Rolling bearing health prognosis using a modified health index based hierarchical gated recurrent unit network

Rolling bearing health prognosis is helpful to improve the operation efficiency and security of rotating machinery. In this paper, a modified health index based hierarchical gated recurrent unit network is proposed for rolling bearing health prognosis. Firstly, in order to effectively depict the degradation process, a modified health index is designed based on kernel principle component analysis (KPCA) and exponentially weighted moving average (EWMA). Then, in order to capture the high nonlinear characteristics and assess the health condition, a hierarchical gated recurrent unit network is constructed by stacking multiple hidden layers. Finally, the proposed method is applied for rolling bearing health prognosis with the experimental bearing data, and the results confirm that it outperforms other existing methods.