A degradation prognostic framework for gas turbine engines

Gas turbine prognostics is a promising technology that is used in aircraft maintenance because of its ability to forecast remaining useful life and likely future circumstances, which leads to the prospects of reliable operation of a system. As the performance of all systems degrades over time, it is essential to forecast the functionality and health condition of critical systems. Model-based prognostic methods using reliable damage modelling methods can accurately forecast the remaining useful life of a system by tracking the trends of a growing deterioration. In this paper, a model based prognostic framework using Particle Filtering method that includes an exponential damage propagation model is applied to a gas turbine system in order to predict the remaining useful life of the system. Bayesian inference is applied in Particle Filtering method to use degradation measurements for estimation and update process of model parameters.