Forecasting algorithm of renewable energy generation facility production

The article considers the solution to the problem of determining the power generation plan for the generating facility, which is a wind power plant. Wind power, as a key renewable energy source, faces high generation uncertainty due to variability of meteorological conditions. In order to solve the task of optimizing the power generation plan at wind power plants, this work applies machine-learning methods that allow for analyzing large amounts of data obtained from various sensors and meteorological stations. The use of machine learning models helps to accurately predict energy production, which in turn allows to optimize the operation of wind power plant, including by adjusting the modes of maximizing the installed capacity utilization factor (ICUF). Several mathematical models are considered – k-nearest neighbor model, decision tree model, random forest model and gradient boosting. These models are chosen because of their algorithmic simplicity, their learning is relatively fast, and also because of their independence from data type. As a result of the analysis of the data obtained from each model, the gradient-boosting model is chosen - the highest coefficient of determination on validation data is obtained for the shortest time of data processing. Also created a virtual interface for easier data entry and visualization of results.

1. Захожий К.А. Возобновляемые источники энергии. Colloquium-journal. 2020;28(80):57-58. https://doi.org/10.24412/2520-2480-2020-2880-57-58

2. Chen X., Zhang X., Dong M., Huang L., Guo Y., He S. Deep learning-based prediction of wind power for multi-turbines in a wind farm. Frontiers in Energy research. 2021;9:723775. https://doi.org/10.3389/fenrg.2021.723775

3. Song D., Zheng S., Yang S., Yang J., Dong M., Su M., et al. Annual energy production estimation for variable-speed wind turbine at high-altitude site. Journal of Modern Power Systems and Clean Energy. 2021;9(3):684-687. https://doi.org/10.35833/MPCE.2019.000240

4. Tukey J.W. Exploratory data analysis. Addison-Wesley Publishing Company, 1977. P. 688. Available at: https://archive.org/details/exploratorydataa0000tuke_7616/page/n3/mode/2up (accessed: 02.09.2025).

5. Флах П. Машинное обучение. Наука и искусство построения алгоритмов, которые извлекают знания из данных. Москва: ДМК Пресс, 2015. 400 с. Режим доступа: https://rusneb.ru/catalog/000199_000009_008646425/?ysclid=mli2b67snu451963270 (дата обращения: 02.09.2025).