Rolling bearing fault diagnosis method based on enhanced lightweight multi-scale CNN

Abstract

The invention provides a rolling bearing fault diagnosis method based on an enhanced lightweight multi-scale CNN. The method comprises the steps of: firstly carrying out multi-dimensional feature extraction on original vibration signals through a built CNN model; then inputting the original vibration signals into a lightweight multi-scale feature learning network to extract multi-scale features; carrying out selective enhancement on the learned multi-scale features by utilizing a discriminative fault feature enhancement mechanism (DFRM) so as to enhance fault features and weaken general features; fusing the enhanced multi-scale features; repeating the above steps for several times, mapping and inputting finally obtained high-level abstract features into a classifier for training; and finally performing fault identification on a to-be-detected sample according to the trained CNN model, thereby realizing fault diagnosis of a rolling bearing.; According to the method, the defect that a traditional CNN algorithm is insufficient in discriminative fault feature extraction capacity under the conditions of complex working conditions and strong noise interference is overcome, and the lightweight requirement of the industrial Internet of Things for a deep learning model is met.

Description

technical field [0001] The invention relates to the technical field of rolling bearing fault diagnosis, in particular to a rolling bearing fault diagnosis method based on an enhanced lightweight multi-scale CNN. Background technique [0002] Rolling bearings are one of the most widely used components in industrial production, and their health status is directly related to the safe and stable operation of mechanical equipment. However, in the long-term operation under complex working conditions of high speed and high load, rolling bearings are prone to failures such as wear and peeling. These failures will lead to rapid degradation of rotating machinery and even cause serious economic losses and casualties. Monitoring and fault diagnosis are very practical. [0003] At present, scholars at home and abroad have done a lot of research work on the fault diagnosis of rolling bearings, but in the face of complex and changeable working conditions and strong noise interference, the...

AI Technical Summary

Problems solved by technology

However, in the long-term operation under complex working conditions with high speed and high load, the rolling bearings are prone to failures such as wear and peeling. These failures will lead to rapid degradation of the rotating machinery and even cause serious economic losses and casualties. Monitoring and fault diagnosis are of great practical significance

[0003] At present, scholars at home and abroad have done a lot of research work on the fault diagnosis of rolling bearings, but in the face of complex and changeable working conditions and strong noise interference, the vibration signal is nonlinear, non-stationary and strongly coupled, and the fault characteristics are as follows: The time scale is diverse and not obvious, which poses a severe challenge to high-performance fault diagnosis based on vibration signals

Images