Rolling bearing fault diagnosis method

Abstract

The invention discloses a rolling bearing fault diagnosis method. The fault diagnosis method comprises the following steps: setting each fault reference probability distribution Yi; respectively solving the optimal transport distance between each signal Xi to be measured and each fault reference probability distribution Yi; establishing an optimal transport distance matrix, and taking the optimaltransport distance matrix as a fault feature matrix T; and performing classification judgment on the fault feature matrix T by using a classifier and outputting a diagnosis result. According to the method, the bearing fault can be effectively diagnosed under the condition of fewer samples and incomplete samples, the accuracy is high, the consumed time is short, the real-time performance is good, and the method is not limited by a classifier; the bearing fault diagnosis method can accurately predict faults under unknown working conditions, and solves the problem of bearing fault diagnosis undercomplex working conditions in the actual operation process.

Description

technical field [0001] The invention relates to the field of mechanical faults, in particular to real-time fault diagnosis of bearing vibration signals. Background technique [0002] The fault diagnosis of rolling bearings is mainly to provide support for reliable bearing operation. The existing methods are mainly divided into four categories: time domain analysis, frequency domain analysis, time-frequency analysis and methods based on intelligent optimization algorithms. Time domain analysis is According to the time waveform of the signal, analyze the composition and characteristic quantities of the signal. Commonly used time domain indicators include peak value, effective value, pulse, margin, kurtosis, etc., but their precision is not good and they are easily affected by noise; frequency domain analysis method It is to transform the fault signal into the frequency domain. By analyzing the spectrum, the commonly used ones are Fourier transform, spectral kurtosis, cepstrum,...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is that in the process of bearing fault diagnosis, especially when the bearing is working in an unknown working condition, the accuracy rate is not high, the training time is long, and the real-time performance is poor. It provides a rolling bearing fault diagnosis based on the optimal transportation theory. diagnosis method

Images