Abstract

Electricity load forecasting is a critical task for the efficient management of power systems and the optimization of electricity generation and distribution. This study proposes a hybrid ARIMA-LSTM-Random Forest model for intelligent electricity load forecasting. The model combines the strengths of classical statistical methods, such as ARIMA, with the deep learning capabilities of LSTM to capture nonlinear temporal dependencies, and Random Forest to enhance prediction accuracy by addressing residual errors. The proposed approach was tested on real electricity consumption data and demonstrated superior accuracy compared to individual ARIMA, LSTM, or Random Forest models. The results highlight the effectiveness of the hybrid method in reducing mean squared error and improving forecast reliability. The findings can be applied to load planning, optimization of power system operations, and integration of renewable energy sources into energy networks.

Keywords: Electricity load, forecasting, hybrid model, ARIMA, LSTM, Random Forest, intelligent methods

References

1. Ye, H., Yang, B., Han, Y., & Chen, N. (2022). State-Of-The-Art Solar Energy Forecasting Approaches: Critical Potentials and Challenges. Frontiers in Energy Research, 10, 268. https://doi.org/10.3389/fenrg.2022.875790
2. Kashtan, V., Hnatushenko, V. (2023). Deep Learning Technology for Automatic Burned Area Extraction Using Satellite High Spatial Resolution Images. In: Babichev, S., Lytvynenko, V. (eds) Lecture Notes in Data Engineering, Computational Intelligence, and Decision Making. ISDMCI 2022. Lecture Notes on Data Engineering and Communications Technologies, vol 149. Springer, Cham. https://doi.org/10.1007/978-3-031-16203-9_37
3. Kashtan, V., Hnatushenko, V. (2023). Automated building damage detection on digital imagery using machine learning. Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu, № 6, 134-140. https://doi.org/10.33271/nvngu/2023-6/134
4. Kyrylenko, O., Shcherba, A., Kucheriava, I. (2021). Intellectual technologies for monitoring of technical state of up-to-date high-voltage cable power lines. Tekhnichna Elektrodynamika, 2021, 29-40. https://doi.org/10.15407/techned2021.06.029
5. Psyuk, V., & Polyanska, A. (2024). The usage of artificial intelligence in the activities of mining enterprises. E3S Web of Conferences, 526, 01016. https://doi.org/10.1051/e3sconf/202452601016
6. Pavlychenko, A., Sala, D., Pyzalski, M., Dybrin, S., Antoniuk, O., & Dychkovskyi, R. (2025). Utilizing Fuel and Energy Sector Waste as Thermal Insulation Materials for Technical Buildings. Energies, 18(9), 2339. https://doi.org/10.3390/en18092339
7. Lewicka, D. (2020). Employee institutional trust as an antecedent of diverse dimensions of organisational commitment. Argumenta Oeconomica, 2019(1), 321–340. https://doi.org/10.15611/aoe.2020.1.13
8. Sytnyk, R., Hnatushenko, V., Hnatushenko, V. (2022). Decentralized Information System for Supply Chain Management Using Blockchain. IntelITSIS'2022: 3nd International Workshop on Intelligent Information Technologies and Systems of Information Security, Khmelnytskyi, Ukraine. http://ceur-ws.org/Vol-3156/paper45.pdf
9. Smyl, S. (2020). A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting. International Journal of Forecasting, 36, 75–85. https://doi.org/10.1016/j.ijforecast.2019.03.017
10. Mohammed, N. A., Al-Bazi, A. (2022). An adaptive backpropagation algorithm for long-term electricity load forecasting. Neural Computing and Applications, 34, 477–491. https://doi.org/10.1007/s00521-021-06384-x
11. Pavićević, M., Popović, T. (2022). Forecasting day-ahead electricity metrics with artificial neural networks. Sensors, 22, 1051. https://doi.org/10.3390/s22031051
12. Qiao, W. (2018). Intelligent pre-estimation method study on short-period power load. Jiangsu Electrical Journal, 13–17. https://doi.org/10.3390/en13226105
13. Mehra, A., Spreeuwers, L., & Strisciuglio, N. (2021). Deepfake Detection using Capsule Networks and Long Short-Term Memory Networks. Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 407–414. https://doi.org/10.5220/0010289004070414
14. Dychkovskyi, R., Falshtynskyi, V., Ruskykh, V., Cabana, E., & Kosobokov, O. (2018). A modern vision of simulation modelling in mining and near mining activity. E3S Web of Conferences, 60, 00014. https://doi.org/10.1051/e3sconf/20186000014
15. Graw, J. H., Wood, W. T., & Phrampus, B. J. (2021). Predicting Global Marine Sediment Density Using the Random Forest Regressor Machine Learning Algorithm. Journal of Geophysical Research: Solid Earth, 126(1). https://doi.org/10.1029/2020jb020135
16. Kholodniak, Y., Havrylenko, Y., Halko, S., Hnatushenko, V., Suprun, O., Volina, T., Shchur, T. (2023). Improvement of the algorithm for setting the characteristics of interpolation monotone curve. Informatyka, Automatyka, Pomiary W Gospodarce I Ochronie Środowiska, 13(4), 44–50. https://doi.org/10.35784/iapgos.5392
17. Dongyan, F., Hai, S., Jun, Y., Kai, Z., Xia, Y., Zhixue, S. (2021). Well production forecasting based on ARIMA-LSTM model considering manual operations. Energy, Volume 220, 119708. https://doi.org/10.1016/j.energy.2020.119708
18. Kasemset, C., Phuruan, K.t, Opassuwan, T. (2023). Shallot Price Forecasting Models: Comparison among Various Techniques. Production Engineering Archives, 29, 348-355. https://doi.org/10.30657/pea.2023.29.40