Іntelligent methods for predicting failures of power electrotechnical equipment based on multilevel digital twins
DOI:
https://doi.org/10.33216/1998-7927-2025-295-9-81-90Keywords:
digital twin, intelligent forecasting, machine learning, residual life, technical diagnostics, power equipmentAbstract
The article explores the use of intelligent approaches to predicting failures of power electrical equipment through the implementation of multilayer digital twins that integrate physical modeling, machine learning technologies, and big data analytics in order to enhance the reliability of power grid operation. The study aims to develop and experimentally validate a comprehensive digital model that synthesizes physics-informed mathematical equations with deep learning algorithms to achieve higher accuracy in residual life prediction and minimize the occurrence of technical failures in transformer equipment and asynchronous electric motors. The methodological framework of the research is based on the formation of a multilayer architecture of a digital twin, which includes: a sensor layer for data aggregation, a physical modeling layer of electromagnetic, thermal, and mechanical phenomena, an analytical layer employing machine learning algorithms, and a decision-making layer incorporating elements of fuzzy logic. The scientific results obtained showed that the use of a hybrid model can reduce the root mean square error (RMSE) to 0.031, reduce the mean absolute percentage error (MAPE) to 2.8%, and increase the F1-score to 0.93, which significantly exceeds the performance of classical diagnostic methods. It has been established that the completeness of sensor information and the frequency of model updates exert a decisive influence on predictive accuracy, while the system’s ability to automatically adapt to load variations has been experimentally confirmed. The developed architecture has shown strong resilience to data deficiency and a high level of generalization (CV < 0.06). The applied significance of the study lies in reducing emergency downtime by 30–40%, optimizing maintenance costs, and transforming the management of equipment condition toward a risk-oriented strategy. Future research directions include the standardization of digital twin technologies, the advancement of explainable AI methods, and the implementation of basic cybersecurity systems for next-generation industrial digital platforms.
References
1. Al-Shetwi A. Q., Atawi I. E., El-Hameed M. A., Abuelrub A. Digital Twin Technology for Renewable Energy, Smart Grids, Energy Storage and Vehicle-to-Grid Integration: Advancements, Applications, Key Players, Challenges and Future Perspectives in Modernising Sustainable Grids. IET Smart Grid. 2025. https://doi.org/10.1049/stg2.70026
2. Amin U., Kim D., Ahmed F. N., Ahmad G., Hossain M. J. Digital Twins for Smart Asset Management in the Energy Industry: State-of-the-Art. Expert Systems With Applications. 2025. Vol. 289. Article 128358. https://doi.org/10.1016/j.eswa.2025.128358
3. Chen S. (2025). Towards Prescriptive Maintenance Using Digital Twins and Artificial Intelligence [Licentiate thesis, Chalmers University of Technology, Department of Industrial and Materials Science, Gothenburg, Sweden]. 190 p. URL: https://research.chalmers.se/publication/547701/file/547701_Fulltext.pdf
4. Falekas G., Karlis A. Digital Twin in Electrical Machine Control and Predictive Maintenance: State-of-the-Art and Future Prospects. Energies. 2021. Vol. 14(18). P. 5933. https://doi.org/10.3390/en14185933
5. Fassi Y., Heiries V., Boutet J., Boisseau S. Towards Physics-Informed Machine-Learning-Based Predictive Maintenance for Power Converters: A Review. IEEE Transactions on Power Electronics. 2023. Vol. 39(2). pp. 2692-2720. https://doi.org/10.1109/TPEL.2023.3328438
6. Hosamo H. H., Nielsen H. K., Kraniotis D., Svennevig P. R., Svidt K. Digital Twin framework for automated fault source detection and prediction for comfort performance evaluation of existing non-residential Norwegian buildings. Energy and Buildings. 2023. Vol. 281. Article 112732. https://doi.org/10.1016/j.enbuild.2022.112732
7. Hu J., Xiao H., Ye Z., Luo N., Zhou M. Research and Prospects of Digital Twin-Based Fault Diagnosis of Electric Machines. Sensors. 2025. Vol. 25(8). p. 2625. https://doi.org/10.3390/s25082625
8. Lang L., Du J., Guo J., Xue G. A Digital Twin-Based Approach for Full Lifecycle Condition Monitoring of Power Grid Equipment. Frontiers in Artificial Intelligence and Applications. 2025. P. 201-210. DOI 10.3233/FAIA250268.
9. Makedon V., Myachin V., Plakhotnik O., Fisunenko N., Mykhailenko O. Construction of a model for evaluating the efficiency of technology transfer process based on a fuzzy logic approach. Eastern-European Journal of Enterprise Technologies. 2024. no 2(13(128)). p. 47-57. https://doi.org/10.15587/1729-4061.2024.300796.
10. Makedon V., Myachin V., Aloshyna T., Cherniavska I., Karavan N. Improving the Readiness of Enterprises to Develop Sustainable Innovation Strategies through Fuzzy Logic Models. Economic Studies (Ikonomicheski Izsledvania). 2025. Vol. 34, No. 5. pp. 165-179. https://archive.econ-studies.iki.bas.bg/2025/2025_05/2025_05_09.pdf
11. Nakti I., Mansouri M., Al-Hmouz R., Khedher A. Artificial Intelligence Techniques With Digital Twin for Fault Diagnosis in Interconnected Systems: A Review. IEEE Access. 2025. Vol. 13. P. 91860-91874. https://doi.org/10.1109/ACCESS.2025.3572563.
12. Rana S. AI-driven fault detection and predictive maintenance in electrical power systems: A systematic review of data-driven approaches, digital twins, and self-healing grids. American Journal of Advanced Technology and Engineering Solutions. 2025. Vol. 1(1), p. 258-289. https://doi.org/10.63125/4p25x993
13. Singh R. R., Bhatti G., Kalel D., Vairavasundaram I., Alsaif F. Building a Digital Twin Powered Intelligent Predictive Maintenance System for Industrial AC Machines. Machines. 2023. Vol. 11(8), 796. https://doi.org/10.3390/machines11080796
14. Thwe M. M., Ştefanov A., Rajkumar V. S., Palensky P. Digital Twins for Power Systems: Review of Current Practices, Requirements, Enabling Technologies, Data Federation, and Challenges. IEEE Access. 2025. Vol. 13. pp. 105517-105540. https://doi.org/10.1109/ACCESS.2025.3580055
15. Xu N., Hu L., Di X., Tang Y., Dong L., Sun X. Research Status and Progress of Digital Twin Models for Electric Power System Equipment. In Proceedings of the 4th International Conference on Cyber Security, Artificial Intelligence and the Digital Economy (CSAIDE 2025), 07–09 March 2025, Kuala Lumpur, Malaysia. ACM, New York, NY, USA, 2025, 7 p. https://doi.org/10.1145/3729706.3729778
16. Zhong D., Xia Z., Zhu Y., Duan J. Overview of predictive maintenance based on digital twin technology. Heliyon. 2023. Vol. 9(4). e14534. https://doi.org/10.1016/j.heliyon.2023.e14534
