Deep Learning-Based Non-Linear Prediction of Remaining Useful Life of Aircraft Engines 


Vol. 14,  No. 12, pp. 1097-1104, Dec.  2025
10.3745/TKIPS.2025.14.12.1097


PDF
  Abstract

The prediction of the remaining useful life (RUL) of aircraft turbofan engines is a critical task in prognostics and health management (PHM), as it enables the early detection of component degradation, the optimization of maintenance schedules, and the prevention of safety incidents. Recent deep learning–based RUL prediction studies have made significant progress. However, most efforts have focused on improving model architectures, while relatively little attention has been paid to the design of RUL labeling functions. This study proposes a novel approach that preserves the existing pre-training structure used in recent research while replacing the target RUL labeling function with a non-linear concave function that more accurately reflects actual degradation patterns. Using the NASA C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) dataset, we evaluate a CAE (convolutional autoencoder)–RNN-based prediction model under various parameter settings and demonstrate that the proposed non-linear labeling model improves prediction accuracy in terms of RMSE (root mean squared error) and simultaneously enhances the NASA S-score safety metric compared to the conventional piecewise-linear model.

  Statistics
  Cite this article

[IEEE Style]

M. Kim and K. Lee, "Deep Learning-Based Non-Linear Prediction of Remaining Useful Life of Aircraft Engines," The Transactions of the Korea Information Processing Society, vol. 14, no. 12, pp. 1097-1104, 2025. DOI: 10.3745/TKIPS.2025.14.12.1097.

[ACM Style]

Min-Jung Kim and Kang-Won Lee. 2025. Deep Learning-Based Non-Linear Prediction of Remaining Useful Life of Aircraft Engines. The Transactions of the Korea Information Processing Society, 14, 12, (2025), 1097-1104. DOI: 10.3745/TKIPS.2025.14.12.1097.

Forms

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

Old Journals

Indexing

All publications of TKIPS is indexed in DOI, EBSCO, Google Scholar, Crossref, and CrossCheck.

TKIPS is also selected as the Journal for Accreditation by NRF (National Research Foundation of Korea).

Related Journals