A Bearing Life Prediction Method Based on Enhanced Particle Filter

Abstract

The invention discloses a bearing life prediction method based on enhanced particle filter. For the problems existing in a traditional particle filter algorithm that particles are degenerated and the particle diversity is lost, a self-adaption importance density function selection algorithm and a particle smoothing algorithm based on a neural network are respectively proposed, a traditional importance density function determination method and a resampling algorithm are improved, on this basis, an enhanced particle filter algorithm is proposed, and the bearing remaining service life prediction method based on the enhanced particle filter is designed. The method can effectively restrain the particle degeneration phenomenon and the particle depletion phenomenon and can accurately predict the bearing remaining service life in advance after the multi-step iteration.

Description

technical field [0001] The invention relates to a bearing life prediction method, in particular to a bearing life prediction method based on enhanced particle filter. Background technique [0002] As we all know, the failure of rotating machinery caused by bearing failure is not uncommon (for example, in the study of motor reliability co-sponsored by General Electric and IEEE Industrial Application Society, it was found that the problems caused by bearings accounted for more than 40% of the total machine failure). Therefore, researches on fault diagnosis and life prediction of bearings have received widespread attention from industry and academia. [0003] Traditional bearing fault diagnosis is to identify and classify faulty bearing signals, while bearing life prediction is to continuously track and monitor mechanical components before a fault occurs, and build a model based on the extracted tracking features to estimate the remaining life of the bearing. Make predictions ...

AI Technical Summary

Problems solved by technology

However, the traditional particle filter algorithm has problems such as particle degradation and loss of particle diversity, which seriously affect the accuracy and reliability of bearing life prediction.

There is currently no public solution to this problem

Images