Systems and Means of Informatics
2025, Volume 35, Issue 4, pp pp 92-110
CONSTRUCTION AND ANALYSIS OF MODEL TO PREDICT THE TECHNICAL STATE OF RAILWAY CAR AXLE BOXES USING INTELLIGENT PREDICTIVE ANALYTICS METHODS
- O. V. Druzhinina
- E. R. Korepanov
- I. V. Makarenkova
- V. V. Maksimova
- A. A. Petrov
Abstract
The paper is devoted to the study of the problem of constructing and analyzing a model for predicting the technical state of axle boxes of railway cars based on the use of artificial intelligence methods. The relevance of this problem is related to the need to create and improve high-tech and energy-efficient data analysis tools for diagnosing the technical condition of elements and systems of transport infrastructure. It is proposed to use the LSTM (Long ShortTerm Memory) neural network architecture to predict the state when processing sequential data (time series). Synthetic datasets for neural network training are generated using the developed simulation stochastic model of thermal control of axle boxes. The performed computer modeling in the PyTorch environment allowed to conduct a comparative analysis of the results of computational experiments and to evaluate the effectiveness of LSTM training in the framework of the problem under consideration. The constructed predictive analytics model can serve as the basis for the ABITech Thermal Forecast Module, a software package for diagnosing the technical state of axle boxes.
[+] References (30)
- Sokolov, I. A. 2019. Theory and practice of application of artificial intelligence methods. Her. Russ. Acad. Sci. 89(2): 115-119. doi: 10.1134/S1019331619020205. EDN: VYXJUC.
- Vaswani, A., N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, L. Kaiser, and I. Polosukhin. 2017. Attention is all you need. Cornell University. 15 p. Available at: https://arxiv.org/pdf/1706.03762v5 (accessed November 18, 2025).
- Smagulova, K., and A. P. James. 2019. A survey on LSTM memristive neural network architectures and applications. Eur. Phys. J. - Spec. Top. 228(10):2313-2324. doi: 10.1140/epjst/e2019-900046-x. EDN: HRKIKB.
- Ghojogh, B., and A. Ghodsi. 2023. Recurrent neural networks and long shortterm memory networks: Tutorial and survey. Cornell University. 15 p. Available at: https://arxiv.org/pdf/2304.11461v1 (accessed November 18, 2025).
- Fedotov, M.V., and V.V. Grachev. 2021. Prediktivnaya analitika tekhnicheskogo sostoyaniya sistem teplovozov s ispol'zovaniem neyrosetevykh prognoznykh modeley [Predictive analytics of the technical condition of diesel locomotive systems using neural network predictive models]. Byulleten' rezul'tatov nauchnykh issledovaniy [Bulletin of Scientific Research Results] 3:102G14. doi: 10.20295/2223-9987-2021-3-102-114. EDN: IKEHIN.
- Chernukhin, A.V., E.A. Bogdanova, T.V. Savitskaya, D. G. Kulakov, and I. R. Pavlov. 2024. Sistema prediktivnoy analitiki tekhnicheskogo sostoyaniya eksgaustera aglomashiny s pomoshch'yu metodov iskusstvennogo intellekta [Predictive analytics system for the technical condition of a sinter extractor using artificial intelligence methods]. Iskusstvennyy intellekt i prinyatie resheniy [Artificial Intelligence and Decision Making] 3:87G03. doi: 10.14357/20718594240307. EDN: HZKRKY.
- Druzhinina, O. V., E. R. Korepanov, A. A. Petrov, I. V. Makarenkova, and V. V. Maksimova. 2023. Postroenie modeli generatsii dannykh dlya resheniya zadach klassifikatsii v diagnostike neispravnostey transportnykh sistem [Design of a data generation model for solving classification problems in the diagnosis of transport system malfunctions]. Nelineynyy mir [Nonlinear World] 21 (3): 16{26. doi: 10.18127/j20700970- 202303-02. EDN: FOIGST.
- Belousov, V. V., O. V. Druzhinina, E. R. Korepanov, I. V. Makarenkova, and V.V. Maksimova. 2021. Podkhod k otsenke tekhnicheskogo sostoyaniya elementov i uzlov transportnykh sistem s primeneniem metodov neyrosetevogo modelirovaniya i tekhnologii tsifrovykh dvoynikov [An approach to assessing the technical condition of elements and nodes of transport systems using neural network modeling methods and digital twin technology]. Neyrokomp'yutery: razrabotka, primenenie [Neurocomputers] 23(5):5{20. doi: 10.18127/j 19998554-202105-01. EDN: EIAAOZ.
