SavedAdd your journal

Journals / Balkan Journal of Electrical and Computer Engineering / 2026 / Cilt: 14

Hybrid CNN–ML System for Wind Speed Level Identification in Complex Terrain: A Case Study from Maden, Turkey

Karmaşık Arazilerde Rüzgar Hızı Seviyesi Belirlemesi için Hibrit CNN-ML Sistemi: Maden, Türkiye'den Bir Vaka Çalışması

H Selcuk Nogay(Bursa Uludag University)0000-0001-9105-508X
Pages
33–40
Publication Date
2026-03-27
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The study highlights the effectiveness of combining deep feature representations with interpretable classifiers in data-scarce, topographically challenging regions, offering a transparent and high-performance alternative for wind energy assessment.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study proposes a hybrid deep and traditional learning framework for wind speed level classification in the complex terrain of the Maden region, Turkey.
📊
Key Findings & Data:A one-dimensional convolutional neural network (1D-CNN) was employed for automatic feature extraction from a 30-day meteorological window, followed by classification using multiple machine learning algorithms.
💡
Conclusion & Impact:The study highlights the effectiveness of combining deep feature representations with interpretable classifiers in data-scarce, topographically challenging regions, offering a transparent and high-performance alternative for wind energy assessment.
🔬Design:Lab & Field Experiment
📍Scope:Comprehensive Sample
⚙️Variables:Özellik tabanlı kategorizasyon · karmaşık arazi · yenilenebilir enerji planlaması

Abstract

This study proposes a hybrid deep and traditional learning framework for wind speed level classification in the complex terrain of the Maden region, Turkey. A one-dimensional convolutional neural network (1D-CNN) was employed for automatic feature extraction from a 30-day meteorological window, followed by classification using multiple machine learning algorithms. Among them, K-Nearest Neighbor (KNN) achieved the highest accuracy (98.75%) when applied to features extracted from the global average pooling (GAP) layer. The hybrid CNN–KNN model significantly outperformed standalone CNN and KNN baselines. The study highlights the effectiveness of combining deep feature representations with interpretable classifiers in data-scarce, topographically challenging regions, offering a transparent and high-performance alternative for wind energy assessment.

Özet

Bu çalışma, Türkiye'nin Maden bölgesinin karmaşık arazisinde rüzgar hızı seviyesi sınıflandırması için hibrit derin ve geleneksel öğrenme çerçevesini önermektedir. 30 günlük bir meteorolojik pencereden otomatik özellik çıkarımı için tek boyutlu bir evrişimli sinir ağı (1D-CNN) kullanılmış ve ardından birden fazla makine öğrenme algoritması kullanılarak sınıflandırma yapılmıştır. Bunlar arasında K-En Yakın Komşu (KNN), küresel ortalama havuzlama (GAP) katmanından çıkarılan özelliklere uygulandığında en yüksek doğruluğu (%98,75) elde etmiştir. Hibrit CNN-KNN modeli, bağımsız CNN ve KNN temel çizgilerinden önemli ölçüde daha iyi performans göstermiştir. Çalışma, veri kıtlığı olan, topoğrafik olarak zorlu bölgelerde derin özellik temsillerini yorumlanabilir sınıflandırıcılarla birleştirmenin etkinliğini vurgulayarak, rüzgar enerjisi değerlendirmesi için şeffaf ve yüksek performanslı bir alternatif sunmaktadır.

