Adaptive Prediction of Remaining Useful Lifetime for Single Equipment Based on Accelerated Degradation Modeling

In view of the few research on remaining useful lifetime (RUL) prediction based on accelerated degradation data of single equipment, an adaptive prediction method of RUL for single equipment based on accelerated degradation modeling is proposed in this paper. Firstly, based on the nonlinear Wiener process, the accelerated degradation model for the equipment is constructed. Next, the degradation state of the equipment is updated by using the Kalman filter (KF) algorithm. And then, the expectation-maximization-Kalman filter (EM-KF) algorithm is used to adaptively estimate the parameters of the accelerated degradation model. Finally, based on the full probability formula, the probability density function (PDF) of RUL is derived. Through the analysis of the numerical example, the effectiveness of the proposed method is verified.