Abstract
The application of machine learning in virtual electric power plants (VES) is an important area of research, thanks to which it is possible to increase the energy facilities reliability and efficiency. Wind farms are complex dynamic structures that require constant monitoring and adaptive management to maintain stability and optimize performance. Within the framework of this work, a solution has been proposed that can determine the required amount of control actions to be transmitted to the power system in response to the received data in order to prevent an emergency situation at the power facility. The research methods used were theoretical analysis of scientific literature on machine learning and power system management, data analysis and preprocessing, including descriptive statistics methods, data visualization and the Boruta algorithm for selecting significant features. The solution to the problem is demonstrated using a modified IEEE 39-Bus test model with the addition of wind turbines, where the task of determining the volume of control actions is formalized as a multiclass classification task. The following machine learning algorithms were used to solve it: CatBoost, random forest, extreme gradient boosting and neural networks based on PyTorch. These algorithms make it possible to predict changes in the energy network based on historical data, which can help in predicting loads and optimizing resource use. The best results were obtained using neural networks, which provided an average accuracy of 98% in determining the required amount of control actions. This indicates the high efficiency of the applied machine learning algorithms in energy systems. The results obtained demonstrate the practical applicability of the method to increase the efficiency of emergency management in stochastic energy generation conditions. This study highlights the importance of machine learning in the energy sector and opens up opportunities for further development of intelligent grid management systems.
Keywords
power system, emergency automation, control actions, mathematical modeling, machine learning, virtual power plants
1. Isaev E.V., Kats P.Ya., Lisitsin A.A., Nikolaev A.V., Ten E.A. Estimator of static stability and control action selection in terms of static stability provision in post-emergency conditions. Isvestiya NTTS Edinoy energeticheskoy sistemy [Proceedings of the Power Grid Research and Development Centre]. 2013, no. 1(68), pp. 48-57. (In Russian)
2. Pavlushko S.A., Zhurkov A.V., Satsuk E.I., Kats P.Ya., Lisitsin A.A. Control action calculation in terms of statistic stability in the centralized system software of the new generation emergency control system. Elekricheskie stantsii [Electrical stations]. 2015, no. 2, pp. 35–40. (In Russian)
3. Lisitsin A.A., Chaplyuk S.V. Automatic Determination of TsSPA starting units. Isvestiya NTTS Edinoy energeticheskoy sistemy [Proceedings of the Power Grid Research and Development Centre]. 2018, no. 2, pp. 6–11. (In Russian)
4. Landman A.K., Pertushkov M.V., Petrov A.E., Sakaev O.O., Subbotin-Chukalskiy A.V. Approaches to fault tolerant distributed control system for emergency control. Nauchnye problem transporta Sibiri i Dalnego Vostoka [Scientific issues of Siberia and Russian Far East Transportation]. 2014, no. 4, pp. 335–340. (In Russian)
5. Landman A.K., Petrov A.M., Petrov A.E., Sakaev O.O. Approaches to development of emergency control systems based on universal hardware and software complexes of emergency automation. Novoe v rossiyskoy elektroenergetike [New issues in Russian electric power industry]. 2012, no. 7, pp. 29-39. (In Russian)
6. Fishov A.G., Kardgaubaev N.A. Decentralized multiagent voltage control in electric power lines. Vestnik Irkutskogo gosudarstvennogo tekhnicheskogo universiteta. [Proceedings of Irkutsk State Technical University]. 2018, vol. 22, no. 6(137), pp. 183–195. (In Russian). doi: 10.21285/1814-3520-2018-6-183-195
7. Aprosin K.I., Khokhrin A.A., Ivanov Yu.V. Assessment of control action for instability prevention automation based on time-phased vector measurements. Releyshchik. [Protection Engineer]. 2021, no. 3(41), pp. 26. (In Russian)
