Bearing fault diagnosis method based on cross-correlation singular value

Abstract

The invention discloses a bearing fault diagnosis method based on cross-correlation singular values, and relates to the technical field of bearing fault signal diagnosis methods. The method comprises the following steps: performing singular value decomposition on the bearing vibration signal reconstructed in phase space to obtain a singular value component signal of the bearing vibration signal; selecting two singular value component signals for cross-correlation, using a cross-correlation function to reduce noise and Highlight the characteristics of the same frequency signal to reduce the noise of the signal, and highlight the fault characteristics to obtain the cross-correlation time-domain signal; select the interval with obvious impact characteristics in the cross-correlation time-domain signal for envelope demodulation, and extract the fault characteristics of the bearing vibration signal frequency. The method can overcome the problem of missing high frequency multiplier features under strong noise background, accurately extract bearing fault characteristic frequency, and can accurately obtain high multiplier frequency.

Description

technical field [0001] The invention relates to the technical field of a bearing fault signal diagnosis method, in particular to a bearing fault diagnosis method based on a cross-correlation singular value. Background technique [0002] In recent years, Singular Value Decomposition (SVD) has been widely used in many fields such as data compression, data mining, signal speech coding and digital watermarking, robot motion control, and filter algorithm design. As a nonlinear filtering method, SVD is also a research hotspot in the field of fault diagnosis. SVD processes signals in a high-dimensional space through a matrix, which can effectively filter out noise and extract signal features. Diagnostic research has also achieved a large number of research results. [0003] The problem of effective singular value determination is the hot spot and difficult point of SVD, which determines the signal processing effect of SVD, but this problem has never been well solved. Most of the ...

AI Technical Summary

Problems solved by technology

Leng Yonggang from Tianjin University used the envelope spectrum to analyze the singular value components, and was able to extract the fault frequency, but the problem of noise interference still existed

Images