Fault Diagnosis Method for Rolling Bearings Based on Fractional Variational Mode Decomposition

Abstract

The invention discloses a rolling bearing fault diagnosis method based on fractional-order variational mode decomposition, and relates to the technical field of bearing fault diagnosis methods. The method includes the following steps: extracting the vibration signal of the rolling bearing, and processing the vibration signal to obtain a multi-component LMF vibration signal; performing fractional variational mode decomposition on the multi-component LMF vibration signal to obtain the mode of the vibration signal The vibration signal is reconstructed according to the modal component of the obtained vibration signal; the reconstructed vibration signal is subjected to wavelet transformation to obtain the instantaneous frequency conversion curve, and the fault information of the rolling bearing is obtained according to the instantaneous conversion frequency curve. The proposed method is successfully applied to the analysis of rolling bearing inner ring fault data, which can accurately extract the instantaneous fault characteristic frequency and has strong noise robustness.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis methods for rolling bearings, in particular to a rolling bearing fault diagnosis method based on fractional variational mode decomposition. Background technique [0002] Bearing is one of the core components of a locomotive. It has the characteristics of complex structure, high speed, and heavy load. It is a typical nonlinear system. There are various types of bearing faults, and they are prone to damage. According to statistics, 40% of the used locomotive bearings are tested every year, and 33% of the bearings need to be replaced. And any small failure of locomotive bearings may bring huge losses to the country and the people. In engineering practice, most of them use the analysis of bearing vibration signals for fault diagnosis. However, the vibration signals collected by sensors will be accompanied by a large number of interference signals, making the signals weak and unstable. Theref...

AI Technical Summary

Problems solved by technology

In the prior art, there is also an algorithm (EMD-FRFT) combining EMD and fractional Fourier transform (EMD-FRFT) to suppress high-power interference in tracking radar signals, which has achieved certain results, but the filtering effect is not good when the interference frequency is high. Obviously; an integrated empirical mode decomposition algorithm (EEMD-FRFT) based on fractional Fourier transform is also proposed in the prior art to detect and estimate parameters of multi-component chirp signals, but due to the integrated empirical mode decomposition Due to the limitations of the algorithm, the mode mixing problem has not been well solved

Images