Remaining useful life prediction for fractional degradation processes under varying modes

Abstract

Degradation processes of practical chemical engineering systems are difficult to model accurately because of complicated nonlinearities, non-stationarities, and non-Markovian properties. Traditional prognostics techniques tend to neglect the multi-mode switching issue, and therefore cannot be used for predicting the remaining useful life (RUL) of a piece of long-life equipment, especially when it is continuously operated under varying modes. Specifically, the stationarity of differential data also plays an important role in depicting the evolutionary trend, and further affects the extrapolation procedure for RUL prediction. In this paper, we construct a multi-mode degradation model with regard to fractional diffusion processes by considering both stationary and non-stationary increments. The drift term is represented as a time-related nonlinear function, while the diffusion term is driven by fractional Brownian motion (FBM) or sub-fractional Brownian motion (sub-FBM), according to the results of the stationary test. Based on the monitored data, the model is identified by combining change-point detection and parameter estimation. The closed-form distribution of RUL is then derived through a weak convergence transformation. A numerical example and a case study of a large blast furnace are provided to evaluate the performance of the proposed prognostics framework. K E Y W O R D S degradation, multi-mode, non-Markovian, remaining useful life, stationary test

Prognostics of fractional degradation processes with state-dependent delay

“…The thermocouple temperatures of the hearth wall are selected as the degradation variables, which can be used for indirectly reflecting the erosion degree of the wall, as its thickness is difficult to be measured under harsh high temperature environments. Also see Xi et al 29 for some introductions and preliminary studies of the concerned temperature degradation processes. Because such data attach to slow time-varying processes, the sampling interval is kept consistent with the operator schedule (8 h), and we can denote t = 1=3 day for clarity.…”

Fault Detection, Supervision, and Safety for Chemical Processes: 2020

“…The prediction of remaining useful life (RUL) was investigated by the authors of both Xi et al and Xie et al Xi et al addressed the multi‐mode switching issue in RUL prediction and tested their proposed method on a large blast furnace. Xie et al combined a particle filter and an extended unbiased finite‐impulse‐response filter to predict the RUL of a lithium‐ion battery.…”