Health stages division and remaining useful life prediction of rolling element bearings based on hidden semi-Markov model

Health stages division and Remaining Useful Life (RUL) prediction are two important parts in safety study of rolling element bearings. In this paper, the Hidden Semi-Markov Model (HSMM) is proposed to divide the degradation stages of rolling element bearings. Firstly, we extract the root mean square feature from the original vibration signal, then utilize Viterbi algorithm to divide the degradation stages. Secondly, Fault occurrence time is determined according to the degradation stage and RUL is predicted with HSMM. In order to verify the effectiveness of this method, IEE-PHM-2012 challenge data sets are adopted and the comparison with the existing methods is carried out.