SavedAdd your journal

Journals / Journal of Information Systems and Management Research / 2025 / Cilt: 7 Sayı: 2

Use of Advanced Optimization and Hybrid Deep Learning Models in Ultra-Short-Term Wind Energy Forecasting: The Case of Hatay, Türkiye

Ultra Kısa Vadeli Rüzgar Enerjisi Tahmininde Gelişmiş Optimizasyon ve Hibrit Derin Öğrenme Modellerinin Kullanımı: Hatay, Türkiye Örneği

Orkun Teke(CELÂL BAYAR ÜNİVERSİTESİ)0000-0003-4390-263X
Tolga Depci(İSKENDERUN TEKNİK ÜNİVERSİTESİ)0000-0001-9562-8068
Pages
214–229
Publication Date
2025-12-31
🏛️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 findings indicate a significant superiority of the hybrid architectures over the baseline (SDAE) model: CNN-LSTM maintains consistently high accuracy across all horizons, while GRU-LSTM yields the lowest error metrics specifically at the shortest horizon (achieving an R²=0.9976 at t+1).

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study presents a performance comparison of hybrid deep learning approaches with an end- to-end data pipeline designed to enhance accuracy and stability in ultra-short-term wind power forecasting.
📊
Key Findings & Data:An advanced hyperparameter optimization tool, Optuna, trained the models (SDAE baseline, CNN-LSTM, and GRU-LSTM) using progressive search and pruning strategies.
💡
Conclusion & Impact:The findings indicate a significant superiority of the hybrid architectures over the baseline (SDAE) model: CNN-LSTM maintains consistently high accuracy across all horizons, while GRU-LSTM yields the lowest error metrics specifically at the shortest horizon (achieving an R²=0.9976 at t+1).
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Rüzgar gücü kestirimi · ultra-kısa dönem tahmin · CNN-LSTM

Abstract

This study presents a performance comparison of hybrid deep learning approaches with an end- to-end data pipeline designed to enhance accuracy and stability in ultra-short-term wind power forecasting. Anomaly removal using DBSCAN and feature selection based on RFECV are applied to multivariate SCADA-based data. An advanced hyperparameter optimization tool, Optuna, trained the models (SDAE baseline, CNN-LSTM, and GRU-LSTM) using progressive search and pruning strategies. Performance is evaluated using MAE, RMSE, and R² metrics from t+1 up to t+6 horizons. The findings indicate a significant superiority of the hybrid architectures over the baseline (SDAE) model: CNN-LSTM maintains consistently high accuracy across all horizons, while GRU-LSTM yields the lowest error metrics specifically at the shortest horizon (achieving an R²=0.9976 at t+1). The stability of the CNN-LSTM is maintained as the forecasting horizon extends, achieving a respectable performance of R²=0.79 even at t+6. This work proposes the operational use of GRU-LSTM for the shortest-term forecasts and CNN-LSTM for more stable predictions as the horizon lengthens. The results demonstrate that hybrid models establish a reliable foundation for industrial applications and suggest further gains are possible through the integration of uncertainty modeling and Numerical Weather Prediction (NWP).

Özet

Bu çalışma, Bu çalışma, ultra kısa vadeli rüzgar enerjisi tahmininde doğruluğu ve kararlılığı artırmak amacıyla tasarlanmış uçtan uca bir veri hattı ile hibrit derin öğrenme yaklaşımlarının performans karşılaştırmasını sunmaktadır. SCADA tabanlı çok değişkenli verilere, anomali giderme işlemi DBSCAN ile, özellik seçimi ise RFECV tabanlı olarak uygulanmıştır. Optuna gibi gelişmiş bir hiperparametre optimizasyon aracı, ilerici arama ve budama stratejileri kullanarak modelleri (SDAE temel, CNN-LSTM ve GRU-LSTM) eğitmiştir. Performans, t+1'den t+6'ya kadar MAE, RMSE ve R² metrikleriyle değerlendirilmiştir. Bulgular, hibrit mimarilerin temel (SDAE) modele kıyasla önemli bir üstünlük sergilediğini göstermektedir: CNN-LSTM tüm ufuklarda sürekli olarak yüksek doğruluk sergilerken, GRU-LSTM özellikle en kısa ufukta (t+1'de R²=0.9976) en düşük hata değerlerini üretmiştir. CNN-LSTM'nin kararlılığı ufuk uzadıkça korunmuş ve t+6'da bile R²=0.79'luk yüksek bir performans elde edilmiştir. Bu çalışma, en kısa vadeli tahminler için GRU-LSTM, ufuk uzadıkça daha kararlı tahminler için ise CNN-LSTM'in operasyonel kullanımını önermektedir. Sonuçlar, hibrit modellerin endüstriyel uygulamalar için güvenilir bir temel oluşturduğunu ve belirsizlik modellemesi ile Sayısal Hava Tahmini (NWP) entegrasyonuyla daha ileri kazanımlar elde edilebileceğini göstermektedir.

