Remaining Useful Life Prediction of Nonlinear Wiener Process-Based Degradation Model with Fusing Failure Data Time

Abstract

Aiming at the imperfection of historical degradation data for equipment, a remaining useful life (RUL) prediction method with fusing failure time data is proposed. Firstly, the nonlinear Wiener process-based degradation mode is used to model the degradation process of equipment. Then, based on the failure time data of congeneric equipment, the expectation maximization (EM) algorithm is used to estimate the unknown parameters in the model, in which the fixed parameters are calculated based on the field degradation data of the evaluated equipment. Finally, the degradation data of lithium-ion batteries are used to verify the proposed RUL prediction method. The experimental results show that for degradation data of equipment with imperfect prior information, the RUL prediction method with fusing failure time data is better than the traditional RUL prediction method.