Abstract
The article considers the task of improving the energy efficiency of the electric power industry, namely, improving the accuracy of forecasting electricity consumption. The relevance and importance of short-term electricity consumption forecasting in sparsely populated areas is substantiated, which consists in an increase in the number of autonomous cottage settlements in many regions of Russia, and at the same time, a small number of scientific papers on this topic. A recurrent neural network LSTM was chosen to implement short-term forecasting. The mathematical model of LSTM has been adapted consisting of six equations describing: accounting for previous values of power consumption (LSTM latent state); exclusion of power consumption values that have not reached the threshold value (LSTM filter); updating the time series of power consumption values (adding new values to the LSTM memory); forecasting values of power consumption with a given step-gnosis (new latent state of LSTM). The architecture of an artificial neural network is proposed, which considers all external factors affecting the nature of electricity consumption in a sparsely populated area, namely: weather data (temperature, wind speed); day type (weekday, weekend, holiday); seasonality (winter, spring, summer, autumn). For the software implementation of recurrent LSTM in the MATLAB modeling environment, an algorithm for short-term forecasting of power consumption was developed using the Deep Learning Toolbox library functions. The developed software was tested using ANN architectures with a different number of hidden layers. The analysis of the obtained results made it possible to determine the most accurate predictive model – the recurrent LSTM model containing four hidden layers. It has been experimentally proven to increase the accuracy of short-term power consumption forecasting by considering a variety of changing environmental parameters.
Keywords
short-term forecasting, sparsely populated areas, recurrent neural network, MATLAB programming environment
1. Ob utverzhdenii kompleksnoy gosudarstvennoy programmy Rossiyskoy Federacii «Energosberezhenie i povyshenie energeticheskoy effektivnosti»: Postanovlenie Pravitel'stva Rossiyskoy Federacii ot 09 sentyabrya 2023 goda № 1473. [Approval of the integrated State program of the Russian Federation «Energy Saving and Energy Efficiency Improvement»: Resolution of the Government of the Russian Federation dated September 09, 2023, No. 1473]. Available at: https://economy.gov.ru/material/dokumenty/postanovlenie_pravitelstva_rossiyskoy_federacii_ot_9_sentyabrya_2023_g_1473.html (accessed 27 February 2026). (in Russian)
2. Energeticheskaya strategiya Rossiyskoy Federacii na period do 2050 goda: Rasporyazhenie Pravitel'stva Rossiyskoy Federacii ot 12 aprelya 2025 goda № 908-r. [Energy Strategy of the Russian Federation for the period up to 2050 Resolution of the Government of the Russian Federation dated April 12, 2025, no. 908-r]. Available at:https://minenergo.gov.ru/ministry/energy-strategy (accessed 27 February 2026). (in Russian)
3. Ob obshchih principah organizacii mestnogo samoupravleniya v Rossiyskoy Federacii: Federalnyy zakon ot 06 oktyabrya 2003 goda № 131-FZ. [On the General principles of the Organization of Local Self-Government in the Russian Federation: Federal Law No. 131-FZ of October 06, 2003]. Available at: http://www.kremlin.ru/acts/bank/20035 (accessed 27 February 2026). (in Russian)
4. Semenova N.G., Mitrofanov N.S. Selection of Optimal Parameters for Forecasting Electricity Consumption in the Distribution Network. Energetika: sostoyanie, problemy, perspektivy [Proceedings of the XVI All Russian Scientific and Technical Conference «Power Engineering: State, Problems, Perspectives»]. Orenburg, OSU Publ., 2025, pp. 292–295. (in Russian).
5. Semenova N.G., Mitrofanov N.S. Choosing a Method for Solving Problems of Short and Long Term Electricity Consumption Forecasting. XXI Mezhdunarodnaya konferenciya po elektromekhanike i robototekhnike: sbornik dokladov [Proceedings of the XXI International Conference on Electromechanics and Robotics]. St. Petersburg, SUAI Publ., 2026, pp. 342–346. (in Russian).
6. Vinogradov D.V., Kosenko E.Y. Analysis of Methods for Forecasting Electricity Consumption in Industrial Enterprises. Perspektivnye sistemy i zadachi upravleniya: materialy XVIII Vserossiyskoy nauchno prakticheskoy konferencii i XIV molodezhnoy shkoly seminara «Upravlenie i obrabotka informacii v tekhnicheskih sistemah» [Proceedings of the XVIII All Russian Scientific and Practical Conference and the XIV Youth School Seminar «Promising Management Systems and Tasks: Management and Information Processing in Technical Systems»]. Taganrog, SFU Publ., 2023, pp. 634–639. (in Russian).
