Bearing fault diagnosis method based on multi-channel vibration characteristics

Abstract

The invention provides a multi-channel vibration characteristic-based bearing fault diagnosis method, which comprises the following steps of: acquiring an original signal to obtain an original vibration signal sample set; performing time domain processing on the original vibration signal sample set to obtain a time domain signal sample set; performing time-frequency transformation on the time-domain signal sample set to obtain a multi-channel frequency spectrum signal; intercepting a fault characteristic frequency by using the multi-channel frequency spectrum signal and constructing sample characteristic graphs, wherein the sample characteristic graphs form a fault diagnosis sample set; and constructing a fault diagnosis model by using the fault diagnosis sample set, inputting the feature graph of the to-be-diagnosed data into the fault diagnosis model, and outputting a diagnosis result by the fault diagnosis model. According to the method, the complexity of time-frequency analysis can be reduced, the influence of subjective factors of personnel on a diagnosis result is reduced, the specific fault frequency information of the bearing and machine learning are fused, the calculation amount and the over-fitting risk are reduced, and the diagnosis accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of bearing fault judgment, in particular to a bearing fault diagnosis method based on multi-channel vibration characteristics. Background technique [0002] With the development of modern industry and the continuous improvement of the level of science and technology, electromechanical equipment is constantly developing in the direction of large-scale, high-speed, continuous, centralized, automated and precise, and its composition and structure are also becoming more and more. Complexity, which directly leads to an increase in the failure rate and extremely difficult diagnosis, among which key components such as rolling bearings, if some minor damage or anomalies are not detected and eliminated in time, it may cause the failure, paralysis of the entire system, and even lead to catastrophic consequences . [0003] The fault diagnosis of the bearing in the prior art is mainly based on the basic theory of vibra...

AI Technical Summary

Problems solved by technology

[0004] The invention solves the problems that it is difficult to obtain a clear frequency spectrum during bearing fault diagnosis, and the fault frequency ratio is affected by subjective factors of personnel, resulting in missed judgments and misjudgments. It proposes a bearing fault diagnosis method based on multi-channel vibration characteristics, which can reduce time-frequency analysis. reduce the influence of human subjective factors on the diagnosis results, integrate bearing-specific fault frequency information and machine learning, reduce the amount of calculation and the risk of over-fitting, and improve the accuracy of diagnosis

Images