Bearing cross-working-condition fault prediction method based on adversarial transfer learning

Abstract

The invention discloses a bearing cross-working-condition fault prediction method based on adversarial transfer learning, and the method comprises the steps: setting different working conditions, carrying out the signal collection of a bearing, and obtaining a vibration signal; recognizing the vibration signal based on a continuous abnormal point detection method of peak measurement to obtain a health stage signal and a degradation stage signal; performing preprocessing and fault feature extraction on the degradation stage signal to obtain a fault feature set; inputting the fault feature set into a domain adversarial transfer learning double-branch neural network and carrying out updating training to obtain a prediction model; and obtaining to-be-tested data and inputting the to-be-tested data into the prediction model to obtain a fault prediction result. By using the method, accurate service life prediction of the high-speed shaft ball bearing of the gear box of the large fan under different working conditions is realized, wherein the service life of the high-speed shaft ball bearing faces to actual working condition characteristics. The bearing cross-working-condition fault prediction method based on adversarial transfer learning can be widely applied to the field of engineering part service life prediction.

Description

technical field [0001] The invention relates to the field of life prediction of engineering components, in particular to a bearing fault prediction method based on adversarial transfer learning across working conditions. Background technique [0002] The development of wind power has given birth to the vigorous development of the wind power industry. Gearbox is an important mechanical part in the transmission chain of large wind turbines, which has the characteristics of complex structure, low speed, heavy load and long service period. Wheels and ring gears, etc.) frequently fail under the action of complex and variable loads and super instantaneous impact. Gearboxes have always been a critical component and weak link in large wind turbines due to the severity of the consequences of failure. Gearbox failures have typical chain characteristics, that is, minor failures can easily expand into catastrophic failures in a short period of time, which generally directly lead to th...

AI Technical Summary

Problems solved by technology

Gearbox faults have typical chain characteristics, that is, minor faults can easily expand into catastrophic faults in a short period of time, which generally directly lead to the shutdown of large wind turbines. Secondly, the difficulty of repairing gearbox faults will result in longer downtime and huge economic losses. losses, thereby significantly reducing the availability and economics of large wind turbines throughout their life cycle

[0003] Due to its complex structure, harsh working environment, and fine assembly process, large-scale wind power gearboxes are different from ordinary rotating machinery systems, making their remaining life prediction methods for engineering practice subject to various challenges. Challenges: It is difficult to effectively extract fault features, the strong randomness of the degradation process makes it difficult to predict the remaining life robustly, and many implementation steps are limited by expert experience, which makes the intelligence level of life prediction research not high

Images