2. Home
  3. ARCHIVE
  4. 2026
  5. VOL 48, NO 2 (2026)
  6. pp. 51-68

Energy AI Software System for Forecasting the Electricity Procurement Portfolio of a Supplier

O. Kliuzko
ORCID: https://orcid.org/0009-0000-3313-0547

Èlektron. model. 2026, 48(2):51-68

ABSTRACT

The Energy AI software system is proposed as an integrated tool to support operational planning and risk management for energy supply companies in the Ukrainian electricity market. The software system combines short-term hourly consumption forecasting (STLF) with subsequent LP/MILP optimization of the procurement portfolio in market segments (RDD/RDN/VDR/BR). The forecasting module provides hourly consumption profiling using the Random Forest method, and the model parameters are tuned using Optuna. The forecast results are used by the Energy AI optimization module to form a procurement portfolio, taking into account limits, product activity masks, discretization steps, and supplier policies. An example of the system's application is given with an assessment of the effectiveness of the supplier's procurement strategy and the impact of imbalances.

Full text: PDF

KEYWORDS

Energy AI, consumption forecasting, portfolio optimization, Random Forest, load-shedding, Optuna, LP/MILP, electricity supplier.

REFERENCES

  1. Blinov, I., & Parus, E. (2022). Wholesale and retail electricity market: calculation work. Textbook for students majoring in 141 "Electric Power Engineering, Electrical Engineering and Electromechanics". Kyiv: Igor Sikorsky Kyiv Polytechnic Institute, 44 p.
  2. Ugbehe, P.O., Diemuodeke, O.E. & Aikhuele, D.O. (2025). Electricity demand forecasting methodologies and applications: a review. Sustainable Energy res. Vol. 12, art 19, pp. 1-32. https://doi.org/10.1186/s40807-025-00149-z
  3. Rangelov, D., Boerger, M., Tcholtchev, N., Lämmel, P. and Hauswirth, M. (2023). Design and Development of a Short-Term Photovoltaic Power Output Forecasting Method Based on RF, Deep Neural Network and LSTM Using Readily Available Weather Features in IEEE Access, vol. 11, pp. 41578-41595, doi: 10.1109/ACCESS.2023.3270714.
  4. Xuan, Y. et al. (2021). Multi-Model Fusion Short-Term Load Forecasting Based on RF Feature Selection and Hybrid Neural Network in IEEE Access, 9, pp. 69002-69009, doi: 10.1109/ACCESS.2021.3051337.
  5. Hu, F.; Feng, X.; Cao, H. (2018). A Short-Term Decision Model for Electricity Retailers: Electricity Procurement and Time-of-Use Pricing. Energies 11, pp. 3258, https://doi>.org/ 10.3390/en11123258
  6. do Prado J.C. and Qiao W. (2019). A Stochastic Decision-Making Model for an Electricity Retailer with Intermittent Renewable Energy and Short-Term Demand Response, in IEEE Transactions on Smart Grid, vol. 10, no. 3, pp. 2581-2592, doi: 10.1109/TSG.2018.2805326
  7. Guo L., Sriyakul T., Nojavan S. and Jermsittiparsert K. (2020). Risk-Based Traded Demand Response Between Consumers’ Aggregator and Retailer Using Downside Risk Constraints Technique, in IEEE Access, vol. 8, pp. 90957-90968 doi: 10.1109/ACCESS.2020.2993868
  8. Golmohamadi H. and Keypour R. (2018). Stochastic optimization for retailers with distributed wind generation considering demand response, in Journal of Modern Power Systems and Clean Energy, vol. 6, no. 4, pp. 733-748, doi: 10.1007/s40565-017-0368-y.
  9. Hongsheng Xu, Jinyu Wen, Qinran Hu, Jiao Shu, Jixiang Lu, Zhihong Yang, (2022). Energy Procurement and Retail Pricing for Electricity Retailers via Deep Reinforcement Learning with Long Short-term Memory, CSEE Journal of Power and Energy Systems, vol. 8, no. 5, pp. 1338-1351. https://ieeexplore.ieee.org/document/9713968
  10. Yuankun Liu, Dongxia Zhang, Hoay Beng Gooi (2021). Data-driven Decision-making Strategies for Electricity Retailers: A Deep Reinforcement Learning Approach, Csee Journal of Power and Energy Systems, vol. 7, no. 2, pp. 358-367. https://ieeexplore.ieee.org/document/9215156
  11. Leandro J. Cesini Silva, Cindy P. Guzman, Marcos J. Rider (2022). Contracting Strategy for Consumers with Distributed Energy Resources in the Liberalized Electricity Market, ACCESS, vol. 10, pp 80437- https://ieeexplore.ieee.org/document/9844706.
  12. Evdokimov, V. & Ivanov, H. (2017). Methods for determining volumes and prices of electricity in contracts in the conditions of a liberalized market. Modeling and Information Technologies, 81, 142-152. http://nbuv.gov.ua/UJRN/Mtit_2017_81_22.
  13. Saukh S., Kliuzko O. (2024). Optimization model for electricity procuremenets and sales portfolio of electrical energy supplier company Electronic modeling –international scientific-theoretical journal ISSN 2616-9525 (online) no. 3. P. 3-21. DOI: https://doi.org/ 15407/emodel.46.03
  14. Saukh S., Kliuzko O. (2024). Forecasting electricity consumption under conditions of rocket and drone attacks on the power system Electronic modeling –international scientific-theoretical journal ISSN 2616-9525 (online) no. 5. P. 87-104. DOI: https://doi.org/10.15407/ emodel.47.05.087