SavedAdd your journal

Journals / Politeknik Dergisi / 2026 / Cilt: 29 Sayı: 4

A Hybrid Deep Learning Model Integrating CNN–LSTM and Metaheuristic Optimization for Seasonal Wind Power Forecasting

Mevsimsel Rüzgâr Gücü Tahmini için CNN–LSTM ve Meta-Sezgisel Optimizasyonu Entegre Eden Hibrit Derin Öğrenme Modeli

Oğuz Taşdemir(KIRŞEHİR AHİ EVRAN ÜNİVERSİTESİ)0000-0003-1782-0024
Salih Ermiş(AHI EVRAN UNIVERSITY)0000-0002-1053-9160
Pages
1–12
Publication Date
2026-06-03
🏛️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)

Overall, the CNN-LSTM model provides a reliable and practical forecasting tool for real-time power prediction and grid applications under varying climatic conditions.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study introduces a medium-term wind power forecasting approach based on hybrid deep learning architectures integrating convolutional and recurrent neural networks with metaheuristic optimization.
📊
Key Findings & Data:Three hybrid models CNN-LSTM, ISO-LSTM, and SO-LSTM were trained using 2024 daily meteorological and power data from a 52.80 MW wind farm in the Marmara Region of Turkey.
💡
Conclusion & Impact:Overall, the CNN-LSTM model provides a reliable and practical forecasting tool for real-time power prediction and grid applications under varying climatic conditions.
🔬Design:Modeling & Simulation
📍Scope:2024 Verileri
⚙️Variables:Hibrit · Derin Öğrenme · CNN-LSTM

Abstract

This study introduces a medium-term wind power forecasting approach based on hybrid deep learning architectures integrating convolutional and recurrent neural networks with metaheuristic optimization. Three hybrid models CNN-LSTM, ISO-LSTM, and SO-LSTM were trained using 2024 daily meteorological and power data from a 52.80 MW wind farm in the Marmara Region of Turkey. Seasonal performance was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) metrics. Among the models, CNN-LSTM achieved the highest accuracy, particularly in summer, with significantly lower error values (MAE = 9.02 MW, MAPE = 4.45%, RMSE = 10.52 MW). ISO-LSTM demonstrated moderate accuracy, while SO-LSTM showed higher errors, especially during autumn and winter due to increased meteorological variability. The strong performance of CNN-LSTM is attributed to its efficient spatial feature extraction and ability to capture medium-term temporal dependencies. Overall, the CNN-LSTM model provides a reliable and practical forecasting tool for real-time power prediction and grid applications under varying climatic conditions.

Özet

Bu çalışma, konvolüsyonel ve tekrarlayan sinir ağlarını meta-sezgisel optimizasyon yöntemleriyle birleştiren hibrit derin öğrenme mimarilerine dayalı orta vadeli rüzgar gücü tahmin yaklaşımı sunmaktadır. CNN-LSTM, ISO-LSTM ve SO-LSTM olmak üzere üç hibrit model, Türkiye’nin Marmara Bölgesi’nde bulunan 52,80 MW’lık bir rüzgar santraline ait 2024 yılı günlük meteorolojik ve üretim verileriyle eğitilmiştir. Modellerin mevsimsel performansı Ortalama Mutlak Hata (OMH), Karesel Ortalama Hata (KOH) ve Ortalama Mutlak Yüzdesel Hata (OMYH) ölçütleri kullanılarak değerlendirilmiştir. Sonuçlar, özellikle yaz döneminde en düşük hata değerlerini sağlayan CNN-LSTM modelinin en yüksek doğruluğa ulaştığını göstermiştir (OMH = 9,02 MW, OMYH = %4,45, KOH = 10,52 MW). ISO-LSTM modeli orta düzey doğruluk sunarken, SO-LSTM modeli özellikle sonbahar ve kış aylarında artan meteorolojik değişkenlik nedeniyle daha yüksek hatalar üretmiştir. CNN-LSTM’nin üstün performansı, teknik özellikleri etkili biçimde çıkarması ve orta vadeli zamansal bağımlılıkları yakalayabilmesine bağlanmaktadır. Genel olarak CNN-LSTM modeli, değişken iklim koşullarında güvenilir ve uygulanabilir bir tahmin aracı sunarak gerçek zamanlı güç tahmini ve şebeke uygulamaları için uygun bir çözüm oluşturmaktadır.

🔖

Cite This Article

Taşdemir, O., Ermiş, S. (2026). A Hybrid Deep Learning Model Integrating CNN–LSTM and Metaheuristic Optimization for Seasonal Wind Power Forecasting. Politeknik Dergisi, Cilt: 29 Sayı: 4, 1–12. https://doi.org/10.2339/politeknik.1822806

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (29)

