the field of fault diagnosis of rotating machinery, in particular to a rolling bearing fault diagnosis method based on manifold learning and s-k-means clustering. Background technique [0002] As a key component of rotating machinery, rolling bearings play a key role in rotating machinery, and their working conditions will directly affect the overall performance of the entire mechanical equipment. Rolling bearing failure is one of the main reasons for the failure of rotating machinery equipment, and it may even lead to major property losses in severe cases. Therefore, in order to avoid mechanical failures caused by bearings and reduce economic losses, it is very necessary to carry out condition monitoring and fault diagnosis of bearings to ensure their normal operation. When a rolling bearing has a local defect failure, a series of shock signals will be generated, which is manifested as a sudden change in the vibration signal....

AI Technical Summary

Problems solved by technology

However, the K-means clustering algorithm relies heavily on the selection of the initial cluster center, is susceptible to the influence of isolated points, and needs to set the number of clusters in advance

Images