Rüzgar Enerjisi Tahmini için Regresyon Modellerinin Karşılaştırmalı Analizi

Öz

Bu çalışmada, Türkiye'deki bir rüzgar türbininden elde edilen SCADA verilerini kullanarak rüzgar türbini üretim gücünü tahmin etmek için çeşitli regresyon modellerinin performansları karşılaştırılmıştır. Çalışmanın ana amacı, rüzgar hızı, rüzgar yönü, teorik güç eğrisi gibi farklı parametrelerin etkisi altında türbinin aktif güç üretimini doğru bir şekilde tahmin etmek için farklı makine öğrenimi algoritmalarını değerlendirmektir. Bu amaçla, CatBoost, XGB, Extra Trees, Random Forest, Decision Tree, Gradient Boosting, XGBRF, Linear, SVR ve AdaBoost dahil olmak üzere toplam 10 farklı regresyon modeli uygulanmıştır. Modellerin başarısı, R² puanı ve RMSE gibi performans ölçütleri ile değerlendirilmiştir. Sonuçlar, CatBoost Regressor modelinin %98,25 R² puanı ile en yüksek doğruluğu sağladığını, XGB Regressor ve Extra Trees Regressor modellerinin ise sırasıyla %98,14 ve %97,76 R² puanları ile iyi performans gösterdiğini ortaya koymuştur. Ayrıca, haftalık analiz, rüzgar hızı ve teorik güç eğrisinin türbinin aktif güç üretimi üzerinde önemli bir etkiye sahip olduğunu, rüzgar yönünün ise daha düşük bir ilişki gösterdiğini ortaya koydu. Bu bulgular, rüzgar türbinleri için güç tahmininin makine öğrenimi teknikleri kullanılarak daha verimli hale getirilebileceğini ve bu tür modellerin güç üretimi tahmininde önemli bir potansiyele sahip olduğunu göstermektedir.

Anahtar Kelimeler

• Rüzgar Türbini
• Güç Tahmini
• Makine Öğrenmesi
• Regresyon modelleri
• SCADA Verileri

Comparative Analysis of Regression Models for Wind Energy Forecasting

Öz

In this study, the performances of various regression models are compared for predicting the wind turbine generation power using SCADA data from a wind-turbine in Turkey. The main objective of the study is to evaluate different machine learning algorithms to accurately predict the active power generation of the turbine under the influence of different parameters such as wind speed, wind direction, theoretical power curve. For this purpose, a total of 10 different regression models including CatBoost, XGB, Extra Trees, Random Forest, Decision Tree, Gradient Boosting, XGBRF, Linear, SVR and AdaBoost are applied. The success of the models was evaluated by performance metrics such as R² score and RMSE. The results revealed that the CatBoost Regressor model provided the highest accuracy with an R² score of 98.25%, while the XGB Regressor and Extra Trees Regressor models performed well with R² scores of 98.14% and 97.76%, respectively. Furthermore, the weekly analysis revealed that wind speed and the theoretical power curve have a significant impact on the active power generation of the turbine, while wind-direction shows a lower relationship. These findings show that power forecasting for wind-turbines can be made more efficient using machine learning techniques and that such models have significant potential in power generation forecasting.

Anahtar Kelimeler

• Wind Turbine
• Power Estimation
• Machine Learning
• Regression Models
• SCADA Data

Kaynakça

1. Abdullah, B. U. D., Khanday, S. A., Islam, N. U., Lata, S., Fatima, H., & Nengroo, S. H. (2024). Comparative analysis using multiple regression models for forecasting photovoltaic power generation. Energies, 17(7), 1564.
2. Alpsalaz, F., Özüpak, Y., Aslan, E., & Uzel, H. (2025). Classification of maize leaf diseases with deep learning: Performance evaluation of the proposed model and use of explicable artificial intelligence. Chemometrics and Intelligent Laboratory Systems, 262, 105412.
3. Altork, Y. (2025). Comparative Analysis of Machine Learning Models for Wind Speed Forecasting: Support Vector Machines, Fine Tree, and Linear Regression Approaches. International Journal of Thermofluids, 101217.
4. Aslan, E. (2025). Temperature prediction and performance comparison of permanent magnet synchronous motors using different machine learning techniques for early failure detection. Eksploatacja i Niezawodność – Maintenance and Reliability, 27(1).
5. Aslan, E., Özüpak, Y., & Alpsalaz, F. (2025). Boiler efficiency and performance optimization in district heating and cooling systems with machine learning models. Journal of the Chinese Institute of Engineers, Series A.
6. Ayele, S. T., Ageze, M. B., Zeleke, M. A., & Miliket, T. A. (2023). Adama II wind farm long-term power generation forecasting based on machine learning models. Scientific African, 21, e01831.
7. Cakiroglu, C., Demir, S., Ozdemir, M. H., Aylak, B. L., Sariisik, G., & Abualigah, L. (2024). Data-driven interpretable ensemble learning methods for the prediction of wind turbine power incorporating SHAP analysis. Expert Systems with Applications, 237, 121464.
8. Çınar, M., Aslan, E., & Özüpak, Y. (2025). Comparison and optimization of machine learning methods for fault detection in district heating and cooling systems. Bulletin of the Polish Academy of Sciences Technical Sciences, 73(3), 1–9.

