@article{ref1,
title="A new approach for safety life prediction of industrial rolling bearing based on state recognition and similarity analysis",
journal="Safety science",
year="2020",
author="Wang, Heng and Chen, Jinhai and Qu, Jiangming and Ni, Guangxian",
volume="122",
number="",
pages="e104530-e104530",
abstract="Rolling bearing is an important part of rotating machinery which operation safety and reliability are directly related to the normal operation of equipment. Remaining useful life is an important index for describing the safe operation of industrial rolling bearing and life prediction technology of machinery equipment is an important method to realize intelligent maintenance and industrial safety operation. According to the idea of data-driven, a new method for predicting the life of rolling bearing was proposed. By using life-cycle data of bearing from normal to failure, based on the new clustering algorithm of K-means and threshold correction, the recognition and partition of different operation states of rolling bearings were realized and the life model was established by defining the state matrix. For the monitoring bearing, comprehensive similarity analysis with the historical data was carried out to construct the life proportional adjustment function and dynamically modify the parameters of the state matrix model, so as to realize the adaptive prediction of the life of the monitoring bearing. The bearing test data of the University of Cincinnati Laboratory Center were used to carry out the applied research. The normal operation state and remaining life of two bearings were predicted by a set of bearing life data. The results showed that the method had better prediction accuracy and generalization compared with hidden Markov model and grey model. This study provides some theoretical guidance and basis for the industrial safe operation and maintenance of rolling bearings during service.<p /> <p>Language: en</p>",
language="en",
issn="0925-7535",
doi="10.1016/j.ssci.2019.104530",
url="http://dx.doi.org/10.1016/j.ssci.2019.104530"
}