The aviation industry places critical emphasis on safety, reliability, and maintenance efficiency. One of the most important aspects of aircraft health management is the prediction of the Remaining Useful Life (RUL) of engines, which enables timely maintenance, minimizes downtime, and prevents unexpected failures. Traditional maintenance schedules based on fixed intervals or threshold limits often fail to account for varying operational conditions, leading to inefficiencies and increased costs. To address these limitations, this study proposes a deep learningâbased predictive model for estimating the RUL of aircraft engines using time-series sensor data.
The proposed system utilizes data from the NASA C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) dataset, which includes multivariate sensor readings from multiple engine units operating under different conditions and fault modes. After preprocessing and normalization, the time-series data are fed into advanced deep learning architectures such as Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNNs) to capture both temporal dependencies and degradation trends. The model is trained to map sensor signal patterns to RUL values, learning complex nonlinear relationships without manual feature engineering.
Experimental results demonstrate that the hybrid CNN-LSTM model achieves superior performance, reducing prediction error compared to traditional regression and shallow machine learning methods. The system effectively forecasts degradation trajectories, offering precise RUL predictions even under variable operational environment
Keywords
Deep learning, prediction, aircraft engine
Conclusion
Deep learning techniques have become widely attractive in engineering use in the last decade, particularly in data analysis for reliability evaluation, which was previously inefficient so it actually needed both expert knowledge of the studied system or the limitations of traditional PHM techniques. There are still many obstacles for reliability related, data-driven applications to maintain improving the estimate of signs of health that can send an accurate diagnostic for systems and facilities. This finding indicates a deep learning way to predict the health of complex systems using large amounts of machine data. The method was taught by the framework using a special topology tile neural network, and it was verified using two separate sets of data. Using the C- MAPSS and Challenge datasets, the proposed framework is validated through the training and testing of several models. The proposed approach is also reliable in predicting how hard both datasets will be usable. The superiority of the proposed topology is illustrated by a comparison of SCG algorithms.
References
[1]Vimala Mathew, Tom Toby, Vikram Singh, B Maheswara Rao, M Goutham Kumar, âPrediction of Remaining Useful Lifetime(RUL) of Turbofan Engine using Machine Learningâ 2017 IEEE International Conference on Circuits and Systems(ICCS 2017). IEEE 2017.
[2]Dong Dong, Xiao-Yang Li, Fu-Qiang Sun âLife Prediction of Jet Engines Based on LSTM-Recurrent Neural Networksâ School of Reliability and Systems Engineering, Science and Technology on Reliability and Environment Engneering. IEEE 2017.
[3]Mei Yuan, Yuting Wu and Li Lin âFault diagnosis and Remaining useful life estimation of aero engine using LSTM neural networkâ 2016 IEEE/CsAA International Conference on Aircraft Utility System(AUS).
[4]Olgun Aydin, Screen Guldamlasioglu âUsing LSTM Networks to Predict Engine Condition on Large Scale Data Processing Frameworkâ 2017 4th International Conference On Electrical and Electronics Engineering.
[5]Zhi Lv, Jian wang ,Guigang Zhang, Huang Jiyang âPrognostic Health Management of Condition-Based Maintenance for Aircraft Engine Systems.
[6]Okoh, C.; Roy, R.; Mehnen, J.; Redding, L. âOverview of Remaining Useful Life Prediction Techniques in Through-Life Engineering Services.â Procedia CIRP 2014, 16, 158â163.
[7]El Mejdoubi, A.; Chaoui, H.; Sabor, J.; Gualous, H. âRemaining Useful Life Prognosis of Supercapacitors under Temperature and Voltage Aging Conditions.â IEEE Trans. Ind. Electron. 2018, 65, 4357â4367.
[8]Dhiman, H.S.; Deb, D.; Carroll, J.; Muresan, V.; Unguresan, M.-L. âWind Turbine Gearbox Condition Monitoring Based on Class of Support Vector Regression Models and Residual Analysis.â Sensors 2020, 20, 6742
[9]Lan, G.; Li, Q.; Cheng, N. âRemaining Useful Life Estimation of Turbofan Engine Using LSTM Neural Networks.â In Proceedings of the IEEE CSAA Guidance, Navigation and Control Conference (CGNCC), Xiamen, China, 10â12 August 2018
[10]Enright, M.P.; McClung, R.C. âA Probabilistic Framework for Gas Turbine Engine Materials ith Multiple Types of Anomalies.â J. Eng. Gas Turbines Power 2011, 133, 082502.
[11]Zhang, B.; Wang, D.; Song, W.; Zhang, S.; Lin, S. âAn Interval-Valued Prediction Method for Remaining Useful Life of Aero Engine.â In Proceedings of the 2020 39th Chinese Control Conference (CCC), Shenyang, China, 27â29 July 2020.
[12]Goebel, K.; Saxena, A. âTurbofan Engine Degradation Simulation Dataset. In NASA Ames Prognostics Data Repository; NASA Ames Research Center: Moffett Field, CA, USA, 2008.â
[13]Chaoui, H.; Kandidayeni, M.; Boulon, L.; Kelouwani, S.; Gualous, H. âReal-Time Parameter Estimation of a Fuel Cell for Remaining Useful Life Assessment. IEEE Trans.â Power Electron. 2021, 36,7470â7479.
[14]Hong, C.W.; Lee, C.; Lee, K.; Ko, M.-S.; Kim, D.E.; Hur, K. âRemaining Useful Life Prognosis for Turbofan Engine Using Explainable Deep Neural Networks with Dimensionality Reduction.â Sensors 2020, 20, 6626.
[15]Yuan, M.; Wu, Y.; Lin, L. Fault diagnosis and remaining useful life estimation of aero engine using LSTM neural network. In Proceedings of the IEEE International Conference on Aircraft Utility Systems (AUS), Beijing, China, 10â12 October 2016.
[16]Saxena, A.; Goebel, K.; Simon, D.; Eklund, N. Damage propagation modeling for aircraft engine run-to-failure simulation. In Proceedings of the 2008 International Conference on Prognostics and Health Management, Denver, CO, USA, 6â9 October 2008.
[17]Huang Z., Xu Z., Ke X., Wang W., Sun
Y. âRemaining useful life prediction for an adaptive skew-Wiener process model. Mechanical Systems and Signal Processing.â 2017;87:294 â306.
Zhang X., Dong Y., Wen L. Remaining useful life estimation based on a new convolutional and recurrent neural network. Proceedings of the IEEE 15th International Conference on Automation Science and Engineering; August 2019; Vancouver, Canada. pp. 317â322.