Rolling bearing health index curve construction method based on AFF-AAKR fusion

Abstract

The invention discloses a rolling bearing health index curve construction method based on AFF-AAKR fusion, and the method comprises the steps of firstly carrying out the feature extraction of a horizontal vibration acceleration sensor signal of a rolling bearing from a time domain, a frequency domain and a time-frequency domain, obtaining a candidate feature set, and screening out optimal features; then, obtaining an observation space vector from the data sequence with the preferable features by adopting an AFF method, extracting a health space vector from the observation space vector, mapping the observation space vector to a health space represented by the health space vector to obtain a mapping vector, calculating a residual error of data at each moment in the observation space vector and the mapping vector to obtain a data sequence, and fitting the residual data to obtain a curve which is a health index curve. According to the method, time-frequency domain feature extraction is carried out on the collected vibration acceleration signal of the rolling bearing, construction of the health index curve is realized through feature fusion based on AFF-AAKR, and the performance of the health index curve is improved.

Description

technical field [0001] The invention belongs to the technical field of rolling bearing life prediction, and more specifically relates to a rolling bearing health index curve construction method based on AFF-AAKR fusion. Background technique [0002] Rolling bearings are the basic parts in rotating machinery and play a pivotal role in rotating machinery. Condition monitoring and remaining service life prediction of rolling bearings are particularly important in industrial production. The performance of the health index curve is a key factor that directly affects the accuracy of remaining life prediction. On the one hand, the purpose of constructing the health index curve is to track the trend of equipment status over time. On the other hand, the health index curve can be used to easily predict the remaining life. The performance of the health index curve is crucial to the prediction of the remaining life. influences. [0003] According to whether the health index curve is ...

AI Technical Summary

Problems solved by technology

At the same time, in today's era of big data, in the practical application of engineering, the existing health index curve construction technology has problems such as inaccurate description of degradation trend, low generalization ability, and inability to achieve high-precision life prediction, etc., which need to be studied and solved

Images