터보팬 엔진의 잔여 유효 수명 예측을 위한 2D CNN 기반 모델

Abstract

In this paper, we propose a method for predicting Remaining Useful Life (RUL) using multivariate time series data. To enhance the relationships between the input and output data, we propose a pre-processing method and construct a model using 2D convolutional layers, bidirectional long short-term memory, and long short-term memory networks. The results demonstrate that the proposed approach provides higher accuracy for the RUL estimation compared to previous methods.