Developing maintenance and refurbishment model of aerial vehicles with artificial neural network applicaion

Aeronautical and Space-Rocket Engineering


DOI: 10.34759/vst-2022-1-19-26

Аuthors

Dolgov O. S.*, Safoklov B. B.**

*e-mail: dolgov@mai.ru
**e-mail: safoklovbb@mai.ru

Abstract

Maintaining the specified safety, reliability and availability characteristics of the aerial vehicles (AV) with long operation life and after-sales service, can significantly exceed their purchase cost. Conceptually new approaches are required nowadays in the industry to ensure the quality improvement level, increase in the safety and economic efficiency of the AV for the aviation industry enterprises. Highly efficient AV with low life cycle cost (LLC) and high utilization factor are economically viable for the aircraft operators (consumers). One of the ways of the LCC reduction consists in optimizing the aircraft maintenance system during operation, refurbishment and overhaul.

Manufacturing companies that are among the first in the aviation industry to integrate predictive maintenance (PM) into the after-sales service (AS) and maintenance and repair systems (MRO), all other things being equal, will be able to provide the most competitive product in the aviation industry. This concept implementation is complicated since the PTO concept involves continuous monitoring of a large number of parameters, which does not allow fully implementing it in the aviation industry due to the lack of global broadband data transmission from the aircraft throughout the entire flight.

Mathematical method of artificial neural networks (ANN) application is the least costly for the incoming big data streaming analysis.

The gist of the ANN utilization consists in processing the information array obtained from the product state monitoring system to predict the available solutions on the product maintenance.

The way to the MRO optimization is integration with the Aircraft Health Monitoring (AHM), in which, the ANN employing as a tool is one of the concepts.

The authors propose application of the developed model of the aircraft maintenance and refurbishment for the ANN utilization, with the ANN employing as a predictive maintenance tool.

Keywords:

aerial vehicle, artificial neural networks, maintenance design, aircraft maintenance and refurbishment, predictive maintenance

References

  1. Pogosyan M.A., Bratukhin A.G., Savel’evskikh E.P., Strelets D.Yu., Zlygarev V.A. Vestnik mashinostroeniya, 2017, no. 5, pp. 60–65.

  2. Dolgov O.S., Safoklov B.B., Aruvelli S. Airdrop of discharged batteries using an unmanned aerial vehicle as a method of increasing the range of an electric aircraft by reducing the mass in flight. Journal of Aeronautics, Astronautics and Aviation, 2021, vol. 53, no. 2, pp. 235-239. DOI: 10.6125/JoAAA.202106_53(2).15

  3. Sudov E.V., Petrov A.N., Petrov A.V., Osyaev A.T., Serebryanskii S.A. Tekhnologii integrirovannoi logisticheskoi podderzhki v protsessakh zhiznennogo tsikla aviatsionnoi tekhniki (Technologies of integrated logistics support in the processes of aviation equipment life cycle), Moscow, Editus, 2018, 174 p.

  4. Abashev O.V., Kuprikov M.Y. An application of artificial neural networks in aircraft design. Aerospace MAI Journal, 2008, vol. 15, no. 5, pp. 27-33.

  5. Buhl A., Hjertén H. Evaluation of Artificial Neural Networks for Predictive Maintenance. Department of Computer Science, Faculty of Engineering LTH, 2018, 77 p.

  6. Legkonogikh D.S. Izvestiya Samarskogo nauchnogo tsentra Rossiiskoi akademii nauk, 2012, vol. 14, no. 4-2, pp. 639-643.

  7. Vozdushnye suda grazhdanskoi aviatsii. Pryamye zatraty na tekhnicheskoe obsluzhivanie pri ekspluatatsii. Metodika rascheta. OST 1 02799–2012 (Civil aviation aircraft. Direct maintenance costs during operation. Calculation method. Industry Standard 1 02799-2012), Moscow, Standartinform, 2013, 16 p.

  8. Ali R., Al-Shamma O. Comparative Study of Cost Estimation Models used for Preliminary Aircraft Design. Global Journal of Researches in Engineering: B Automotive Engineering, 2014, vol. 14, no. 4. URI: https://globaljournals.org/GJRE_Volume14/2-A-Comparative-Study.pdf

  9. Curran R., Chou S.-Y., Trappey A. Collaborative Product and Service Life Cycle Management for a Sustainable World. 15th ISPE International Conference on Concurrent Engineering. Springer, 2008, 643 p.