9. Chen, X., Zhang, X., Dong, M., Huang, L., Guo, Y., & He, S. (2021). Deep learning-based prediction of wind power for multi-turbines in a wind farm. Frontiers in Energy Research, 9, 723775.
10. Chaka, M. D., et al. (2024). Improving wind speed forecasting at Adama wind farm II in Ethiopia through deep learning algorithms. Case Studies in Chemical and Environmental Engineering, 9, 100594.
11. Deng, J., et al. (2024). Wind turbine short-term power forecasting method based on hybrid probabilistic neural network. Energy, 134042.
12. Kaggle. (2025, April 5). Wind turbine SCADA dataset. Retrieved from https://www.kaggle.com/datasets/berkerisen/wind-turbine-scada-dataset
13. Lin, W. H., Wang, P., Chao, K. M., Lin, H. C., Yang, Z. Y., & Lai, Y. H. (2021). Wind power forecasting with deep learning networks: Time-series forecasting. Applied Sciences, 11(21), 10335.
14. Liu, Z. H., Wang, C. T., Wei, H. L., Zeng, B., Li, M., & Song, X. P. (2024). A wavelet-LSTM model for short-term wind power forecasting using wind farm SCADA data. Expert Systems with Applications, 247, 123237.
15. Mostafa, K., Zisis, I., & Moustafa, M. A. (2022). Machine learning techniques in structural wind engineering: A state-of-the-art review. Applied Sciences, 12(10), 5232.
16. Nakhchi, M. E., Win Naung, S., & Rahmati, M. (2023). Wake and power prediction of horizontal-axis wind farm under yaw-controlled conditions with machine learning. Energy Conversion and Management, 296, 117708.
17. Oyucu, S., & Aksöz, A. (2024). Integrating machine learning and MLOps for wind energy forecasting: a comparative analysis and optimization study on Türkiye’s wind data. Applied Sciences, 14(9), 3725.
18. Özüpak, Y., Alpsalaz, F., & Aslan, E. (2025). Air quality forecasting using machine learning: Comparative analysis and ensemble strategies for enhanced prediction. Water, Air, & Soil Pollution, 236(7), 1–17.
19. Özüpak, Y., & Aslan, E. (2024). Using artificial neural networks to improve the efficiency of transformers used in wireless power transmission systems for different coil positions. Revue Roumaine des Sciences Techniques – Série Électrotechnique et Énergétique, 69(2), 195–200.
20. Özüpak, Y., & Mansurov, S. (2025). Optimizing electricity demand forecasting with a novel RNN-LSTM hybrid model. Energy Sources, Part B: Economics, Planning and Policy, 20(1).
21. Pacal, I., & Kunduracioglu, I. (2025). ASMM-Net: A hybrid CNN with adaptive scale modulation for robust ischemic stroke lesion segmentation. Iranian Journal of Science and Technology, Transactions of Electrical Engineering.
22. Peng, X., Yang, Z., Li, Y., Wang, B., & Che, J. (2023). Short-term wind power prediction based on stacked denoised auto-encoder deep learning and multi-level transfer learning. Wind Energy, 26(10), 1066–1081.
23. Qu, Z., Li, J., Hou, X., & Gui, J. (2023). A D-stacking dual-fusion, spatio-temporal graph deep neural network based on a multi-integrated overlay for short-term wind-farm cluster power multi-step prediction. Energy, 281, 128289.
24. Ramadevi, B., Kasi, V. R., & Bingi, K. (2024). Enhancement of Texas wind turbine power predictions using fractional order neural network by incorporating machine learning models to impute missing data. Knowledge-Based Systems, 300, 112176.
25. Su, M., Sun, B., & He, J. (2024). Wind power prediction through acoustic data-driven online modeling and active wake control. Energy Conversion and Management, 319, 118920.
26. Sun, Y., Zhou, Q., Sun, L., Sun, L., Kang, J., & Li, H. (2024). CNN–LSTM–AM: A power prediction model for offshore wind turbines. Ocean Engineering, 301, 117598.
27. Tarek, Z., et al. (2022). Wind power prediction based on machine learning and deep learning models. Computers, Materials & Continua, 74(1), 715–732.
28. Ti, Z., Deng, X. W., & Zhang, M. (2021). Artificial neural networks based wake model for power prediction of wind farm. Renewable Energy, 172, 618–631.
29. Wang, Q., Hu, J., Yang, S., Dong, Z., Deng, X., & Xu, Y. (2025). Towards machine learning applications for structural load and power assessment of wind turbine: An engineering perspective. Energy Conversion and Management, 324, 119275.
30. Yegin, A. E., & Karcioglu, A. A. (2025). Dam water levels prediction using advanced hybrid deep learning model based on Bayesian Optimization approach. Egyptian Informatics Journal, 31, 100760.
31. Yilmaz, M. T., Algul, E., & Pacal, I. (2025). A comparative study of advanced deep learning architectures for breast cancer classification on ultrasound and histological images. Results in Engineering, 28, 107600.
32. Zehtabiyan-Rezaie, N., Iosifidis, A., & Abkar, M. (2023). Physics-guided machine learning for wind-farm power prediction: Toward interpretability and generalizability. PRX Energy, 2(1), 013009.
33. Zhang, J., & Zhao, X. (2023). Digital twin of wind farms via physics-informed deep learning. Energy Conversion and Management, 293, 117507.
34. Zhang, L., et al. (2024). Research on wind turbine icing prediction data processing and accuracy of machine learning algorithm. Renewable Energy, 237, 121566.
35. Zhou, H., Qiu, Y., Feng, Y., & Liu, J. (2022). Power prediction of wind turbine in the wake using hybrid physical process and machine learning models. Renewable Energy, 198, 568–586.