A.Yu. Chesalov1, S.V. Gromov2
1, 2 Atlansis Software LLC (Tver, Russia)

1 achesalov@mail.ru, 2 s.gromov.dev@gmail.com

In the face of growing demands for reliability, cost-effectiveness, and safety of industrial assets, technologies that enable the transition from scheduled and reactive maintenance to data-driven equipment condition management are becoming increasingly important. One of the most promising areas in this field is predictive maintenance, where one of the key tasks is estimating the remaining service life of industrial equipment. Accurate predictions enable optimization of maintenance and repair schedules, minimization of unscheduled downtime, reduction in operating costs, and prevention of catastrophic consequences from industrial equipment failure. Modern approaches to predicting the remaining service life of industrial equipment have various advantages and limitations. However, industrial equipment often exhibits complex degradation mechanisms influenced by numerous internal and external factors. Under these conditions, the challenge of creating new architectures and algorithms for the development of industrial automated systems is particularly pressing.

The purpose of the study is to develop a basic algorithm and architecture for an automated predictive maintenance system that implements a combined method for assessing the technical condition and predicting the remaining service life of industrial equipment.

The proposed architecture and algorithm have been implemented in the Python programming language as a separate software module. A distinctive feature of the software implementation is its versatility, allowing for the handling of a large number of sensors and large volumes of data, as well as the ability to integrate with other systems and maintain independence from any hardware or software platform.

The results of the study and work can be used to improve the performance of industrial automated predictive or prescriptive maintenance systems, as well as computerized maintenance management systems.

Chesalov A.Yu., Gromov S.V. Development of a basic architecture of an automated system for predictive maintenance and assessment of the remaining useful life of industrial equipment. Neurocomputers. 2026. V. 28. № 2. P. 21–33. DOI: https://doi.org/10.18127/ j19998554-202602-02 (in Russian)

