Reliability Assessment Method By Integrating Accelerated Degradation and Life Data Based on Nonlinear Wiener Process

Abstract

A performance degradation process is often considered as a stochastic process that is affected by environment, stresses, internal materials, and many other random factors. The Wiener process represents a continuous incremental random process and has the favorable nature of computing and analysis, and many scholars use the linear Wiener processes to describe the performance degradation process. However, not all degradation processes are linear time functions, which means that the nonlinear degradation process needs to be emphasized. At the same time, for some products, accelerated performance degradation data and accelerated life data can be available, and both of them contain reliability information. Therefore, this paper proposes a reliability modeling method for nonlinear Wiener degradation failure process by integrating accelerated performance degradation data and life data. Firstly, the nonlinear process is transformed into a linear process by time scale function, and then two independent modeling processes based on accelerated degradation data and accelerated life data are given respectively. Based on the joint modeling of two independent processes, a one-dimensional nonlinear model is established. Then as a example, the Maximum Likelihood Estimation (MLE) method is used to obtain the point estimate of the model parameters. The simulation data of a kind of electrical connector are used to prepare for the estimation results. The results show that the proposed joint modeling method can accurately describe the nonlinear Wiener process of product performance degradation.