NARNET-based Prognostics Modeling for Deteriorating Systems under Dynamic Operating Conditions

Abstract

This paper presents a new prognostics modeling method based on a nonlinear autoregressive neural network (NARNET) for computing the remaining useful life (RUL) of a deteriorating system under dynamic operating conditions. Our approach consists of two processes: (1) an offline training process is built to model the degradation law and failure zones based on the dataset of hundreds of identical units with run-to-failure time-series sensor measurements; (2) an online prediction and testing process is developed to predict the RUL of a test unit. We particularly investigate how the degradation process is affected by the unit-specific operating conditions. The operating conditions are forecasted by a NARNET model based on the unit's historical operating conditions. We show that the prognostics model integrating the operating condition forecast provides more accurate and efficient RUL prediction. The aircraft turbine engine degradation dataset is utilized to demonstrate the model and test the model performance.