  1. H. N. D. Şenyapar, U. Cetinkaya, and R. Bayındır, "Renewable Energy Incentives and Future Implications for Turkey: A Comparative Bibliometric Analysis," Politeknik Dergisi, vol. 27, no. 1, pp. 329-342, (2024).
  2. Z. Öztürk and A. Demirci, "Yenilenebilir enerji kaynaklı hibrit güç sistemlerinin farklı penetrasyon ve şebeke tarifeleri altında optimizasyonu," Politeknik Dergisi, vol. 26, no. 3, pp. 1267-1275, (2023).
  3. T. Kåberger, "Progress of renewable electricity replacing fossil fuels," Global Energy Interconnection, vol. 1, no. 1, pp. 48-52, (2018).
  4. A. Dhabi, "Irena," Renewable energy statistics, (2025).
  5. TÜREB, Türkiye Rüzgar Enerjisi İstatistik Raporu, (2025).
  6. D. Hdidouan and I. Staffell, "The impact of climate change on the levelised cost of wind energy," Renewable energy, vol. 101, pp. 575-592, (2017).
  7. S. Niu, Z. Zhang, X. Ke, G. Zhang, C. Huo, and B. Qin, "Impact of renewable energy penetration rate on power system transient voltage stability," Energy Reports, vol. 8, pp. 487-492, (2022).
  8. R. Tawn and J. Browell, "A review of very short-term wind and solar power forecasting," Renewable and Sustainable Energy Reviews, vol. 153, 111758, (2022).
  9. Q. Chen and K. Folly, "Wind Power Forecasting," IFAC-PapersOnLine, vol. 51, no. 28, pp. 414–419, (2018).
  10. Y. Tang, K. Yang, Y. Zheng, L. Ma, S. Zhang, and Z. Zhang, "Wind power forecasting: A transfer learning approach incorporating temporal convolution and adversarial training," Renewable Energy, vol. 224, 120200, (2024).
  11. I. Karijadi, S.Y. Chou, and A. Dewabharata, "Wind power forecasting based on hybrid CEEMDAN-EWT deep learning method," Renewable Energy, vol. 218, 119357, (2023).
  12. Y. Zhang, X. Kong, J. Wang, H. Wang, and X. Cheng, "Wind power forecasting system with data enhancement and algorithm improvement," Renewable and Sustainable Energy Reviews, vol. 196, 114349, (2024).
  13. Y. Zhao, S. Pan, Y. Zhao, H. Liao, L. Ye, and Y. Zheng, "Ultra-short-term wind power forecasting based on personalized robust federated learning with spatial collaboration," Energy, vol. 288, 129847, (2024).
  14. X. He, B. He, T. Qin, C. Lin, and J. Yang, "Ultra-short-term wind power forecasting based on a dual-channel deep learning model with improved coot optimization algorithm," Energy, vol. 305, 132320, (2024).
  15. L. Liu, X. Wang, X. Dong, K. Chen, Q. Chen, and B. Li, "Interpretable feature-temporal transformer for short-term wind power forecasting with multivariate time series," Applied Energy, vol. 374, 124035, (2024).
  16. M. Öztürk, R. Kayabaşı, and O. Taşdemir, "Kirşehir’in Rüzgar Enerjisi Potansiyeli ve İç Anadolu Bölgesi Kurulu Rüzgar Enerjisi Santrallerinin Güç Analizi," Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, vol. 28, no. 1, pp. 189-201, (2025).
  17. O. Taşdemir, İ. F. Tepe, and E. Irmak, "Optimization of Seasonal Wind Energy Forecasts with PSO Algorithm," in 2024 6th Global Power, Energy and Communication Conference (GPECOM), IEEE, pp. 543-548, (2024).
  18. S. Hanifi, H. Zare-Behtash, A. Cammarano, and S. Lotfian, "Offshore wind power forecasting based on WPD and optimised deep learning methods," Renewable Energy, vol. 218, 119241, (2023).
  19. U. Yuzgec, E. Dokur, and M. Balci, "A novel hybrid model based on empirical mode decomposition and echo state network for wind power forecasting," Energy, vol. 300, 131546, (2024).
  20. H. Zhao, P. Xu, T. Gao, J. J. Zhang, J. Xu, and D. W. Gao, "CPTCFS: CausalPatchTST incorporated causal feature selection model for short-term wind power forecasting of newly built wind farms," International Journal of Electrical Power & Energy Systems, vol. 160, 110059, (2024).
  21. S. Mahata, P. Harsh, and V. Shekher, "Comparative study of time-series forecasting models for wind power generation in Gujarat, India," e-Prime-Advances in Electrical Engineering, Electronics and Energy, vol. 8, 100511, (2024).
  22. L. Li, Z.Y. Cen, M.L. Tseng, Q. Shen, and M. H. Ali, "Improving short-term wind power prediction using hybrid improved cuckoo search arithmetic-support vector regression machine," Journal of Cleaner Production, vol. 279, 123739, (2021).
  23. Y. LeCun et al., "Backpropagation applied to handwritten zip code recognition," Neural computation, vol. 1, no. 4, pp. 541-551, (1989).
  24. Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, (2002).
  25. S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural computation, vol. 9, no. 8, pp. 1735-1780, (1997).
  26. F. A. Hashim and A. G. Hussien, "Snake Optimizer: A novel meta-heuristic optimization algorithm," Knowledge-Based Systems, vol. 242, 108320, (2022).
  27. Y. Gu, M. Chen, and L. Wang, "A self-learning discrete salp swarm algorithm based on deep reinforcement learning for dynamic job shop scheduling problem," Applied Intelligence, vol. 53, no. 15, pp. 18925-18958, (2023).
  28. H. Wang, Z. Chen, G. Kozinets, P. Kozinets, M. Alexey, and P. Cao, "Multi-strategy improved snake optimization algorithm for predicting the ice melting rate of a thermal coring bit," International Journal of Heat and Mass Transfer, vol. 252, pp. 127491, (2025).
  29. Y. Zhou, R. Huang, Q. Lin, Q. Chai, and W. Wang, "Probabilistic optimization based adaptive neural network for short-term wind power forecasting with climate uncertainty," International Journal of Electrical Power & Energy Systems, vol. 157, pp. 109897, (2024).