Abstract
Efficient management of the industrial electric power supply system operational reliability requires timely forecasting of the restoration time (MTTR, Тv) and the failure flow parameter (ω) for different voltage classes and equipment types. These values determine the damage from underproduction and make it possible to form the budget for capital and current repairs of electrical grid and substation equipment. Traditional statistical methods (exponential smoothing, moving average, linear regression) do not always capture the nonlinear relationships between failures, weather conditions and substation load. This paper proposes an approach based on a neural network for predicting MTTR and ω for the next year. A multilayer perceptron with eight input features and two output features, namely MTTR and ω, was used to ensure stable operation over short time series. A sliding window method was used to evaluate the forecast accuracy. The initial data include emergency outage logs for 2008 2020, information on equipment quantity and substation capacity for the considered node (Central Power Station – Steam Blowing Power Station –Substation No. 95). For each voltage class (3, 6, 10, 35, 110, 220 kV) and equipment type (transformer, cable line, overhead line, circuit breaker), restoration time distribution laws are constructed and their evolution is assessed. The developed algorithm also calculates the undelivered electrical energy for the given power node. The results are visualized in multi-page reports (PDF, Excel) containing graphs of MTTR and ω dynamics, distribution law tables and the comparison of the predicted values with the actual data. The proposed method can be used by dispatch and maintenance services for planning preventive maintenance and assessing the probable damage from emergency outages.
Keywords
power supply reliability, failure flow parameter, restoration time, neural network forecasting, undelivered electricity, power node, restoration time distribution laws, emergency outages
1. Bolshev V.E. Using machine learning to predict power outages on 110 kV transmission lines based on the line parameters. Zhurnal Sibirskogo federalnogo universiteta. Tekhnika i tekhnologii [Journal of Siberian Federal University. Engineering & Technologies], 2025, vol. 18, no. 1, pp. 19–32. (In Russian).
2. Ziniev Sh.Z., Kargin V.A., Krotov I.V., Ezirbaev T.B. Forecasting of 35 kV overhead line failures based on neural network to ensure reliability of power supply for the agro industrial complex. Vestnik Altayskogo gosudarstvennogo agrarnogo universiteta [Bulletin of Altai State Agricultural University], 2026, no. 2 (256), pp. 79–85. (In Russian). doi: 10.53083/1996-4277-2026-256-2-79-85
3. Lukovenko A.S., Zenkov I.V. Methods for calculating reliability of a power supply system. Vestnik Irkutskogo gosudarstvennogo tekhnicheskogo universiteta [Proceedings of Irkutsk State Technical University], 2021, vol. 25, no. 1(156), pp. 57–65. (In Russian). doi: 10.21285/1814-3520-2021-1-57-65
4. Kovalev S.P. Application of deep learning neural networks in mathematical support of digital twins in electric power distribution systems. Sistemy i sredstva informatiki [Systems and Means of Informatics], 2021, vol. 31, no. 1, pp. 133–144. (In Russian). doi: 10.14357/08696527210111
5. Sukhachev I.S., Sidorov S.V., Sushkov V.V., Antropova V.R. Development of a model for predicting the technical condition of oilfield electromechanical equipment using an artificial neural network. Elektrotekhnicheskie i informatsionnye kompleksy i sistemy [Electrical and data processing facilities and systems], 2022, vol. 18, no. 1, pp. 88–94. (In Russian). doi: 10.17122/1999-5458-2022-18-1-88-94
6. Kolodenkova A.E., Vereshchagina S.S. Development of an integrated approach to detecting electrical equipment faults using convolutional neural networks. Izvestiya YuFU. Tekhnicheskie nauki [Proceedings of SFedU. Engineering Sciences], 2025, no. 1(243), pp. 279–287. (In Russian). doi: 10.18522/2311-3103-2025-1-279-287
7. Nekrasov I.V., Konstantinovsky Yu.D., Kukin N.S. Optimization of probabilistic operational characteristics of a neural network algorithm for predictive diagnostics of industrial equipment failures through automated training set processing. Dinamika i vibroakustika [Journal of Dynamics and Vibroacoustics], 2024, vol. 10, no. 3, pp. 81–92. (In Russian). doi: 10.18287/2409-4579-2024-10-3-81-92
8. Boyarkin D.A., Krupenyov D.S., Yakubovsky D.V. Machine learning methods in assessing the reliability of electric power systems by the Monte Carlo method. Vestnik Yuzhno Uralskogo gosudarstvennogo universiteta. Seria: Matematicheskoe modelirovanie i programmirovanie [Bulletin of the South Ural State University. Series “Mathematical Modelling, Programming & Computer Software”], 2018, vol. 11, no. 4, pp. 146–153. (In Russian). doi: 10.14529/mmp18041
9. Yang Y., Wang X., Li Y., Zhang Z., Chen J. Failure Prediction Based on Deep Neural Networks in Power Backbone Networks. 2024 22nd International Conference on Optical Communications and Networks (ICOCN). IEEE, 2024. doi: 10.1109/ICOCN63276.2024.10648539
10. Zhang Yu, Shi J., Huang Sh., Geng L., Wang Z., Sun H., Yu Q., Liu D., Yao X., Zuo W., Guo M., Che X. SHAP-Enhanced Artificial Intelligence Machine Learning Framework for Data-Driven Weak Link Identification in Regional Distribution Grid Power Supply Reliability. Energies, 2025, no. 18 (20), art. no. 5372. doi: 10.3390/en18205372
11. Varganova A.V., Efimova V.A., Makushin E.V., Pozin D.O., Mugalimov R.G. Search for Optimal Recloser Installation Locations in the 6-10 kV Distribution Network. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2024, no. 2(63), pp. 66-73. (In Russian). doi: 10.18503/2311-8318-2024-2(63)-66-73
Malafeev A.V., Zaitsev Ya.A. Predicting the Reliability of Industrial Enterprise Power Supply System Using Artificial Intelligence. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2026, no. 3(72), pp. 22-26. (In Russian). https://doi.org/10.18503/2311-8318-2026-3(72)-22-26