Wind Speed Prediction in Bingol Region: A Deep Learning and Stacking Ensemble Approach

Abstract

This study aims to develop a novel model for wind speed prediction by integrating advanced deep learning techniques with ensemble methods using wind speed data collected from various districts of the Bingol region. The methodology includes rigorous data preprocessing, time‐based feature engineering, STL decomposition, and standardization – all mathematically modeled. A hybrid deep learning model comprising Conv1D, LSTM, and attention mechanisms is implemented alongside a stacking ensemble approach that integrates predictions from Ridge, Random Forest, XGBoost, LightGBM, CatBoost, SVR, and MLP regressors. Model performance is evaluated using RMSE, MAE, R², and EVS, with each district’s data supported by specific mathematical analyses.

Keywords

• Wind speed
• Deep learning
• LSTM
• STL decomposition

Bingöl Bölgesinde Rüzgar Hızı Tahmini: Derin Öğrenme ve Katmanlı Bütünleşik Yaklaşım

Öz

Bu çalışma, Bingöl bölgesinin çeşitli ilçelerinden toplanan rüzgar hızı verilerini kullanarak, ileri düzey derin öğrenme tekniklerini ensemble yöntemleriyle entegre eden yenilikçi bir rüzgar hızı tahmin modeli geliştirmeyi amaçlamaktadır. Metodoloji, titiz veri ön işleme, zamana dayalı özellik mühendisliği, STL ayrıştırması ve standardizasyonu içermekte olup, tümü matematiksel olarak modellenmiştir. Conv1D, LSTM ve dikkat mekanizmalarını içeren hibrit bir derin öğrenme modeli, Ridge, Random Forest, XGBoost, LightGBM, CatBoost, SVR ve MLP regresörlerinden elde edilen tahminleri birleştiren bir stacking ensemble yaklaşımıyla birlikte uygulanmıştır. Model performansı, RMSE, MAE, R² ve EVS kullanılarak değerlendirilmiş ve her ilçenin verileri belirli matematiksel analizlerle desteklenmiştir.

Anahtar Kelimeler

• Rüzgar Hızı
• Derin öğrenme
• UKSB
• STL dönüşümü

Kaynakça

1. Polat S, Alpaslan N, Hallaç İR. Wind Speed Prediction Using Meteorological Measurements for Elazığ Province. Computer Science 2023; Vol:8: 110–120.
2. Arabi S, Asgarimehr M, Kada M, et al. Hybrid CNN-LSTM Deep Learning for Track-Wise GNSS-R Ocean Wind Speed Retrieval. Remote Sensing 2023, Vol 15, Page 4169 2023; 15: 4169.
3. Chen G, Li L, Zhang Z, et al. Short-term wind speed forecasting with principle-subordinate predictor based on Conv-LSTM and improved BPNN. IEEE Access 2020; 8: 67955–67973.
4. Li X, Li K, Shen S, et al. Exploring Time Series Models for Wind Speed Forecasting: A Comparative Analysis. Energies 2023, Vol 16, Page 7785 2023; 16: 7785.
5. Zuluaga CD, Álvarez MA, Giraldo E. Short-term wind speed prediction based on robust Kalman filtering: An expe rimental comparison. Applied Energy 2015; 156: 321–330.
6. Lin KP, Pai PF, Ting YJ. Deep belief networks with genetic algorithms in forecasting wind speed. IEEE Access 2019; 7: 99244–99253.
7. Zhu Q, Chen J, Zhu L, et al. Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach. Energies 2018, Vol 11, Page 705 2018; 11: 705.
8. Chen M-Y, Liou Y-F, Chien H. Applications of deep-learning on TRITON data: results and findings. https://doi.org/101117/123042757 2025; 13268: 6–8.

