Remaining Useful Life (RUL) Prediction Based on the Bivariant Two-Phase Nonlinear Wiener Degradation Process

Recent advancements in science and technology have resulted in products with enhanced reliability and extended lifespans across the aerospace and related sectors. Traditional statistical models struggle to assess their reliability accurately, prompting increased interest in predicting product lifespans during service. These products, characterized by intricate structures and diverse functionalities, exhibit complex, multistage, multiperformance, and nonlinear degradation processes. To address these challenges, this paper proposes a framework for multiperformance, multi-phase Wiener process modeling and reliability analysis. It introduces a two-phase nonlinear Wiener degradation model and identifies change points via the Schwarz information criterion (SIC). The analytical formula for remaining useful life (RUL) is obtained from the concept of the first hitting time (FHT), which considers the stochastic nature of the degradation amount at the change point. The Akaike information criterion (AIC) is then utilized, and an appropriate copula function is chosen to analyze the correlation between two performance indices, given an established complexity with parameters in the degradation model. A two-step method for estimating these uncertain parameters is presented in this paper. Validation through a turbine engine case study underscores its potential to advance reliability theory and engineering practices.