A system of integrated technical diagnostics, redundancy and forecasting of the technical condition of aircraft on-board equipment based on machine learning methods

Аuthors

Savchuk А. D.^*, Savchenko A. Y., Mironov S. A., Bukirev A. S.^**

Air force academy named after professor N.E. Zhukovskogo and Y. A. Gagarin, 54a Starye Bolshevikov str., Voronezh, 394064, Voronezh Region

*e-mail: atsavchuk@mai.education
**e-mail: bukirev@inbox.ru

Abstract

The necessity of increasing the depth of localization of failures of on-board equipment of aircraft using machine learning methods is substantiated. A system of integrated technical diagnostics, redundancy and forecasting of technical condition based on interaction with a multiplex channel of information exchange has been developed. Key features of the system include: an algorithm for automatically determining learning parameters for each monitored element at the diagnostic stage.; using a recurrent neural network with long-term short-term memory to predict the state; a mechanism for dynamic redundancy of failed elements. 

Keywords:

on-board equipment, machine learning, diagnostics, redundancy, forecasting, technical condition

References

1. Mozgalevsky A.V., Kalyavin V.P., Kostandi G.G. Diagnostics of electronic systems / edited by A.V. Mozgalevsky. – L.: Sudostroenie, 1984. – 224 p.
2. GOST 24029-80. Technical diagnostics. Categories of controllability of diagnostic objects. – M.: Publishing House of Standards, 1980. – 11 p.
3. Golubets, A.A. Introduction of machine learning and artificial intelligence in the aviation industry in order to improve flight safety A.A. Golubets // Internauka. – 2024. – № 41-1(358). – Pp. 34-36.
4. Application of machine learning methods in solving technical diagnostics problems / V.N. Klyachkin, I.N. Karpunina, Yu.E. Kuvaiskova, A.S. Khoreva // Scientific Bulletin of the Institute of Applied Sciences, 2016, No. 8, pp. 158-161.
5. Kushniruk, A.S. Diagnostics of motor-axial bearings of a wheel-motor unit with a neural network reference model: specialty 05.22.07 "Rolling stock of railways, train traction and electrification": dissertation for the degree of candidate of Technical Sciences / Kushniruk Alexey Sergeevich. – Khabarovsk, 2021. – 155 p.
6. Makarov, A.S. Algorithms for monitoring and diagnostics of aviation GTE control systems based on neural network models and fuzzy logic: specialty 05.13.01 "System analysis, management and information processing (by industry)": dissertation for the degree of Candidate of Technical Sciences / Makarov Andrey Sergeevich. Ufa, 2011. – 207 p.
7. Legkonogikh, D. N. According to S. N. The application of neural network technologies in diagnostic systems of aviation power plants / D. N. According to S. N. Legkonogikh / / Proceedings of the Samara Scientific Center of the Russian Academy of Sciences. – 2012. - T. N. 14. – № 4-2. -pp. 639-643.
8. Dobrodeev, I.P. Improving the efficiency of neural network models in systems for diagnosing the technical condition of gas turbine engines based on functional adaptation: specialty 05.13.01 "System analysis, management and information processing (by industry)": dissertation for the degree of Candidate of Technical Sciences / Dobrodeev Ilya Pavlovich. – Rybinsk, 2010. – 218 p.
9. GOST R 52070-2003. The backbone serial interface of the electronic module system. General requirements. Publishing House of Standards. Moscow: Standartinform, 2003. 27 p.
10. Vyugin, V.V. Mathematical foundations of machine learning and forecasting. 2nd Ed., ispr. and add. Moscow: ICNMO, 2018. – 384 p.
11. Certificate of state registration of a computer program No. 2023669933 Russian Federation. The Manchester-2 data decoding program of the aircraft information and control system: No. 2023669137: application 15.09.2023: published 22.09.2023 / A.S. Bukirev.
12. Patent No. 242787 C1 Russian Federation, IPC G06F 11/00, G06N 20/00, G05B 23/02. A device for diagnosing and predicting the technical condition of an aircraft on-board equipment complex based on machine learning methods: No. 2025135065: application 08/12/2025: published 10/04/2026 / A.D. Savchuk
13. Patent No. 2816667 C1 Russian Federation, IPC G06F 11/00, G06N 20/00, G05B 23/02. A method for diagnosing a complex of on-board equipment of aircraft based on machine learning and a device for its implementation: No. 2023116413: application 06/21/2023: published 04/03/2024 / A.S. Bukirev, A.Yu. Savchenko, S.V. Ippolitov [et al.].
14. Certificate of state registration of a computer program No. 2024692313 Russian Federation. A program for assessing the technical condition of an on-board equipment complex based on predicting a controlled diagnostic parameter using a recurrent neural network with long-term short-term memory: No. 2024691930: application 24.12.2024: published 27.12.2024 / A.D. Savchuk.
15. Kuvaiskova, Yu. N.E. Application of ensemble machine learning methods for predicting the technical condition of an object /Yu. n.E. Kuvaiskova / / Proceedings of the Samara Scientific Center of the Russian Academy of Sciences. – 2021. - T. N. 23, No. 1(99). -pp. 111-114.
16. Volkov, V.L. Modeling of processes and systems in instrument engineering: a textbook for students of technical specialties of full-time, evening and correspondence education / V.L. Volkov; Arzamas Polytechnic Institute. Arzamas: Association of Scientists, LLC, 2008, 143 p.
17. Forecasting the quality of functioning of a technical facility using machine learning / M.I. Kornilova, S.V. Busygin, V.N. Kovalnogov, V.N. Klyachkin // Reliability and quality of complex systems. – 2023. – № 4(44). – Pp. 152-158.
18. Bengio, Y., Simard, P., Frasconi, P. Learning Longterm Dependencies with Gradient Descent // IEEE Transactions on Neural Networks. 1994. №2. p. 157 – 166.
19. Wei, W., Li, P. Multi-Channel LSTM with Different Time Scales for Foreign Exchange Rate Prediction. Proceedings of the international conference on Advanced Information Science and System. 2019.
20. Isaev, G.G. Intelligent information technologies in systems of technical diagnostics of aviation equipment / G.G. Isaev, N.P. Kozak // Information technologies in education: Collection of scientific papers. Ulyanovsk: Publisher Kachalin Alexander Vasilyevich, 2018. pp. 57-61.
21. Certificate of state registration of a computer program № 2024664709. Russian Federation. The program of technical diagnostics of information-transforming elements of onboard equipment of the aviation complex based on machine learning without a teacher with automatic determination of training parameters of models "DIAGNOSTICS-93": № 2024662926: application 06.06.2024: published 24.06.2024 / A.S. Bukirev.
22. Fedorinov A. Y., Ivanov Yu.P. A method for improving the accuracy and reliability of navigation and measurement systems based on complex optimally invariant filtering of arbitrary signals in conditions of redundancy of information processing devices / A.Y. Fedorinov, Yu.P. Ivanov // Proceedings of MAI. - 2024. - No. 139.
23. Bukirev A.S. A method for diagnosing a complex of on-board equipment of aircraft based on machine learning / A.S. Bukirev // Proceedings of MAI. – 2023. – No. 133.

Download