Methodology for diagnosing the technical condition of aviation gas turbine engines using recurrent neural networks (RNN) and long short-term memory networks (LSTM)

This study presents a method for diagnosing the technical condition of aviation gas turbine engines (GTE) using recurrent neural networks (RNN) and long short-term memory networks (LSTM). The primary focus is on comparing the effectiveness of these models for forecasting key operating parameters of GTEs, such as vibrations, turbine-inlet temperatures, and rotor speeds of low and high pressure. The research involved thorough data cleaning and normalization, including handling missing values, normalization using Min-Max Scaling, outlier removal, data decorrelation, and time series smoothing. The RNN and LSTM models were trained using the backpropagation through time (BPTT) algorithm to accurately forecast GTE operating parameters. The results show that both models demonstrate high forecasting accuracy, but the RNN models perform better in most parameters. For vibration parameters (VIB_N1FNT1, VIB_N1FNT2, VIB_N2FNT1, and VIB_N2FNT2), RNN models achieved lower RMSE and MAE values, confirming their higher accuracy. For temperature parameters (EGT1 and EGT2), RNN models also showed higher accuracy rates. Meanwhile, LSTM models achieved better results for some rotor speed parameters (N21 and N22). The findings emphasize the necessity of choosing the appropriate model based on the nature of data and the specifics of the parameters to be forecast. Future research may focus on developing hybrid approaches that combine the advantages of both models to achieve optimal results in diagnosing the technical condition of GTEs.

1. Fentaye, A.D., Zaccaria, V., Kyprianidis, K. (2021). Aircraft engine performance monitoring and diagnostics based on deep convolutional neural networks. Machines, vol. 9, issue 12, ID: 337. DOI: 10.3390/machines9120337 (accessed: 27.02.2024).

2. Al-Tekreeti, W.K.F., Kashyzadeh, K.R., Ghorbani, S. (2024). Advancements in gas turbine fault detection: a machine learning approach based on the temporal convolutional network-autoencoder model. Applied Sciences, vol. 14, issue 11, ID: 4551. DOI: 10.3390/app14114551 (accessed: 27.02.2024).

3. Berghout, T., Mouss, M.-D., Mouss, L.-H., Benbouzid, M. (2023). ProgNet: A transferable deep network for aircraft engine damage propagation prognosis under real flight conditions. Aerospace, vol. 10, issue 1, ID: 10. DOI: 10.3390/aerospace10010010 (accessed: 27.02.2024).

4. Hochreiter, S., Schmidhuber, J. (1997). Long short-term memory. Neural computation, vol. 9, issue 8, pp. 1735–1780. DOI: 10.1162/neco.1997.9.8.1735

5. Zhao, J., Li, Y.-G., Sampath, S. (2023). Convolutional neural network denoising autoencoders for intelligent aircraft engine gas path health signal noise filtering. Journal of Engineering for Gas Turbines and Power, vol. 145, issue 6, ID: 061013. DOI: 10.1115/1.4056128 (accessed: 27.02.2024).

6. Garg, S., Simon, D. (2012). Challenges in aircraft engine gas path health management. In: Proceedings of the Tutorial on Aircraft Engine Control and Gas Path Health Management, Cleveland, OH, USA, 64 p. Available at: https:// ntrs.nasa.gov/api/citations/20150009565/downloads/20150009565.pdf (accessed: 15.02.2024).

7. Mohammadi, R., Naderi, E., Khorasani, K., Hashtrudi-Zad, S. (2011). Fault diagnosis of gas turbine engines by using dynamic neural networks. 2011 IEEE International Conference on Quality and Reliability. Bangkok, Thailand, pp. 25–30. DOI: 10.1109/ICQR.2011.6031675

8. Goodfellow, I., Bengio, Y. Courville, A. (2016). Deep learning. The MIT Press, 800 p.

9. Clifton, D. (2006). Condition monitoring of gas-turbine engines. Transfer Report, Department of Engineering Science, University of Oxford, 60 p. Available at: https://www.robots.ox.ac.uk/~davidc/pubs/transfer.pdf (accessed: 27.02.2024).

10. Upadhyay, A., Li, J., King, S., Addepalli, S. (2023). A deep-learning-based approach for aircraft engine defect detection. Machines, vol. 11, issue 2, ID: 192. DOI: 10.3390/machines11020192 (accessed: 27.02.2024).

11. Zhou, D., Yao, Q., Wu, H., Ma, S., Zhang, H. (2020). Fault diagnosis of gas turbine based on partly interpretable convolutional neural networks. Energy, vol. 200, ID: 117467. DOI: 10.1016/j.energy.2020.117467 (accessed: 27.02.2024).

12. Falsetti, C., Sisti, M., Beard, P.F. (2021). Infrared thermography and calibration techniques for gas turbine applications: A review. Infrared Physics & Technology, vol. 113, ID: 103574. DOI: 10.1016/j.infrared.2020.103574 (accessed: 27.02.2024).

13. Zhao, F., Chen, L., Xia, T., Ye, Z., Zheng, Y. (2019). Gas turbine exhaust system health management based on recurrent neural networks. Procedia CIRP, vol. 83, no. 12, pp. 630–635. DOI: 10.1016/j.procir.2019.04.122

14. Pitkänen, J., Hakkarainen, T., Jeskanen, H., Kuusinen, P., Lahdenperä, K., Särkiniemi, P. (2000). NDT methods for revealing anomalies and defects in gas turbine blades. In: 15th World Conference on Nondestructive Testing, Italy, Roma, 15–21 October. Available at: https://www.ndt.net/article/wcndt00/papers/idn629/idn629.htm (accessed: 27.02.2024).

15. Loboda, I. (2016). Neural networks for gas turbine diagnosis. In book: Artificial Neural Networks-Models and Applications. DOI: 10.5772/63107 (accessed: 27.02.2024).

16. Pineda, F.J. (1987). Generalization of back-propagation to recurrent neural networks. Physical Review Letters, vol. 59, issue 19, ID: 2229. DOI: 10.1103/PhysRevLett.59.2229 (accessed: 27.02.2024).

17. Pankov, E.A., Chayka, N.F. (2016). The possibilities of spectral methodes for aircraft engines diagnosis. Interexpo Geo-Siberia, no. 9, pp. 8–13. (in Russian)