A Light Gradient Boosting Machine for Remainning Useful Life Estimation of Aircraft Engines

Abstract

The turbofan engine is a crucial component of the aircraft. In order to provide an appropriate maintenance for the turbofan engine to improve the reliability of the system, it is necessary to estimate the remaining useful life (RUL) of the engine. In this paper, a data-based RUL prediction method is proposed using the Light Gradient Boosting Machine (LightGBM). To capture more degradation information, the time window of row data and runtime of the turbofan engine are used as inputs to the proposed method after normalization. LightGBM works very well with these high-dimensional inputs and the model is easy to interpret. Moreover, LightGBM is insensitive to noise and can deal with redundant information in the time-window-size data, which facilitate the application of the proposed method to turbofan engines. In order to demonstrate the effectiveness of the proposed method, experiments are performed on C-MAPSS datasets provided by NASA, and the high prognostic accuracy is obtained. By comparing with the existing methods using the same dataset, the promising prospects of the proposed method in industry applications are illustrated.