- Mironov, A. A., V. L. Obraztsov, and A. E. Pavlyukov. 2012. Teoriya i praktika beskontaktnogo teplovogo kontrolya buksovykh uzlov v poezdakh [Theory and practice of noncontact thermal monitoring of axle boxes in trains]. Ekaterinburg: Assorti. 396 p.
- Mironov, A. A., V. L. Obraztsov, V. S. Mityushev, and K.V. Grigor'ev. 2011. Teplovoy kontrol' buksovykh uzlov sredstvami KTSM-02 [Thermal control of axle boxes by means of KTSM-02]. Avtomatika, svyaz', informatika [Automation, Communications, Informatics] 12:9GB EDN: OMRLCN.
- KTSM-03 - kompleks tekhnicheskikh sredstv mnogofunktsional'nyy dlya diagnostiki khodovykh chastey zheleznodorozhnogo podvizhnogo sostava [KTSM-03: Multifunctional technical complex for diagnostics of running gear of railway rolling stock]. Available at: https://infotecs-at.ru/products/7 (accessed November 18, 2025).
- Belousov, V. V., O. V. Druzhinina, E. R. Korepanov, I. V. Makarenkova, and V. V. Maksimova. 2022. Primenenie neyronnykh setey dlya resheniya zadach klassifikatsii pri vyyavlenii neispravnostey transportnykh sistem [The use of neural networks for solving classification problems in diagnosing malfunctions of transport systems]. Neyrokomp'yutery: razrabotka, primenenie [Neurocomputers] 24(4):18G7. doi: 10.18127/j 19998554-202204-02. EDN: ZZAOWY.
- Gavrilov, V. E., and A. A. Zatsarinny. 2022. Problemy i ugrozy vnedreniya nekotorykh novykh tsifrovykh tekhnologiy [Problems and threats of some new digital technologies implementation]. Sistemy i Sredstva Informatiki - Systems and Means of Informatics 32(3):15{25. doi: 10.14357/08696527220302.EDN: QKESJR.
- Gavrilov, V. E., and A. A. Zatsarinny. 2024 Osobennosti obespecheniya funktsional'noy bezopasnosti avtomatizirovannykh sistem s primeneniem tekhnologii iskusstvennogo intellekta [Specific features of ensuring the functional safety of automated systems using artificial intelligence technology]. Sistemy i Sredstva Informatiki - Systems and Means of Informatics 34(3):23-34. doi: 10.14357/08696527240303. EDN: RPQAIE.
- WaggonTracker. 2025. Intelligent rail freight transport featured by WaggonTracker. PJ Monitoring GmbH. Available at: https://pjm.co.at/en/waggontracker/ (accessed November 18, 2025).
- Danilenko, D. V., A. N. Strelchenko, and A. V. Grigoriev. 2023. Sistema monitoringa sostoyaniya gruzovogo zheleznodorozhnogo vagona [System for monitoring the condition of a freight railway car]. Patent RF No. 2798214. EDN: DMKCVZ.
- IEEE. 2015. Standard for low-rate wireless networks 802.15.4-2015. Available at: https://ieeexplore.ieee.org/document/7460875 (accessed November 18, 2025).
- LoRaWAN specification v1.1. 2017. LoRa Alliance. 101 p. Available at: https://resources.lora-alliance.org/technical-specifications/lorawan- specification-v1-1 (accessed November 18, 2025).
- Monagle, D., E.A. Ponce and S.B. Leeb. 2024. Rule the Joule: An energy management design guide for self-powered sensors. IEEE Sens. J. 24(1):6{15. doi: 10.1109/JSEN.2023.3336529.
- Zewe, A. 2024. Self-powered sensor automatically harvests magnetic energy. MIT News. Available at: https://news.mit.edu/2024/self-powered-sensor-harvests- magnetic-energy-0118 (accessed November 18, 2025).
- Prokhorov, A., and M. Lysachev. 2020. Tsifrovoy dvoynik: analiz, trendy, mirovoy opyt [Digital twin: Analysis, trends, and global experience]. Ed. A. Borovkov. Moscow: Al'yans Print. 401 p.