8. Zhu Q., Dang J., Chen J. A method for power system transient stability assessment based on deep belief networks. Proc. CSEE. 2018, vol. 38, pp. 735-743. doi: 10.3390/en17235864
9. Azman S.K., Isbeih Y.J., Moursi M.S.E. A Unified Online Deep Learning Prediction Model for Small Signal and Transient Stability. IEEE Transactions on Power Systems. 2020, vol. 35, no. 6, pp. 4585-4598. doi: 10.1109/TPWRS.2020.2999102
10. Liu X., Liu J., Zhao Y. A Deep Reinforcement Learning Framework for Automatic Operation Control of Power System Considering Extreme Weather Events. IEEE Power & Energy Society General Meeting (PESGM). IEEE, 2022. doi: 10.1109/PESGM48719.2022.9916843
11. Xia Y., Xu Y., Wang Y. A Safe Policy Learning-Based Method for Decentralized and Economic Frequency Control in Isolated Networked-Microgrid Systems. IEEE Transactions on Sustainable Energy. 2022, vol. 13, no. 4, pp. 1982-1993. doi: 10.1109/TSTE.2022.3178415
12. Zhang Q., Dehghanpour K., Wang Z. A Learning-Based Power Management Method for Networked Microgrids Under Incomplete Information. IEEE Transactions on Smart Grid. 2020, vol. 11, no. 2, pp. 1193-1204. doi: 10.1109/TSG.2019.2933502
13. Zhou T., Wang Y., Xu Y. Applications of Reinforcement Learning in Frequency Regulation Control of New Power Systems. International Conference on Cyber-Physical Social Intelligence (ICCSI). IEEE, 2022, pp. 501-506. doi: 10.1109/ICCSI55536.2022.9970560
14. MacDougall P., Kosek A.M., Bindner H. Applying machine learning techniques for forecasting flexibility of virtual power plants. IEEE Electrical Power and Energy Conference (EPEC). IEEE, 2016. doi: 10.1109/EPEC.2016.7771738
15. Sierla S., Pourakbari-Kasmaei M., Vyatkin V. A taxonomy of machine learning applications for virtual power plants and home/building energy management systems. Automation in Construction. 2022, vol. 136, 104174. doi: 10.1016/j.autcon. 2022.104174
16. Li X., Luo F., Li C. Multi-agent deep reinforcement learning-based autonomous decision-making framework for community virtual power plants. Applied Energy. 2024, vol. 360, 122813. doi: 10.1016/j.apenergy.2024.122813
17. Athay T., Podmore R., Virmani S. A Practical Method for the Direct Analysis of Transient Stability. IEEE Transactions on Power Apparatus and Systems. 1979, vol. PAS-98, no. 2, pp. 573-584.doi: 10.1109/tpas.1979.319407
18. Pai A. Energy Function Analysis for Power System Stability. Springer, 1989.doi: 10.1007/978-1-4613-1635-0
19. Kursa M.B., Rudnicki W.R. Feature selection with the Boruta package. Journal of Statistical Software. 2010, vol. 36, pp. 1–13. doi: 10.18637/jss.v036.i11
20. Prokhorenkova L., Gusev G., Vorobev A. CatBoost: unbiased boosting with categorical features. Advances in Neural Information Processing Systems(NeurIPS). 2018, vol. 31.
21. Chen T., Guestrin C. XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016, pp. 785-794. doi: 10.1145/2939672.29397
22. Kurilenko I.E., Nikonov I.E. Classification of messages in voice interaction systems. Vestnik MEI [Bulletin of MEI], 2020, no. 5, pp. 132. (In Russian). doi: 10.24160/1993-6982-2020-5-132-139
23. Mohammed A., Kora R. A comprehensive review on ensemble deep learning: Opportunities and challenges. Journal of King Saud University - Computer and Information Sciences. 2023, vol. 35, no. 2, pp. 757-774. doi: 10.1016/j.jksuci. 2023.01.014
24. Gil-González W., Montoya O.D., Escobar-Mejía A. LQR-Based Adaptive Virtual Inertia for Grid Integration of Wind Energy Conversion System Based on Synchronverter Model. Electronics. 2021, no. 10(9), 1022. doi: 10.3390/electronics10091022
25. Senyuk M., Beryozkina S., Zicmane I. Bulk Low-Inertia Power Systems Adaptive Fault Type Classification Method Based on Machine Learning and Phasor Measurement Units Data. Mathematics. 2025, no. 13(2), 316. doi: 10.3390/math13020316
Bezzubov A.A., Senyuk M.D., Aksyonov K.A. Using Machine Learning to Identify the Control Effects of Virtual Power Plants. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2025, no. 4(69), pp. 79-87. (In Russian). https://doi.org/10.18503/2311-8318-2025-4(69)-79-87