Hostname: page-component-f554764f5-sl7kg Total loading time: 0 Render date: 2025-04-16T07:48:38.078Z Has data issue: false hasContentIssue false
  • English
  • Français

Deep learning and similarity-based models for predicting turbofan engine remaining useful life: insights from the CMAPSS dataset

Published online by Cambridge University Press:  14 April 2025

F. Isbilen Open the ORCID record for F. Isbilen [Opens in a new window] ,
O. Bektas Open the ORCID record for O. Bektas [Opens in a new window]  and
M. Konar Open the ORCID record for M. Konar [Opens in a new window]
F. Isbilen
Affiliation:
Graduate School of Natural and Applied Sciences, Erciyes University, Kayseri, Türkiye Aircraft Technology, Rumeli University, Istanbul, Türkiye
O. Bektas
Affiliation:
Faculty of Engineering and Natural Sciences, Istanbul Medeniyet University, Istanbul, Türkiye
M. Konar*
Affiliation:
Faculty of Aeronautics and Astronautics, Erciyes University, Kayseri, Türkiye
*
Corresponding author: M. Konar; Email: [email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

Turbofan engines are having a growing role in modern aircraft maintenance. Due to this increase, estimation of remaining useful life (RUL) of these engines is an important area of study in the field of reliability and maintenance optimisation. In this work, we propose a hybrid approach that combines deep learning models with similarity-based methods for accurate RUL estimation. For a better comparison, we evaluate four architectures: dropout long short-term memory (LSTM), bidirectional LSTM, convolutional neural network 1D (CNN 1D), and multi-layer LSTM. The FD002 subset of NASA’s Commercial Modular Aero-Propulsion System Simulation dataset is used in the case study. Root mean square error (RMSE) and mean absolute error (MAE) were used for performance metrics. The main output of the study suggests that the dropout LSTM model achieves the best prediction accuracy with an RMSE score of 26.547 and a MAE score of 18.749. It is worth noting that these are achieved despite requiring higher computational resources compared to multi-layer LSTM. Furthermore, all models had difficulties with smaller test trajectory lengths such as 50–100 due to training data imbalance. Overall, the findings highlight the promise of hybrid deep learning and similarity-based approaches for RUL prediction. However, potential advancements such as hyperparameter optimisation and data augmentation still hold potential for further improvements.

Keywords

CMAPPSpredictive maintenanceartificial neural networksimilarity base
Type
Research Article
Information
The Aeronautical Journal , First View , pp. 1 - 32
Check for updates
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Royal Aeronautical Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