- Kovalyov, S. P. 2021. Primenenie neyronnykh setey glubokogo obucheniya v matematicheskom obespechenii tsifrovykh dvoynikov elektroenergeticheskikh sistem [Employing deep learning neural networks in mathematical basis of digital twins of electrical power systems]. Sistemy i Sredstva Informatiki - Systems and Means of Informatics 31(1):133-144. doi: 10.14357/08696527210111. EDN: NQPUXR.
- Kolla, B., and G. R. Kanagachidambaresan, eds. 2021. Programming with TensorFlow. Solution for edge computing applications. Cham, Switzerland: Springer. 284 p. doi: 10.1007/978-3-030- 57077-4.
- Adaptirovannye neyrosetevye algoritmy na otechestvennykh protsessorakh [Adapted neural network algorithms on domestic processors]. 2025. Brain-Tools.RU. Available at: https://www.braintools.ru/article/13697 (accessed November 18, 2025).
- Efimov, I. N., E. A. Morozov, and K. M. Selivanov. 2014. Komp'yuternoe modelirovanie dinamicheskikh system [Computer modeling of dynamic systems]. Izhevsk: Institute of Computer Science. 134 p.
- Kuznetsov, S. 2014. Matematicheskie modeli protsessov i sistem tekhnicheskoy ekspluatatsii avioniki kak markovskie tsepi [Processes and systems of avionics technical operation as Markov chains mathematical models]. Nauchnyy vestnik MGTU GA [Civil Aviation High Technologies] 201:56-64. EDN: RYFUTT.
- Galazhinskaya, O.N., and S. P. Moiseeva. 2016. Teoriya sluchaynykh protsessov. Ch. 2: Markovskie protsessy [Theory of random processes. Part 2: Markov processes]. Tomsk: TGU. 126 p.
- Druzhinina, O.V., E.R. Korepanov, A. A. Petrov, I.V. Makarenkova, and V.V. Maksimova. 2024. Razrabotka instrumental'no-metodicheskogo obespecheniya dlya otsenki tekhnicheskogo sostoyaniya uzlov transportnykh sistem s primeneniem neyronnykh setey, algoritmov nechetkogo i stokhasticheskogo modelirovaniya [Development of instrumental and methodological support for assessing the technical condition of transport system nodes using neural networks, fuzzy and stochastic modeling algorithms]. Neyrokomp'yutery: razrabotka, primenenie [Neurocomputers] 26(2): 13{22. doi: 10.18127/j 19998554-202402-02. EDN: EPQPVE.
- Druzhinina, O.V., A. A. Petrov, I.V. Makarenkova, and V.V. Maksimova. 2024. Programma sinteza dannykh i pereobucheniya neyroseti dlya tseli monitoringa i diagnostiki buksovykh uzlov zheleznodorozhnykh vagonov [The program of data synthesis and retraining of a neural network for the purpose of monitoring and diagnostics of axle boxes of railway wagons]. Certificate RF of State Registration of Computer Programs No. 2024663875. EDN: DAQSRM.
- Grandini, M., E. Bagli, and G. Visani. 2020. Metrics for multi-class classification: An overview. Cornell University. 17 p. Available at: https://arxiv.org/pdf/2008.05756
(accessed November 18, 2025).
[+] About this article
Title
CONSTRUCTION AND ANALYSIS OF MODEL TO PREDICT THE TECHNICAL STATE OF RAILWAY CAR AXLE BOXES USING INTELLIGENT PREDICTIVE ANALYTICS METHODS
Journal
Systems and Means of Informatics
Volume 35, Issue 4, pp 92-110
Cover Date
2025-12-25
DOI
10.14357/08696527250407
Print ISSN
0869-6527
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
data analysis; predictive analytics; computer modeling; neural networks LSTM; time series; machine learning algorithms; technical state assessment; intelligent transport systems
Authors
O. V. Druzhinina  , E. R. Korepanov  , I. V. Makarenkova  , V. V. Maksimova  ,
and A. A. Petrov
Author Affiliations
 Federal Research Center "Computer Science and Control", Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
 Yelets State University named after I. A. Bunin, 28 Kommunarov Str., Yelets 399770, Lipetsk Region, Russian Federation
|