Research on Fault Prediction Method of Rolling Bearing Based on Health Indicator and Dynamic Model

Aiming at the problem of rolling bearing fault prediction in gradual degradation mode, this paper proposes a rolling bearing fault prediction method based on Health Indicator (HI) and dynamic model from the perspective of initial fault detection and prediction model construction. Firstly, multiple His that can represent the degradation state of rolling bearings are mined. Then, based on the amplitude variation trend of multiple His, rolling bearing health status identification is carried out to complete the initial fault detection. Secondly, aiming at the problem of model construction in different degradation stages of rolling bearings, the dynamic model-based strategy is adopted to optimize the parameters of the prediction model based on fault observation samples, so as to complete the prediction and evaluation of Remaining Useful Life (RUL) of rolling bearings. The experimental data verification results of rolling bearings show that the proposed method can achieve relatively ideal fault prediction results.