A Bearing Fault Diagnosis Method Based on Sparse Theory

Abstract

The present disclosure discloses a bearing fault diagnosis method based on sparse theory, including: decomposing the collected original vibration signal X(t) of the bearing into N eigenmode functions through empirical mode decomposition, and calculating each eigenmode function mode function c i Energy ratio of (t) γ i And in order from large to small, select the eigenmode function corresponding to the maximum energy ratio as the analysis signal x(t); use the analysis signal x(t) to extract the i-th order natural frequency f in the bearing ni evoked free decay response x i (t), through the optimization algorithm from this free decay response x i (t) extract the modal parameters; use the impulse response function of the mass-spring second-order damping system as the atomic model, collect the extracted modal parameters to form a parameter set, input the parameter set into the atomic model, and construct a sparse dictionary ; use the matching pursuit algorithm in combination with the constructed sparse dictionary to solve the reconstructed signal, and perform envelope demodulation analysis on the reconstructed signal to form an envelope spectrum. When there is a fault characteristic frequency of the bearing in the envelope spectrum, then Realize fault diagnosis.

Description

technical field [0001] The disclosure belongs to the field of fault diagnosis, and in particular relates to a bearing fault diagnosis method based on sparse theory. Background technique [0002] As a key component, rolling bearings play an important role in rotating machinery, and their failure directly affects the stable operation of the equipment, and sometimes even causes equipment damage and shutdown. Due to the particularity of its working environment, for example, many mechanical equipments operate under variable loads, high temperature and other complex working conditions for a long time. The interference of noise reduces the signal-to-noise ratio of the vibration signal and increases the difficulty of extracting fault features. [0003] The key to fault diagnosis of rolling bearings is to extract hidden fault features from vibration signals containing harmonic and noise interference. Traditional spectral analysis methods such as Fourier transform can effectively ex...

AI Technical Summary

Problems solved by technology

The two key problems of sparse representation lie in the design of sparse dictionary and the solution of sparse coefficients. Too large a sparse dictionary can guarantee a certain degree of redundancy, but it will reduce the solution efficiency. If it is too small, it can improve the solution efficiency but reduce the matching accuracy.

[0005] In the previous fault diagnosis based on sparse representation, no research has been found on the method of constructing a sparse dictionary by identifying the modal parameters of the system

Images