1. Chesalov A.Yu. Primenenie proryvnykh tekhnologij iskusstvennogo intellekta v promyshlennykh ekosistemakh Industrii 4.0. Perspektivnye integratsionnye protsessy v mirovoj ekonomike: noopodkhod. Sb. materialov IX Sankt-Peterburgskogo mezhdunarodnogo ekono­micheskogo kongressa (SPEK-2024). Pod obshch. red. S.D. Bodrunova. T. 2. M.: INIR im. S.Yu. Vitte. 2024. S. 176–184. (in Russian)
2. Palyukh B.V., Chesalov A.Yu. Sovremennye podkhody k sozdaniyu avtomatizirovannykh sistem prognoziruemogo obsluzhivaniya mnogostadijnykh tekhnologicheskikh protsessov v promyshlennosti. Sb. nauch. trudov XXVII Rossijskoj nauch. konf. «Inzhiniring predpriyatij i upravlenie znaniyami» (IP&UZ–2024). Pod nauch. red. Yu.F. Tel'nova. M.: FGBOU VO «REU im. G.V. Plekhanova». 2024. T. 1. S. 351–357. (in Russian)
3. Palyukh B.V., Chesalov A.Yu. Rol' sovremennykh tekhnologij iskusstvennogo intellekta v sozdanii i razvitii avtomatizirovannykh sistem prognoziruemogo i predpisyvayushchego obsluzhivaniya v promyshlennosti. Sovremennaya nauka: aktual'nye problemy teorii i praktiki. Seriya «Estestvennye i tekhnicheskie nauki». 2025. № 5. S. 147–155. DOI 10.37882/2223-2966.2025.05.29. (in Russian)
4. Ullah S. et al. Fault diagnosis of a multistage centrifugal pump using explanatory ratio linear discriminant analysis. Sensors. 2024. V. 24. № 6. P. 1830. DOI: 10.3390/s24061830.
5. Barkov A.V., Zhukovskij D.L. Sovremennye podkhody v ispol'zovanii diagnostiki dlya effektivnoj ekspluatatsii nasosno-kompressornogo oborudovaniya. Severo-Zapadnyj uchebnyj tsentr [Elektronnyj resurs]. URL: https://vibro-expert.ru/sovremennie-podxodi-v-ispolzovanii-diagnostiki-dlya-effektivnoie-ekspluatacii-nasosno-kompressornogo-oborudovaniya.-doklad.html (data obrashcheniya: 16.01.2026). (in Russian)
6. Bashirov M.G., Prakhov I.V., Samorodov A.V. Opredelenie tekhnicheskogo sostoyaniya nasosno-kompressornogo oborudovaniya po znacheniyam parametrov vysshikh garmonicheskikh sostavlyayushchikh tokov i napryazhenij, generiruemykh dvigatelem elektroprivoda. Fundamental'nye issledovaniya. 2010. № 12. S. 62–65. (in Russian)
7. Meng F. et al. A novel fault diagnosis strategy for diaphragm pumps based on signal demodulation and PCA-ResNet. Sensors. 2024. V. 24. № 5. P. 1578. DOI: 10.3390/s24051578.
8. Gao Y., Ahmad Z., Kim J.-M. The prediction of the remaining useful life of rotating machinery based on an adaptive maximum second-order cyclostationarity blind deconvolution and a convolutional LSTM autoencoder. Sensors. 2024. V. 24. P. 2382. DOI: 10.3390/s24082382.
9. Liu J., Hou B., Lu M., Wang D. Box–Cox transformation-based state-space modeling as a unified prognostic framework for degradation linearization and RUL prediction enhancement. Reliability Engineering & System Safety. 2024. V. 244. № 109952. DOI: 10.1016/j.ress.2024.109952.
10. Hyndman R.J., Athanasopoulos G. Forecasting: principles and practice. Ed. 3rd. Melbourne: OTexts. 2021 [Electronic resource]. URL: https://otexts.com/fpp3/ (data obrashcheniya 14.11.2025).
11. Koroleva E.V., Gorelik A.Yu. Prognozirovanie kotirovok finansovykh aktivov s uchetom ekzogennykh faktorov: ispol'zovanie modelej ARIMAX i SARIMAX. Estestvenno-gumanitarnye issledovaniya. 2025. № 2 (58). S. 575–579. (in Russian)
12. Durbin J., Siem J.K. Time series analysis by state space methods. Ed. 2nd. Oxford Statistical Science Series. Oxford. 2012.
13. Rostilov D.A., Kaurkin M.N., Ibraev R.A. Sravnenie metodov usvoeniya dannykh na osnove klassicheskogo, ansamblevogo i lokal'nogo fil'tra Kalmana na primere uravneniya advektsii i zadachi Lorentsa. Vychislitel'nye metody i programmirovanie. 2018. T. 19. № 4. S. 507–515. DOI: 10.26089/NumMet.v19r445. (in Russian)
14. Feng D., Guan W., Wan W. Remaining useful life estimation based on Wiener degradation process with mixed random effects. Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling. 2020. P. 1–6. DOI: 10.1109/APARM49247.2020.9209431.
15. Chetvertakova E.S. Issledovanie degradatsionnykh gamma-modelej so sluchajnym i fiksirovannym effektami. Vestnik IrGTU. 2018. T. 22. № 6 (137). S. 120–128. (in Russian)
16. Galar D., Kumar U., Lee J. et al. Remaining useful life estimation using time trajectory tracking and support vector machines. Journal of Physics: Conference Series. 2012. P. 364. DOI: 10.1088/1742-6596/364/1/012063.
17. Mikulik I.I., Blagoveshchenskaya E.A. O prognosticheskoj sposobnosti rasshirennoj modeli Koksa: sravnitel'nyj analiz. Intellektual'nye tekhnologii na transporte. 2025. № 2 (42). S. 58–70. DOI: 10.20295/2413-2527-2025-242-58-70. (in Russian)
18. Bai S., Kolter J.Z., Koltun V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv:1803. 01271. 2018 [Electronic resource]. URL: https://arxiv.org/abs/1803.01271.