🔖

Cite This Article

Nogay, H. S. (2026). Hybrid CNN–ML System for Wind Speed Level Identification in Complex Terrain: A Case Study from Maden, Turkey. Balkan Journal of Electrical and Computer Engineering, Cilt: 14, 33–40. https://doi.org/10.17694/bajece.1820323

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (34)

  1. X.Y. Chen, L. Zheng, …, J.X. He, “Global perspectives on wind energy innovation: Policy impacts and component-level analysis”, Energy, Vol.319, 2025, 129476.
  2. J.A. Carta, P. Ramírez, S. Velázquez, “A review of wind speed probability distributions used in wind energy analysis: Case studies in the Canary Islands”, Renewable & Sustainable Energy Reviews, Vol.13, No.5, 2009, pp.933–955.
  3. L. Le Toumelin, I. Gouttevin, N. Helbig, “A two-fold deep-learning strategy to correct and downscale winds over mountains”, Nonlinear Processes in Geophysics, Vol.31, No.1, 2024, pp.75–97.
  4. R. Shen, B. Li, K. Li, B. Yan, Y. Zhang, “Reconstruction of wind speed fields in mountainous areas using a full convolutional neural network”, Wind and Structures, Vol.38, No.4, 2024, pp.231–244.
  5. J. Kim, H.-J. Shin, K. Lee, J. Hong, “Enhancement of ANN-based wind power forecasting by modification of surface roughness parameterization over complex terrain”, Journal of Environmental Management, Vol.362, 2024, 121246.
  6. K. Wang, X.-Y. Tang, S. Zhao, “Robust multi-step wind speed forecasting based on a graph-based data reconstruction deep learning method”, Expert Systems with Applications, Vol.238(B), 2024, 121886.
  7. Z. Li, W. Sun, Y. Xiang, G.P. Harrison, “Transfer Strategy for Power Output Estimation of Wind Farm at Planning Stage Based on a SVR Model”, CSEE Journal of Power and Energy Systems, Vol.9, No.4, 2022, pp.1460–1471.
  8. K. Reinhardt, C. Samimi, “Comparison of different wind data interpolation methods for a region with complex terrain in Central Asia”, Climate Dynamics, Vol.51, 2018, pp.3635–3652.
  9. Z. Sun, J. Li, R. Guo, et al., “Machine learning-based temperature and wind forecasts in the Zhangjiakou Competition Zone during the Beijing 2022 Winter Olympic Games”, Journal of Meteorological Research, Vol.38, 2024, pp.664–679.
  10. E. Pauli, H. Andersen, J. Bendix, J. Cermak, S. Egli, “Determinants of fog and low stratus occurrence in continental central Europe – a quantitative satellite-based evaluation”, Journal of Hydrology, Vol.591, 2020, 125451.
  11. X. Gong, Y. Zhu, Y. Wang, E. Li, Y. Zhang, Z. Zhang, “Safety status prediction model of transmission tower based on improved Coati optimization-based support vector machine”, Buildings, Vol.14, 2024, 3815.
  12. S. Mulewa, A. Parmar, A. De, “A novel Bagged-CNN architecture for short-term wind power forecasting”, International Journal of Green Energy, Vol.21, No.12, 2024, pp.2712–2723.
  13. A. Casallas, C. Ferro, N. Celis, et al., “Long short-term memory artificial neural network approach to forecast meteorology and PM2.5 local variables in Bogotá, Colombia”, Modeling Earth Systems and Environment, Vol.8, 2022, pp.2951–2964.
  14. N. Ohana-Levi, S. Munitz, A. Ben-Gal, A. Schwartz, A. Peeters, Y. Netzer, “Multiseasonal grapevine water consumption – drivers and forecasting”, Agricultural and Forest Meteorology, Vol.280, 2020, 107796.
  15. P. Huang, Q. Huang, J. Wang, et al., “Predicting surface soil pH spatial distribution based on three machine learning methods: A case study of Heilongjiang Province”, Environmental Monitoring and Assessment, Vol.197, 2025, 367.
  16. S. Wang, M. Sun, G. Wang, et al., “Simulation and reconstruction of runoff in the high-cold mountains area based on multiple machine learning models”, Water, Vol.15, 2023, 3222.