Baohui, J., Chunhui, X. and Yaohua, L. Study on optimization method of aircraft maintenance plan based on longest path, J Appl Sci, 2013, 13, (16), pp 3354–3357.CrossRefGoogle Scholar
Karaburun, N.N., Hatipoğlu, S.A. and Konar, M. SOC estimation of Li-Po battery using machine learning and deep learning methods, J Aviat, 2024, 8, (1), pp 26–31.CrossRefGoogle Scholar
Maulana, F., Starr, A. and Ompusunggu, A.P. Explainable data-driven method combined with bayesian filtering for remaining useful lifetime prediction of aircraft engines using NASA CMAPSS datasets. Machines, 2023, 11, (2), 163.CrossRefGoogle Scholar
Xiongzi, C., Jinsong, Y., Diyin, T. and Yingxun, W. Remaining useful life prognostic estimation for aircraft subsystems or components: a review, IEEE 2011 10th International Conference on Electronic Measurement & Instruments, Vol. 2, pp 94–98. IEEE, 2011.CrossRefGoogle Scholar
Adam, S.P., Alexandropoulos, S.A.N., Pardalos, P.M. and Vrahatis, M.N. No free lunch theorem: a review, Approximation and Optimization: Algorithms, Complexity and Applications, pp 57–82, 2019.CrossRefGoogle Scholar
Esteghamati, M.Z. and Flint, M.M. Do all roads lead to Rome? A comparison of knowledge-based, data-driven, and physics-based surrogate models for performance-based early design, Eng Struct, 2023, 286, p 116098.CrossRefGoogle Scholar
An, D., Kim, N.H. and Choi, J.H. Practical options for selecting data-driven or physics-based prognostics algorithms with reviews, Reliab Eng Syst Saf, 2015, 133, pp 223–236.CrossRefGoogle Scholar
Lei, Y., Li, N., Guo, L., Li, N., Yan, T. and Lin, J. Machinery health prognostics: A systematic review from data acquisition to RUL prediction. Mech Syst Sig Process, 2018, 104, pp 799–834.CrossRefGoogle Scholar
Byington, C.S., Watson, M. and Edwards, D. Data-driven neural network methodology to remaining life predictions for aircraft actuator components, In 2004 IEEE aerospace conference proceedings (IEEE Cat. No. 04TH8720) (Vol. 6, pp 3581–3589). IEEE, 2004.CrossRefGoogle Scholar
Volponi, A. Data fusion for enhanced aircraft engine prognostics and health management (No. E-15412), 2005.Google Scholar
Watson, M., Byington, C., Edwards, D. and Amin, S. Dynamic modeling and wear-based remaining useful life prediction of high power clutch systems, Tribol Trans, 2005, 48, (2), pp 208–217.CrossRefGoogle Scholar
Goebel, K., Saha, B., Saxena, A., Mct, N. and Riacs, N. A comparison of three data-driven techniques for prognostics, 62nd Meeting of the Society for Machinery Failure Prevention Technology (MFPT), pp 119–131, 2008.Google Scholar
Bolander, N., Qiu, H., Eklund, N., Hindle, E. and Rosenfeld, T. Physics-based remaining useful life prediction for aircraft engine bearing prognosis, Annual Conference of the PHM Society, Vol. 1, No. 1, 2009.Google Scholar
Downey, A., Lui, Y.H., Hu, C., Laflamme, S. and Hu, S. Physics-based prognostics of lithium-ion battery using non-linear least squares with dynamic bounds, Reliab Eng Syst Saf, 2019, 182, pp 1–12.CrossRefGoogle Scholar
Garga, A.K., McClintic, K.T., Campbell, R.L., Yang, C.C., Lebold, M.S., Hay, T.A. and Byington, C.S. Hybrid reasoning for prognostic learning in CBM systems, 2001 IEEE Aerospace Conference Proceedings (Cat. No. 01TH8542) (Vol. 6, pp 2957–2969). IEEE, 2001.CrossRefGoogle Scholar
Swanson, D.C. A general prognostic tracking algorithm for predictive maintenance, 2001 IEEE Aerospace Conference Proceedings (Cat. No. 01th8542), Vol. 6, pp 2971–2977. IEEE, 2001.CrossRefGoogle Scholar
Celaya, J.R., Saxena, A., Saha, S., & Goebel, K.F. Prognostics of power MOSFETs under thermal stress accelerated aging using data-driven and model-based methodologies. In Annual Conference of the PHM Society Vol. 3, No. 1, 2011.CrossRefGoogle Scholar
Isbilen, F., Bektas, O., Avsar, R. and Konar, M. Improved machine learning models with a similarity-based approach for remaining useful life prediction, Aeronaut J, 2024, 129, (1332), pp 1–21.Google Scholar
Wang, T., Yu, J., Siegel, D. and Lee, J. A similarity-based prognostics approach for remaining useful life estimation of engineered systems, 2008 International Conference on Prognostics and Health Management, pp 1–6. IEEE, 2008.CrossRefGoogle Scholar
Hassoun, M.H. Fundamentals of Artificial Neural Networks, The MIT Press, 1995.Google Scholar
Kurucan, M., Özbaltan, M., Yetgin, Z. and Alkaya, A. Applications of artificial neural network based battery management systems: a literature review. Renew Sustain Energy Rev, 2024, 192, p 114262.CrossRefGoogle Scholar
Barzegar Gerdroodbary, M., et al. Optimizing aerodynamic stability in compressible flow around a vibrating cylinder with deep reinforcement learning, Phys Fluids, 2024, 36, p 124143.CrossRefGoogle Scholar
Bektas, O., Marshall, J. and Jones, J.A. Comparison of computational prognostic methods for complex systems under dynamic regimes: a review of perspectives. Arch Comput Methods Eng, 2020, 27, (4), pp 999–1011.CrossRefGoogle Scholar
Bektas, O., Jones, J.A., Sankararaman, S., Roychoudhury, I. and Goebel, K. A neural network filtering approach for similarity-based remaining useful life estimation. Int J Adv Manuf Technol, 2019, 101, pp 87–103.CrossRefGoogle Scholar
