Future Vibration Estimation Using LSTM for Condition-Based Maintenance of Aircraft Systems

Hüseyin Şahin, Ömer Faruk Göktaş

2025

Abstract

Abstract: This study presents a deep learning-based approach for enhancing Condition-Based Maintenance (CBM) strategies in aircraft systems by utilizing Long Short-Term Memory (LSTM) networks to forecast future vibration trends. Using high-resolution time-series data from the NASA IMS Bearing Dataset, the proposed LSTM model successfully captures complex temporal dependencies that characterize degradation behaviour in aircraft components. Experimental results demonstrate that the model achieves high prediction accuracy with a low Mean Absolute Error (MAE) of 0.0010, enabling timely detection of incipient faults and minimizing unnecessary maintenance interventions. Compared to traditional models, LSTM networks offer high performance in learning nonlinear patterns and maintaining predictive reliability under varying operational conditions. The integration of LSTM-based forecasting into CBM frameworks supports proactive maintenance planning, reduces lifecycle costs, and increases aircraft safety. This study contributes to the literature by validating the practical implementation of LSTM in real-world aerospace maintenance workflows, offering a scalable and intelligent solution for predictive maintenance in both civil and military aviation contexts.

Paper Citation

in Harvard Style

Şahin H. and Göktaş Ö. (2025). Future Vibration Estimation Using LSTM for Condition-Based Maintenance of Aircraft Systems. In Proceedings of the 2nd International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences - Volume 1: ICEEECS; ISBN 978-989-758-783-2, SciTePress, pages 169-175. DOI: 10.5220/0014299800004848

in Bibtex Style

@conference{iceeecs25,
author={Hüseyin Şahin and Ömer Faruk Göktaş},
title={Future Vibration Estimation Using LSTM for Condition-Based Maintenance of Aircraft Systems},
booktitle={Proceedings of the 2nd International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences - Volume 1: ICEEECS},
year={2025},
pages={169-175},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0014299800004848},
isbn={978-989-758-783-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences - Volume 1: ICEEECS
TI - Future Vibration Estimation Using LSTM for Condition-Based Maintenance of Aircraft Systems
SN - 978-989-758-783-2
AU - Şahin H.
AU - Göktaş Ö.
PY - 2025
SP - 169
EP - 175
DO - 10.5220/0014299800004848
PB - SciTePress