Motor bearing fault diagnosis method based on feature optimization and GWOA-XGBoost

Abstract

The invention relates to a motor bearing fault diagnosis method based on feature optimization and GWOA-XGBoost, and the method comprises the following steps: carrying out the feature decomposition of a vibration signal of a motor bearing through employing an optimized VMD, obtaining IMF components, calculating the MSE value of each IMF component based on a multi-scale entropy theory, and carrying out the feature reconstruction of the vibration signal through the MSE values; a WOA algorithm is improved by using a GA algorithm to obtain a GWOA model, and the XGBoost model is optimized through the GWOA model; and inputting the vibration signal subjected to feature reconstruction into the optimized XGBoost model to obtain a motor bearing fault diagnosis result. According to the motor bearing fault diagnosis method, the motor bearing fault diagnosis precision can be improved.

Description

technical field [0001] The invention relates to the technical field of motor fault diagnosis, in particular to a motor bearing fault diagnosis method based on feature optimization and GWOA-XGBoost. Background technique [0002] Bearings play the role of supporting and guiding the rotation of the motor. According to statistics, about 40% of motor failures are caused by bearing failures. They are important objects for motor health monitoring and directly affect the overall performance of motor operation. Therefore, it is of great significance to carry out research on fault diagnosis of motor bearings. Since the motor bearing is located inside the motor, its fault status is difficult to identify, resulting in limited accuracy of fault diagnosis. [0003] At present, the bearing fault diagnosis based on the vibration signal is the most common method. When the motor bearing fails, the vibration signal will have obvious shock fluctuations. If the time-frequency signal of the vibr...

AI Technical Summary

Problems solved by technology

[0011] Aiming at the technical problems existing in the prior art, the present invention provides a motor bearing fault diagnosis method based on feature optimization and GWOA-XGBoost to solve the problem of improving the accuracy of motor bearing fault diagnosis

Images