19. Rapakov G.G., Gorbunov V.A., Dianov S.V., Elizarova L.V. Issledovanie LSTM-nejrosetevogo podkhoda pri modelirovanii vremennykh ryadov. Vestnik Cherepovetskogo gosudarstvennogo universiteta. 2023. № 3 (114). S. 47–54. (in Russian)
20. Bai R., Noman K., Yang Y. et al. Towards trustworthy remaining useful life prediction through multi-source information fusion and a novel LSTM‑DAU model. Reliability Engineering & System Safety. 2024. V. 245. Article 110047. DOI: 10.1016/j.ress.2024.110047.
21. Gorid'ko K.A., Timashev E.O., Volkov M.G. i dr. Obzor opyta prognozirovaniya otkazov UETsN metodami mashinnogo obucheniya. Zhurnal «Neftegaz.RU». 2025. № 1. S. 58–64. (in Russian)
22. Yumaguzin U.F., Bashirov M.G. Prognozirovanie ostatochnogo resursa oborudovaniya predpriyatij neftegazovoj otrasli. Fundamental'nye issledovaniya. 2014. № 3-2. S. 277–280. (in Russian)
23. Liu J., Hou B., Lu M., Wang D. Box–Cox transformation-based state-space modeling as a unified prognostic framework for degradation linearization and RUL prediction enhancement. Reliability Engineering & System Safety. 2024. V. 244. Article 109952. DOI: 10.1016/j.ress. 2024.109952.
24. Zhang K., Xia T., Xu Y. et al. A hybrid prognostic & health management framework across multi-level engineering systems with scalable convolution neural networks and adjustable functional regression models. Advanced Engineering Informatics. 2024. V. 61. Article 102566. DOI: 10.1016/j.aei.2024.102566.
25. Bai R., Noman K., Yang Y. et al. Towards trustworthy remaining useful life prediction through multi-source information fusion and a novel LSTM-DAU model. Reliability Engineering & System Safety. 2024. V. 245. Article 110047. DOI: 10.1016/j.ress.2024.110047.
26. Zhu Y., Zhou X., Cheng J. et al. Data and model synergy-driven rolling bearings remaining useful life prediction approach based on deep neural network and Wiener process. Journal of Manufacturing Science and Engineering. 2025. V. 147. № 4. DOI: 10.1115/1.4067092.
27. Rukin M.V., Molchanova V.A., Urazakov K.R. Metodika opredeleniya narabotki na otkaz ustanovki elektrotsentrobezhnogo nasosa. Izves­tiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov. 2022. T. 333. № 12. S. 219–229. DOI: 10.18799/24131830/ 2022/12/3792. (in Russian)
28. Feng D., Guan W., Wan W. Remaining useful life estimation based on Wiener degradation process with mixed random effects. Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling. 2020. P. 1–6. DOI: 10.1109/APARM49247.2020.9209431.
29. Galar D., Kumar U., Lee J. et al. Remaining useful life estimation using time trajectory tracking and support vector machines. Journal of Physics: Conference Series. 2012. P. 364. DOI: 10.1088/1742-6596/364/1/012063.
30. Mikulik I.I., Blagoveshchenskaya E.A. O prognosticheskoj sposobnosti rasshirennoj modeli Koksa: sravnitel'nyj analiz. Intellektual'nye tekhnologii na transporte. 2025. № 2 (42). S. 58–70. DOI: 10.20295/2413-2527-2025-242-58-70. (in Russian)
31. Gorid'ko K.A., Timashev E.O., Volkov M.G. i dr. Obzor opyta prognozirovaniya otkazov UETsN metodami mashinnogo obucheniya. Zhurnal «Neftegaz.RU». 2025. № 1. S. 58–64 [Elektronnyj resurs]. URL: https://magazine.neftegaz.ru/archive/874627/ (data obrashcheniya: 19.01.2026). (in Russian)
32. Koroleva E.V., Gorelik A.Yu. Prognozirovanie kotirovok finansovykh aktivov s uchetom ekzogennykh faktorov: ispol'zovanie modelej ARIMAX i SARIMAX. Estestvenno-gumanitarnye issledovaniya. 2025. № 2 (58). S. 575–579. (in Russian)
33. Rostilov D.A., Kaurkin M.N., Ibraev R.A. Sravnenie metodov usvoeniya dannykh na osnove klassicheskogo, ansamblevogo i lokal'nogo fil'tra Kalmana na primere uravneniya advektsii i zadachi Lorentsa. Vychislitel'nye metody i programmirovanie. 2018. T. 19. № 4. S. 507–515. DOI: 10.26089/NumMet.v19r445. (in Russian)
34. Rapakov G.G., Gorbunov V.A., Dianov S.V. i dr. Issledovanie LSTM-nejrosetevogo podkhoda pri modelirovanii vremennykh ryadov. Vestnik Cherepovetskogo gosudarstvennogo universiteta. 2023. № 3 (114). S. 47–54. (in Russian)
35. Chesalov A.Yu. Matematicheskaya model' snizheniya neopredelennosti na osnove teorii svidetel'stv Dempstera–Shafera na urovne sbora dannykh v avtomatizirovannykh sistemakh prognoziruemogo obsluzhivaniya. Dinamika slozhnykh sistem. 2025. T. 19. № 4. S. 62–74. DOI: 10.18127/j19997493-202504-07. (in Russian)
36. Palyukh B.V., Chesalov A.Yu. Novaya arkhitektura avtomatizirovannykh sistem prognoziruemogo obsluzhivaniya na osnove tekhnologij iskusstvennogo intellekta i algoritmov potentsiala negativnosti rassoglasovaniya. Sb. statej XIV Mezhdunarodnoj nauchno-prakticheskoj konf. «Abalkinskie chteniya». Pod red. S.D. Valenteya. M.: FGBOU VO «REU im. G.V. Plekhanova». 2025. S. 283–290. (in Russian)
37. Chesalov A.Yu. Tendentsii razvitiya periferijnogo iskusstvennogo intellekta v avtomatizatsii tekhnologicheskikh protsessov. Avtomatizatsiya v promyshlennosti. 2025. № 7. S. 9–14. (in Russian)