9. Chen X, Wang Y, Zhang H, et al. A novel hybrid forecasting model with feature selection and deep learning for wind speed research. Journal of Forecasting 2024; 43: 1682–1705.
10. Feng L, Wang Y, Yan Y, et al. Wind Speed Prediction Model Based on Deep Learning. E3S Web of Conferences 2023; 466: 01011.
11. Shi X, Lei X, Huang Q, et al. Hourly Day-Ahead Wind Power Prediction Using the Hybrid Model of Variational Model Decomposition and Long Short-Term Memory. Energies 2018, Vol 11, Page 3227 2018; 11: 3227.
12. Hossain MA, Gray EMA, Islam MR, et al. Forecasting very short-term wind power generation using deep learning, optimization and data decomposition techniques. ICEMS 2021 - 2021 24th International Conference on Electrical Machines and Systems 2021; 323–327.
13. Imani M, Fakour H, Lan WH, et al. Application of Rough and Fuzzy Set Theory for Prediction of Stochastic Wind Speed Data Using Long Short-Term Memory. Atmosphere 2021, Vol 12, Page 924 2021; 12: 924.
14. Dolatabadi A, Abdeltawab H, Mohamed YARI. Hybrid Deep Learning-Based Model for Wind Speed Forecasting Based on DWPT and Bidirectional LSTM Network. IEEE Access 2020; 8: 229219–229232.
15. Li Y, Chen X, Li C, et al. A Hybrid Deep Interval Prediction Model for Wind Speed Forecasting. IEEE Access 2021; 9: 7323–7335.
16. Di Piazza A, Di Piazza MC, Vitale G. Estimation and Forecast of Wind Power Generation by FTDNN and NARX-net based models for Energy Management Purpose in Smart Grids. RE&PQJ 2014; 12: 995–1000.
17. Lawal A, Rehman S, Alhems LM, et al. Wind Speed Prediction Using Hybrid 1D CNN and BLSTM Network. IEEE Access 2021; 9: 156672–156679.
18. Li Y, Chen X, Li C, et al. A Hybrid Deep Interval Prediction Model for Wind Speed Forecasting. IEEE Access 2021; 9: 7323–7335.
19. Mohandes M, Rehman S, Nuha H, et al. Accuracy of wind speed predictability with heights using Recurrent Neural networks. FME Transactions 2021; 49: 908–918.
20. Tuerxun W, Xu C, Guo H, et al. An ultra-short-term wind speed prediction model using LSTM based on modified tuna swarm optimization and successive variational mode decomposition. Energy Science & Engineering 2022; 10: 3001–3022.
21. Cheng L, Zang H, Ding T, et al. Ensemble Recurrent Neural Network Based Probabilistic Wind Speed Forecasting Approach. Energies 2018, Vol 11, Page 1958 2018; 11: 1958.
22. Xu X, Ma S, Huang C. Data denoising and deep learning prediction for the wind speed based on NOA optimization. Epub ahead of print 2 August 2024. DOI: 10.21203/RS.3.RS-4699260/V1.
23. Shirzadi N, Nasiri F, Menon RP, et al. Smart Urban Wind Power Forecasting: Integrating Weibull Distribution, Recurrent Neural Networks, and Numerical Weather Prediction. Energies 2023, Vol 16, Page 6208 2023; 16: 6208.
24. García-Puente B, Rodríguez-Hurtado A, Santos M, Sierra-García JE. Evaluation of XGBoost vs. other Machine Learning models for wind parameters identification. Renew Energy Power Qual J [Internet]. 2023 Jul;21(1):388–93. Available from: https://www.icrepq.com/icrepq23/334-23-garcia.pdf
25. Li X, Li K, Shen S, Tian Y. Exploring Time Series Models for Wind Speed Forecasting: A Comparative Analysis. Energies 2023, Vol 16, Page 7785 [Internet]. 2023 Nov 27 [cited 2025 Mar 9];16(23):7785. Available from: https://www.mdpi.com/1996-1073/16/23/7785/htm
26. UÇAR M, İNCETAŞ MO, BAYRAKTAR I, ÇİLLİ M. Using Machine Learning Algorithms for Jumping Distance Prediction of Male Long Jumpers. J Intell Syst Theory Appl [Internet]. 2022 Sep 1;5(2):145–52. Available from: https://dergipark.org.tr/en/doi/10.38016/jista.1078474