Method for diagnosing an aircraft on-board equipment complex based on machine learning

Аuthors

Bukirev A. S.

Air force academy named after professor N.E. Zhukovskii and Y.A. Gagarin, Voronezh, Russia

e-mail: bukirev@inbox.ru

Abstract

The article presents an analysis of existing methods and control means employed onboard a modern aircraft. The author substantiates the necessity and possibility of increasing the depth of search for the failure location by the machine learning methods application, allowing automatically create and employ diagnostic models being difficult to formalize. A modified algorithm for the onboard equipment information-transforming elements diagnosing of was developed. The algorithm is based on machine learning through interaction with a multiplex information exchange channel, with a modification of the algorithm in terms of using a block for automatically assigning optimal training parameters, according to the criterion of ensuring its full autonomy (training without a teacher), due to preliminary analysis of the training sample for each information-transforming element. The article considers the problem of the external disturbing impacts effect on the result of the information-converting elements of onboard equipment diagnosing. To compensate for these impacts, a modified Kalman filter with automatic determining of optimal filtering parameters is applied for each information-transforming element, due to the training sample preliminary analysis. The developed algorithm combines the integration (ensembling) of the three machine-learning models, with the majority principle of generating at the output the control result of each information-transforming element using the “two out of three” method, to increase the control results reliability, as well as minimize the likelihood of first and second errors of the second kind when diagnosing. In this work, by information-converting elements the onboard equipment performing its functions through the multiplex information exchange channel is meant. The aircraft recovery time is expected herewith to be reduced by minimizing the time for the failure location searching, which will allow increasing the main complex indicator of the aircraft reliability, namely the availability factor.

Keywords:

Method for diagnosing an aircraft on-board equipment complex based on machine learning

References

1. Uatt Dzh., Borkhani R., Katsaggelos A. Mashinnoe obuchenie: osnovy, algoritmy i praktika primeneniya (Machine learning: fundamentals, algorithms and application practice), Saint Petersburg, BKhV-Peterburg, 2022, 640 p.

2. Interfeis magistral'nyi posledovatel'nyi sistemy elektronnykh modulei. Testirovanie seriinykh obraztsov interfeisnykh modulei, funktsioniruyushchikh v rezhime okonechnogo ustroistva. GOST R 52075-2003. (Main serial interface of the electronic module system. Testing of serial samples of interface modules operating in terminal device mode. State Standard 52075-2003), Moscow, Standartinform, 2003, 25 p.

3. Kol'tsov Yu.V., Dobychina E.M. Uspekhi sovremennoi radioelektroniki, 2019, no. 8, pp. 29-45. DOI: 10.18127/j20700784-201908-03

4. Zelentsov A.V., Naletov E.V., Chemodanov M.A. Voennyi nauchno-prakticheskii vestnik, 2023, no. 1 (17), pp. 46-52.

5. Kol'tsov, Yu.V. Dobychina E.M. Uspekhi sovremennoi radioelektroniki, 2020, no. 3, pp. 47-54. DOI: 10.18127/j20700784-202003-06

6. Miroshnikov I.I., Konoval'tsev E.V., Khlon' E.P. Elektronnoe uchebnoe posobie «Aviatsionnoe oborudovanie samoleta Su-30SM» // Mezhvuzovskii sbornik nauchnykh trudov Tom 24.: (Electronic textbook «Aviation equipment of the Su-30SM aircraft». Vol. 24), Krasnodar, Krasnodarskoe vysshee voennoe aviatsionnoe uchilishche letchikov imeni Geroya Sovetskogo Soyuza A.K. Serova, 2020, pp. 143-148.

7. Interfeis magistral’nyi posledovatel’nyi sistemy elektronnykh moduleii, GOST R 52070-2003 (Main serial interface of the system of electronic modules, State Standard R 52070-2003), Moscow, Standarty, 2003.

8. Bukirev A.S., Savchenko A.Yu., Yatsechko M.I. Vestnik Voenno-vozdushnoi Akademii, 2023, no. 1 (51), pp. 237-244.

9. Mozgalevskii A.V., Kalyavin V.P., Kostandi G.G. Diagnostirovanie elektronnykh sistem (Diagnosis of electronic systems), Leningrad, Sudostroenie, 1984, 224 p.

10. Tekhnicheskaya diagnostika. Kategorii kontroleprigodnosti ob"ektov diagnostirovaniya. GOST 24029-80 (Technical diagnostics. Categories of testability of diagnostic objects State Standard 24029-80), Moscow, Izdatel'stvo standartov, 1980, 11 p.

11. Koziratskii Yu.L. Meshcheryakov D.V., Panov S.A., Kalinin V.S. Metodicheskie osnovy analiza i oformleniya nauchnykh rezul'tatov voennykh issledovanii (Methodological foundations for the analysis and presentation of scientific results of military research), Voronezh: VUNTs VVS «VVA», 2021, 459 p.

12. Flakh P. Mashinnoe obuchenie. Nauka i iskusstvo postroeniya algoritmov, kotorye izvlekayut znaniya iz dannykh (Machine learning. The science and art of building algorithms that extract knowledge from data), Moscow, DMK Press, 2015, 400 p.

13. Travin A.A., Kalashnikov E.A., Bakradze L.G. Trudy MAI, 2022, no. 127. URL: https://trudymai.ru/eng/published.php?ID=170352. DOI: 10.34759/trd-2022-127-23

14. Sokolov D.Yu. Trudy MAI, 2022, no. 123. URL: https://trudymai.ru/eng/published.php?ID=165563. DOI: 10.34759/trd-2022-123-17

15. Zheleznyakov A.O., Sidorchuk V.P., Podrezov S.N. Trudy MAI, 2022, no. 123. URL: https://trudymai.ru/eng/published.php?ID=165538. DOI: 10.34759/trd-2022-123-26

16. Vneshnie vozdeistviya. Dannye o vozdeistvii na elektrotekhnicheskoe oborudovanie vibratsii i udarov. Oborudovanie, transportiruemoe reaktivnym samoletom s neizmenyaemoi geometriei kryla. GOST R 57211-2016. (External influences. Data on the effects of vibration and shock on electrical equipment. Equipment transported by a jet aircraft with fixed wing geometry. State Standard R 57211-2016), Moscow, Standartinform, 2016, 35 p.

17. Bukirev A.S. Svidetel'stvo o gosudarstvennoi registratsii programmy dlya EVM № 2023611857 RF, 25.01.2023.

18. Bukirev A.S., Ippolitov S.V., Kryachkov V.N., Savchenko A.Yu. Patent № 2802976 C1 RF. MPK G06N 3/02, 05.09.2023.

19. Bukirev A.S. Svidetel'stvo o gosudarstvennoi registratsii programmy dlya EVM № 2023669752 RF, 20.09.2023.

20. Bukirev A.S. Svidetel'stvo o gosudarstvennoi registratsii programmy dlya EVM № 2023661942 RF, 05.06.2023.

Download