  17. R.M. Adnan, Z. Liang, S. Heddam, et al., “Least square support vector machine and multivariate adaptive regression splines for streamflow prediction in mountainous basin using hydro-meteorological data”, Journal of Hydrology, Vol.586, 2020, 124371.
  18. Y.Y. Dai, M.J. Zhang, M.Y. Liu, L. Wang, “Short-term wind speed forecasts through hybrid model based on improved variational mode decomposition”, International Journal of Green Energy, Vol.21, No.10, 2024, pp.2281–2298.
  19. K.P. Natarajan, J.G. Singh, “Wind speed forecasting using a combined deep learning model with slime mould optimization”, International Journal of Green Energy, 2025.
  20. D. Mahajan, S. Sharma, K. Saini, “Real-time rainfall projection in hilly areas using ARLSTM model: A case of Dharamshala, India”, Engineering Research Express, Vol.6, No.4, 2024, 045261.
  21. A.K. Tran, Q.M. Nguyen, H.S. Hoang, et al., “Wireless sensor networks and machine learning meet climate change prediction”, International Journal of Communication Systems, Vol.34, No.3, 2021, e4687.
  22. D. Boateng, S.G. Mutz, “pyESDv1.0.1: An open-source Python framework for empirical-statistical downscaling of climate information”, Geoscientific Model Development, Vol.16, No.22, 2023, pp.6479–6514.
  23. E. Ammar, G. Xydis, “Wind speed forecasting using deep learning and preprocessing techniques”, International Journal of Green Energy, Vol.21, No.5, 2024, pp.988–1016.
  24. E. Gürsoy, M. Gürdal, E. Gedik, “Wind speed prediction by utilizing geographic information system and machine learning approach: A case study of Karabük province in Türkiye”, International Journal of Green Energy, Vol.22, No.9, 2025, pp.1799–1815.
  25. T.C. Akinci, H.S. Noğay, M. Penchev, A.A. Martinez-Morales, A. Raju, “A hybrid approach to wind power intensity classification using decision trees and large language models”, Renewable Energy, Vol.250, 2025, pp.1–11.
  26. X. Zhang, X. Li, L. Li, et al., “Environmental factors influencing snowfall and snowfall prediction in the Tianshan Mountains, Northwest China”, Journal of Arid Land, Vol.11, 2019, pp.15–28.
  27. D. Kim, J. Lim, S. Kim, “Wind resource assessment in South Korean mountainous regions using ensemble learning”, Renewable Energy, Vol.206, 2023, pp.1156–1168.
  28. J.Y. Lee, K. Lee, “A comparative analysis of machine learning algorithms for wind speed prediction over coastal and inland sites”, Energy Conversion and Management, Vol.224, 2020, 113353.
  29. Z. Zhang, C. Wang, B. Lv, “Comparative analysis of ecological sensitivity assessment using the coefficient of variation method and machine learning”, Environmental Monitoring and Assessment, Vol.196, 2024, 1000.
  30. X. Pan, J. Yang, H. Xu, “Wind speed forecasting using an improved deep belief network in mountainous regions”, IEEE Access, Vol.9, 2021, pp.123456–123467.
  31. R. Ahmed, M. Khalid, A. Hussain, “Deep ensemble learning for wind speed forecasting: A case from mountainous northern Pakistan”, Sustainable Energy Technologies and Assessments, Vol.40, 2020, 100781.
  32. S. Ghavidel, R. Azizipanah-Abarghooee, “Application of hybrid artificial intelligence models for wind power forecasting in complex terrains”, Applied Soft Computing, Vol.113, 2022, 107919.
  33. N. Garcia-Hernando, L.A. Fernandez-Jimenez, “Short-term wind speed prediction using XGBoost and feature engineering in complex terrains”, Energies, Vol.13, No.24, 2020, 6611.
  34. Y. Zhang, Y. Guo, “Wind speed modeling using hybrid CNN-LSTM networks in non-uniform terrain”, Applied Energy, Vol.205, 2017, pp.1344–1356.