Arik, S., Turkmen, I. and Oktay, T. Redesign of morphing UAV for simultaneous improvement of directional stability and maximum lift/drag ratio, Adv Electric Comput Eng, 2018, 18, (4), pp 57–63.CrossRefGoogle Scholar
Konar, M., Arık Hatipoğlu, S. and Akpınar, M. Improvement of UAV thrust using the BSO algorithm-based ANFIS model. Aeronaut J, 2024, 128, (1328), pp 2364–2373.CrossRefGoogle Scholar
Konar, M. and Arık Hatipoğlu, S. Maximisation the autonomous flight performance of unmanned helicopter using BSO algorithm, Aeronaut J, 2024, 128, (1329), pp 2656–2667.CrossRefGoogle Scholar
Karaburun, N.N., Arık Hatipoğlu, S. and Konar, M. Aircraft takeoff speed prediction with machine learning: parameter analysis and model development. Aeronaut J. Published online 2025, pp 1–16. https://doi.org/10.1017/aer.2024.164 CrossRefGoogle Scholar
Ma, M., Sun, C. and Chen, X. Deep coupling autoencoder for fault diagnosis with multimodal sensory data, IEEE Trans Ind Inf, 2018, 14, (3), pp 1137–1145.CrossRefGoogle Scholar
Razavi-Far, R., Farajzadeh-Zanjani, M., Chakrabarti, S. and Saif, M. Data-driven prognostic techniques for estimation of the remaining useful life of lithium-ion batteries, 2016 IEEE International Conference on Prognostics and Health Management (ICPHM), pp 1–8. IEEE, 2016.CrossRefGoogle Scholar
Ma, M. and Mao, Z. Deep-convolution-based LSTM network for remaining useful life prediction, IEEE Trans Ind Inf, 2020, 17, (3), pp 1658–1667.CrossRefGoogle Scholar
Ozkat, E.C., Bektas, O., Nielsen, M.J. and la Cour-Harbo, A. A data-driven predictive maintenance model to estimate RUL in a multi-rotor UAS, Int J Micro Air Veh, 2023, 15, p 17568293221150171.CrossRefGoogle Scholar
Donahue, J., Anne Hendricks, L., Guadarrama, S., Rohrbach, M., Venugopalan, S., Saenko, K. and Darrell, T. Long-term recurrent convolutional networks for visual recognition and description, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 2625–2634, 2015.CrossRefGoogle Scholar
Abdeljaber, O., Avci, O., Kiranyaz, M.S., Boashash, B., Sodano, H. and Inman, D.J. 1-D CNNS for structural damage detection: Verification on a structural health monitoring benchmark data, Neurocomputing, 2018, 275, pp 1308–1317.CrossRefGoogle Scholar
McCulloch, W.S. and Pitts, W. A logical calculus of the ideas immanent in nervous activity, Bull Math Biophysics 1943, 5, pp 115–133.CrossRefGoogle Scholar
NASA. Diagnostics & Prognostics, 2023 [Online accessed 29 March 2025]. https://data.nasa.gov/Aerospace/CMAPSS-Jet-Engine-Simulated-Data/ff5v-kuh6/about_data.Google Scholar
Saxena, A., Celaya, J., Balaban, E., Goebel, K., Saha, B., Saha, S. and Schwabacher, M. Metrics for evaluating performance of prognostic techniques, 2008 International Conference on Prognostics and Health Management, pp 1–17. IEEE, 2008.CrossRefGoogle Scholar
Uckun, S., Goebel, K. and Lucas, P.J. Standardizing research methods for prognostics, 2008 International Conference on Prognostics and Health Management, pp 1–10. IEEE, 2008.Google Scholar
Coble, J.B. Merging data sources to predict remaining useful life–an automated method to identify prognostic parameters, 2010.Google Scholar
Peel, L. Data driven prognostics using a Kalman filter ensemble of neural network models, 2008 International Conference on Prognostics and Health Management, pp 1–6. IEEE, 2008.CrossRefGoogle Scholar
Lei, Y., Li, N., Guo, L., Li, N., Yan, T. and Lin, J. Machinery health prognostics: A systematic review from data acquisition to rul prediction. Mech Syst Sig Process, 2018, 104, pp 799–834.CrossRefGoogle Scholar
Coble, J. and Hines, J.W. Identifying optimal prognostic parameters from data: a genetic algorithms approach, Annual Conference of the PHM Society, Vol. 1, No. 1, 2009.Google Scholar
Kiranyaz, S., Ince, T. and Gabbouj, M. Real-time patient-specific ECG classification by 1-D convolutional neural networks, IEEE Trans Biomed Eng, 2015, 63, (3), pp 664–675.CrossRefGoogle ScholarPubMed
Ince, T., Kiranyaz, S., Eren, L., Askar, M. and Gabbouj, M. Real-time motor fault detection by 1-D convolutional neural networks, IEEE Trans Ind Electron, 2016, 63, (11), pp 7067–7075.CrossRefGoogle Scholar
Jiang, H. Discriminative training of HMMs for automatic speech recognition: A survey, Comput Speech Lang, 2010, 24, (4), pp 589–608.CrossRefGoogle Scholar
Dyke, S.J., Bernal, D., Beck, J. and Ventura, C. Experimental phase II of the structural health monitoring benchmark problem, Proceedings of the 16th ASCE Engineering Mechanics Conference, 2003.Google Scholar
Bektash, O. and la Cour-Harbo, A. First-order layer in artificial pain pathway, Neural Process Lett, 2023, 55, (1), pp 319–343.CrossRefGoogle Scholar
Xue, B., et al. Improved similarity based prognostics method for turbine engine degradation with degradation consistency test, Appl Intell, 2022, 52, pp 10181–10201.CrossRefGoogle Scholar
Peng, C., Hu, X. and Tang, Z. Turbofan engine health assessment based on spatial–temporal similarity calculation, Sensors 2023, 23, 24, p 9748.CrossRefGoogle ScholarPubMed
Baharshahi, M., Seyedhosseini, S.M. and Limon, S.M. Remaining useful life prediction using the similarity-based integrations of multi-sensors data, Qual Eng 2024, 36, (1), pp 36–53.CrossRefGoogle Scholar