7. Isaev Y.N., Arhipova O.V., Kovalev V.Z., Hamitov R.N. Adaptive Short Term Forecasting of Electricity Consumption by Autonomous Power Systems of Small Rural Settlements Based on Correlation Analysis Methods. Izvestiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov [Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering], 2023, vol. 334, no. 2, pp. 224–239. (in Russian). doi: 10.18799/24131830/2023/2/4076
8. Semenova N.G., Vlackaya L.A. Iskusstvennyy intellekt v zadachakh elektroenergetiki [Artificial intelligence in the tasks of the electric power industry]. Orenburg, Orenburg State University Publ., 2025. 234 p. (in Russian)
9. Mamurov B.Zh., Abdullaev Zh.Zh. Regression analysis as a means of studying the dependence between variables. European Science, 2021, no. 2(58), pp. 7–10. (in Russian).
10. Kovalev V.Z., Arkhipova O.V., Shitselov A.V., Shvetsov S.Yu. Research of approaches to short-term forecasting of electricity consumption by small northern settlements as part of the ROETC. Izvestiya Tulskogo gosudarstvennogo universiteta. Tekhnicheskie nauki [Izvestiya Tula State University], 2023, no. 8, pp. 196–204. (in Russian). doi: 10.24412/2071-6168-2023-8-196-197
11. Tavarov S.Sh. Artificial neural network learning algorithm for factor forecasting of household energy consumption. Elektrichestvo [Elektrichestvo], 2022, no. 3, pp. 30–38. (in Russian). doi: 10.24160/0013-5380-2022-3-30-38
12. Reddy G.V., Aitha L.J., Poojitha C., Shreya A.N. Electricity Consumption Prediction Using Machine Learning. E3S Web of Conferences, 2023, no. 391, pp. 1-9. doi: 10.1051/e3sconf/202339101048
13. Rani K, Praveen V., Gupta A., Timmasarti A. Power Consumption Prediction Using Machine Learning. Journal of Machine Learning and Deep Learning (JMLDL), 2024, no. 1 (1), pp. 47-52. doi: 10.64820/AEPJMLDL.11.47.51.122024
14. Frikha M., Taouil K., Fakhfakh A., Derbel F. Predicting Power Consumption Using Deep Learning with Stationary Wavelet. Forecasting, 2024, no. 6(3), pp. 864-884. doi: 10.3390/forecast6030043
15. Serebryakov N.A. Short-term forecasting of hourly power consumption of the guaranteeing supplier with the help of neural network models. Yuzhno-Sibirskiy nauchnyy vestnik [South-Siberian Scientific Bulletin], 2023, no. 5(51), pp. 166–169. (in Russian). doi: 10.25699/SSSB.2023.51.5.023
16. Magasumov M.A., Snytnikov A.V. Forecasting power consumption in the 0.4 kV distribution network section of the Bolshoe Isakovo village. Aktualnye voprosy inzhenernoy otrasli: Sbornik nauchnyh trudov [A collection of scientific papers "Current issues of the engineering industry"]. Kaliningrad, KSTU Publ., 2024, pp. 63–72. (in Russian)
17. Kong Y., Wang Z., Nie Y., Zhou T., Zohren S. Unlocking the Power of LSTM for Long Term Time Series Forecasting. 39th Annual AAAI Conference on Artificial Intelligence (AAAI-25). 2024. Pp. 11968-11976. doi: 10.48550/arXiv. 2408.10006
18. Li Y., Xu J., Anastasiu D. C. An Extreme-Adaptive Time Series Prediction Model Based on Probability-Enhanced LSTM Neural Networks. 37th Annual AAAI Conference on Artificial Intelligence (AAAI-23). 2022. Pp. 8684-8691. doi: 10.48550/arXiv.2211.15891
19. Severiche-Maury Z., Uc-Rios C. E., Arrubla-Hoyos W., Cama-Pinto D. Forecasting Residential Energy Consumption with the Use of Long Short-Term Memory Recurrent Neural Networks. Energies, 2025, no. 18 (5), 1247. doi: 10.3390/ en18051247
20. Rakhmonov I.U., Ushakov V.Ya., Niezov N.N., Kurbonov N.N. Forecasting electricity consumption using neural networks with LSTM. Izvestiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov [Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering], 2023, vol. 334, no. 12, pp. 125–133. (in Russian). doi: 10.18799/24131830/2023/12/4407
21. Lindemann B., Müller T., Vietz H., Jazdi N., Weyrich M. A survey on long short-term memory networks for time series prediction. 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering. ProcediaCIRP, 2021, vol. 99, pp. 650-655. doi:10.1016/j.procir.2021.03.088
22. Semenova N.G., Mitrofanov N.S. Kratkosrochnoe prognozirovanie potrebleniya elektricheskoy energii [Short-term forecasting of electric energy consumption]. Computer program RF, no. 202666591, 2026.
23. POWER| Data Access Viewer. Available at: https://power.larc.nasa.gov/data-access-viewer/ (accessed 13 January 2026).
Semenova N.G. Mitrofanov N.S. Short-Term Forecasting of Electricity Consumption in Sparsely Populated Areas Using a Recurrent Neural Network. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2026, no. 3(72), pp. 34-41. (In Russian). https://doi.org/10.18503/2311-8318-2026-3(72)-34-41