Remaining useful life prediction method based on two-phase adaptive drift Wiener process

The degradation process of components often shows as two-phase in reality, and the two-phase Wiener process has been widely used to model component degradation. However, previous studies have always assumed that the drift coefficient of each phase is constant, failing to capture the effects of external variations, which reduces the predictive performance of model. Thus, this paper establishes a two-phase adaptive drift Wiener process model to characterize the degradation of components. First, a phasing method is proposed that adaptively identifies the change point and uses fitting metrics to analyze determine if the point is anomalous data. Additionally, the adaptive drift method is innovatively introduced into the developed two-phase Wiener process model for updates. Then, the approximate analytical expression of the probability density function of the remaining useful life is derived and extended to the cases where uncertainty in the state at the change point and heterogeneity are considered. Finally, the feasibility of the proposed method is validated through numerical simulation and actual examples in the laboratory.