TITLE
PREDICTING WIND ENERGY PRODUCTION IN THE SHORT-TERM USING MACHINE LEARNING ALGORITHM
AUTHOR(S)
Hilmi KUŞÇU, Taşkın TEZ
ABSTRACT
The prediction of electricity generation from wind power plays a critical role in the formulation and management of future energy production plans. These predictions are highly important for wind energy facilities to achieve optimal performance, meet energy demands, and stabilize energy prices. Therefore, in this study, the Support Vector Machine (SVM) Regression Algorithm, a traditional machine learning algorithm, was preferred to forecast the weekly electricity production of wind power plants. Performance metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and R-Squared (R²) were utilized to assess the accuracy of the predictions. The results of this study indicated that the SVM Regression Algorithm with the Radial function yielded the best outcomes. Consequently, it is recommended to employ the SVM Regression Algorithm with the Radial function for weekly electricity production predictions.
DOI
www.doi.org/10.70456/QNJJ9806
PAGES
23-27
DOWNLOAD
https://unitechsp.tugab.bg/images/2023/1-EE/s1_p9_v6.pdf
How to cite this article:
Hilmi KUŞÇU, Taşkın TEZ, PREDICTING WIND ENERGY PRODUCTION IN THE SHORT-TERM USING MACHINE LEARNING ALGORITHM, UNITECH – SELECTED PAPERS - 2024, 23-27