A stacked deep convolutional neural network to predict the remaining useful life of a turbofan engine

  • 2021-11-24 18:36:28
  • David Solis-Martin, Juan Galan-Paez, Joaquin Borrego-Diaz
  • 1

Abstract

This paper presents the data-driven techniques and methodologies used topredict the remaining useful life (RUL) of a fleet of aircraft engines that cansuffer failures of diverse nature. The solution presented is based on two DeepConvolutional Neural Networks (DCNN) stacked in two levels. The first DCNN isused to extract a low-dimensional feature vector using the normalized raw dataas input. The second DCNN ingests a list of vectors taken from the former DCNNand estimates the RUL. Model selection was carried out by means of Bayesianoptimization using a repeated random subsampling validation approach. Theproposed methodology was ranked in the third place of the 2021 PHM ConferenceData Challenge.

 

Quick Read (beta)

loading the full paper ...