  10. Makarov V.V. Uchenye zapiski fizicheskogo fakul’teta Moskovskogo universiteta, 2017, no. 5. URI: http://uzmu.phys.msu.ru/file/2017/5/1751406.pdf

  11. Karim Md.R. TensorFlow: Powerful Predictive Analytics with TensorFlow: Predict valuable insights of your data with TensorFlow. Packt Publishing, 2018, 165 p.

  12. Bautin A.A., Svirskiy Y.A. Neural networks technologies application in problems of critical places status monitoring of transport aircraft structure. Aerospace MAI Journal, 2020, vol. 27, no. 4, pp. 81-91. DOI: 10.34759/vst-2020-4-81-91

  13. Aircraft Health Monitoring (AHM) integration in MSG-3. International Maintenance Review Board Policy Board (IMRBPB), 2018, 33 p.

  14. Rivas A., Fraile J.M., Chamoso P., González-Briones A., Sittón I., Corchado J.M. A Predictive Maintenance Model Using Recurrent Neural Networks. In: Martínez Álvarez F., Troncoso Lora A., Sáez Muñoz J., Quintián H., Corchado E. (eds) 14th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2019). Advances in Intelligent Systems and Computing. Vol. 950. Springer, Cham. DOI: 10.1007/978-3-030-20055-8_25

  15. Mezhgosudarstvennyi standart. Sistema tekhnicheskogo obsluzhivaniya i remonta tekhniki. Terminy i opredeleniya. GOST 18322-2016 (Interstate Standard. System of technical maintenance and repair of equipment. Terms and definitions. State Standard 18322-2016), Moscow, Standartinform, 2013, 16 p.

  16. Natsional’nyi standart Rossiiskoi Federatsii. Sistemy promyshlennoi avtomatizatsii i integratsiya. Sistemy tekhnicheskogo obsluzhivaniya i remonta. Terminy i opredeleniya. GOST R 57329-2016/EN 13306:2010 (Automation systems and integration. Maintenance and repair systems. Terms and definitions. State Standard R 57329-2016/EN 13306:2010), Moscow, Standartinform, 2020, 23 p.

  17. Mezhgosudarstvennyi standart. Sistema tekhnicheskogo obsluzhivaniya i remonta tekhniki pokazateli dlya otsenki remontoprigodnosti. Terminy i opredeleniya. GOST 21623-76 (System of technical maintenance and repair of equipment. Characteristics for evaluation of maintainability and repairability. Terms and definitions. State Standard 21623-76), Moscow, Standartinform, 2006, 16 p.

  18. Bronnikov A.M. Nauchnyi Vestnik MGTU GA, 2017, vol. 20, no. 6, pp. 89-98. DOI: 10.26467/2079-0619-2017-20-6-89-98

  19. Demidova L.A. Recurrent neural networks’ configurations in the predictive maintenance problems. Workshop on Materials and Engineering in Aeronautics (16–17 October 2019; Moscow). Moscow, Institute of Physics Publishing, 2020. DOI: 10.1088/1757-899X/714/1/012005

  20. Vlasov A.I, Grigor’ev P.V., Krivoshein A.I. Nadezhnost’ i kachestvo slozhnykh system, 2018, no. 2(22), pp. 26-35. DOI 10.21685/2307-4205-2018-2-4

  21. Krenek J., Kuca K., Blazek P., Krejcar O., Jun D. Application of Artificial Neural Networks in Condition Based Predictive Maintenance. In: Król D., Madeyski L., Nguyen N. (eds) Recent Developments in Intelligent Information and Database Systems. Studies in Computational Intelligence, 2016, vol. 642. Springer, Cham. DOI: 10.1007/978-3-319-31277-4_7

  22. Günnemann N., Pfeer J. Predicting defective engines using convolutional neural networks on temporal vibration signals. Proceedings of Machine Learning Research, 2017, vol. 74, pp. 92-102.

mai.ru — informational site of MAI

Copyright © 1994-2024 by MAI