Rolling Bearing Health State Assessment Based on K-Means and Ensemble HMM

Abstract

The division and identification of rolling bearing health states are the basis for Condition-based Maintenance, which effectively guarantee the safe and stable operation of the equipment. In order to accurately divide the normal and failure states, analyze failure occurrence time and identify the current state, the K-Means clustering method is used to cluster the data of the full life cycle, and the ensemble Hidden Markov Model (HMM) method for pattern recognition of online data. The experimental bearing life cycle data set provided by the Institute of Design Science and Basic Component at Xi’an Jiaotong University (XJTU) and the Changxing Sumyoung Technology Co. Ltd. (SY) is selected to verify the effectiveness of the proposed method. The results show that the data consist of different states can get a good clustering effect and each state data can also be accurately identified.