Keywords
aggregators
local electricity markets
neural networks
virtual power plants
How to Cite
Abstract
This study presents a comparative evaluation of three approaches to short-term net-load forecasting for virtual aggregations of heterogeneous energy facilities, comprising both building loads and photovoltaic units. The objective is to assess how different levels of data aggregation influence the forecasting accuracy of total consumption, PV generation, and the resulting net load. All experiments were conducted on the UCSD Microgrid dataset using an identical Transformer-based model, which allowed isolating the effect of data structure rather than model architecture. The results show that disaggregated bottom-up modelling yields the highest accuracy for building consumption, whereas partial aggregation—separate forecasting of total load and PV generation—provides the best performance for net-load prediction. PV generation remains the most challenging component due to structural asymmetry and extended zero-generation intervals; in this case, aggregated forecasts demonstrate lower energy error compared to bottom-up models. The findings provide practical guidance for selecting appropriate aggregation schemes and forecasting strategies in virtual power plants and emerging local energy markets. Ref. 14, fig. 4.
References
1.Kyrylenko O., Denysiuk S., Blinov I. Digital transformation of the energy industry: current trends and task. Pratsi Instytutu Elektrodynamiky NAN Ukrainy. 2022. V. 65. Pp. 5–14. (Ukr) DOI: https://doi.org/10.15407/publishing2022.65.005
2.Blinov I.V., Parus Ye.V., Shymaniuk P.V., Vorushylo A.O.. 2024. Optimization model of microgrid functioning with solar power plant and energy storage system. Tekhnichna Elektrodynamika. 2024. No 5. Pp, 069. DOI: https://doi.org/10.15407/techned2024.05.069
3.Parus Ye.V., Blinov I.V. Optimization of the use of available energy resources of the microgrid under the condition of supporting readiness for isolated mode. Tekhnichna Elektrodynamika. 2025. No 5. Pp. 56–69. DOI: https://doi.org/10.15407/techned2025.05.056
4.Belokha H.S., Sychova V.V.Optimization of electric load schedules by an aggregator in local power systems. Pratsi Instytutu Elektrodynamiky NAN Ukrainy, 2023. No 66. Pp. 84–89. DOI: https://doi.org/10.15407/publishing2023.66.084
5.Blinov I., Radziukynas V., Shymaniuk P., Dyczko A., Stecuła K., Sychova V., Miroshnyk V., & Dychkovskyi R. Smart management of energy losses in distribution networks using deep neural networks. Energie., 2025, No 18. Pp. 3156. DOI: https://doi.org/10.3390/en18123156
6.Chernenko P.O., Martyniuk O.V., & Miroshnyk V.O. Modelling and short-term forecasting of the technological component of the electrical load of a regional power system. Tekhnichna Elektrodynamika. 2016. No 4. Pp. 68–70. DOI: https://doi.org/10.15407/techned2016.04.068
7.Peñaloza A. A., Leborgne R. C., & Balbinot A. Comparative analysis of residential load forecasting with different levels of aggregation. Engineering Proceedings. 2022., No 18(1). Pp. 29. DOI: https://doi.org/10.3390/engproc2022018029
8.Beichter M., Phipps K., Frysztacki M. M., Mikut R., Hagenmeyer V., & Ludwig N. Net load forecasting using different aggregation levels. Energy Informatics. 2022, No 5(S1). Pp. 19. DOI: https://doi.org/10.1186/s42162-022-00213-8
9.Peng Y., Wang Y., Lu X., Li H., Shi D., Wang Z., & Li J. Short-term load forecasting at different aggregation levels with predictability analysis. IEEE Innovative Smart Grid Technologies – Asia (ISGT Asia) 2019. Pp. 3385–3390. DOI: https://doi.org/10.1109/ISGT-Asia.2019.8881343
10.Sevlian R., & Rajagopal R. Short term electricity load forecasting on varying levels of aggregation. arXiv preprint. 2017. arXiv:1404.0058. DOI: https://doi.org/10.1016/j.ijepes.2017.10.032
11.Anand H., Nateghi R., & Alemazkoor N. Bottom-up forecasting: Applications and limitations in load forecasting using smart-meter data. Data-Centric Engineering, 2023. No 4. Pp. e14. DOI: https://doi.org/10.1017/dce.2023.10
12.Bandyopadhyay S., Ganu T., Khadilkar H., & Arya V. Individual and aggregate electrical load forecasting: One for all and all for one. Proceedings of the 6th International Conference on Future Energy Systems (e-Energy ’15. Pp. 257–267 DOI: https://doi.org/10.1145/2768510.2768539
13.Silwal S., Mullican C., Chen Y.A., Ghosh A., Dilliott J., and Kleissl J. Open-source multi-year power generation, consumption, and storage data in a microgrid, Journal of Renewable and Sustainable Energy. 2021. Vol. 13. No. 2. Pp. 025301. DOI: https://doi.org/10.1063/5.0038650.
14.Vaswani A. et al. Attention is all you need, I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, Eds., Curran Associates, Inc., In Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS'17). Curran Associates Inc., Red Hook, NY, USA, 6000–6010.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Copyright (c) 2025 V. Miroshnyk, V. Sychova