Open AccessReal-time Online Prediction of Data Driven Bearing Residual Life
Author(s) -
Wang Hai-ya,
Yu Zhou,
Lanping Guo
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1437/1/012025
Subject(s) - residual , downtime , bearing (navigation) , computer science , data mining , artificial neural network , principal component analysis , predictive modelling , artificial intelligence , machine learning , reliability engineering , engineering , algorithm
In order to realize the predictive maintenance of key components under massive vibration data, real-time online prediction of the remaining life of different types of bearings, a data driven real-time online prediction method for bearing residual life is introduced. The method realizes the construction of the data-driven bearing residual life prediction model by selecting the bearing vibration data Spearman characteristic parameter selection, principal component analysis (PCA), health index fusion, and BP neural network fitting. The built model is continuously updated by real-time online acquisition of data to achieve real-time online prediction of bearing residual life. The accuracy and feasibility of the method for predicting the remaining life of different types of bearings are verified by experiments. Using this method to predict the remaining life of the bearing helps to achieve predictive maintenance of critical components, reduce unplanned downtime, increase production efficiency, and reduce production costs.