Fault fusion diagnosis method for rolling bearing based on improved D-S evidence theory

Abstract

The invention provides a fault fusion diagnosis method for a rolling bearing based on an improved D-S evidence theory. The method comprises the steps of: carrying out classification identification ondata of different fault states by adopting a plurality of fault diagnosis methods, and reflecting different fault characteristics of the rolling bearing from multiple angles to obtain a plurality of primary diagnosis results; constructing a correlation matrix between each evidence by utilizing a conflict factor in the D-S evidence theory, and calculating reliability of each evidence; according tothe reliability of each evidence, classifying similar evidences and conflict evidences by adopting a nearest neighbor rule; retaining probability distribution of similar evidences while modifying basic probability distribution of conflicting evidences; and fusing similar evidences and modified conflict evidences by adopting a D-S combinational rule to obtain a final diagnosis result. According tothe fault fusion diagnosis method for the rolling bearing based on the improved D-S evidence theory, fault diagnosis is performed from multiple angles, then multiple decisions are fused, advantages ofeach diagnostic method can be fully preserved, at the same time one-sidedness caused by single diagnostic method is preserved substantially, so that fault diagnosis rate and diagnosis reliability areimproved.

Description

technical field [0001] The invention relates to the technical field of bearing detection, in particular to a rolling bearing fault fusion diagnosis method based on the improved D-S evidence theory. Background technique [0002] The generator is the core component of the wind power generating set, and its operation reliability directly affects the stability of the generating set, and the rolling bearing is a key part of the generator, so it is of great significance to carry out fault diagnosis on the rolling bearing. At present, for rolling bearing fault diagnosis, different signal processing techniques (such as Fourier transform, wavelet transform, EMD decomposition, etc.) are usually used to extract features to construct fault feature sets, and then different pattern recognition algorithms are used to identify and realize automatic diagnosis. This fault diagnosis method of constructing feature sets first and then pattern recognition has achieved good diagnostic results, but...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the defects in the prior art, the present invention provides a rolling bearing fault fusion diagnosis method based on the improved D-S evidence theory, which solves the problem in the prior art that the fault classification method can only be sought from a single perspective

Images