🔖

Cite This Article

Teke, O., Depci, T. (2025). Use of Advanced Optimization and Hybrid Deep Learning Models in Ultra-Short-Term Wind Energy Forecasting: The Case of Hatay, Türkiye. Journal of Information Systems and Management Research, Cilt: 7 Sayı: 2, 214–229. https://doi.org/10.59940/jismar.1787678

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (60)

  1. Behera, P., Sethi, N., Dash, D. P., Usman, M., & Sahu, P. K. (2025). “Pathways to achieve carbon neutrality in emerging economies: Catalyzing the role of renewable energy, green growth”, ICT, and political risk. Renewable Energy, 243, 122514.
  2. Kok, B., Benli, H., "Energy diversity and nuclear energy for sustainable development in Turkey", Renewable energy, 111 (2017): 870-877.
  3. M. C. Senel, E. Koc, 2015. “Wind Energy in the World and Turkey Condition-General Evaluation,” Journal of Engineers and Machinery, volume 56, issue 663, p. 46-56.
  4. Teke, O. (2025). “Analysis of Performance Results of Models Created for Wind Power Estimation (PhD. Thesis)”. Iskenderun Technical University Energy Institute, Hatay.
  5. Ayar B., Gülten Yalçın Z., Dağ M., (2023). “Harvesting the Wind: A Study on the Feasibility and Advancements of Wind Energy in Turkey”. European Journal of Science and Technology Special Issue 49, 43-49.
  6. Albostan, A., Çekiç, Y. and Eren, L. 2009. “The Impact of Wind Energy on Turkey's Supply Security”, Gazi University Faculty of Engineering and Architecture Journal, 24 (4): 641-649.
  7. Zheng, Z., Yang, Y., Liu, J., Dai, H.-N., & Zhang, Y. (2019). “Deep and embedded learning approach for traffic flow prediction in urban informatics”. IEEE Transactions on Intelligent Transportation Systems, 21(10), 1–13. https://doi.org/10.1109/TITS.2019.2909904
  8. Pandit, R., Astolfi, D., & Durazo-Cardenas, I. (2023). “A review of predictive techniques used to support decision making for maintenance operations of wind turbines”. Energies, 16(4), 1654. https://doi.org/10.3390/en16041654. [9] Tekin, S. F., Fazla, A., & Kozat, S. S. (2021). “Numerical weather forecasting using convolutional-LSTM with attention and context matcher mechanisms”. arXiv. https://doi.org/10.48550/arXiv.2102.00696
  9. Liu, Z., Guo, H., Zhang, Y., & Zuo, Z. (2025). A “Comprehensive Review of Wind Power Prediction Based on Machine Learning: Models, Applications, and Challenges”. Energies, 18(2), 350. https://doi.org/10.3390/en18020350
  10. Kang, B., Xu, Z., He, W., Ding, G., Han, W., Sun, M., et al. (2024). “Research on the double-layer clustering method of residential energy use characteristics under the background of energy system energy savings and carbon reduction”. Front. Energy Res. 11, 1326257. https://doi:10.3389/fenrg.2023.1326257
  11. Kariniotakis, G., Pinson, P., Siebert, N., Giebel, G., & Barthelmie, R. (2004, October). “The state of the art in short term prediction of wind power – from an offshore perspective”. In SeaTech Week – Ocean Energy Conference ADEME-IFREMER, Brest, France. HAL. https://hal.science/hal-00529338. Last Access Date: 10.02.2025
  12. Yang, T., Yang, Z., Li, F., & Wang, H. (2024). “A short-term wind power forecasting method based on multivariate signal decomposition and variable selection”. Applied Energy, 360, 122759. https://doi.org/10.1016/j.apenergy.2024.122759
  13. Yi, L., Chen, B., Zhan, J., Wang, Y., Huang, Y., Long, J., & Sun, T. (2025). “Multi-scale spatial–temporal interactive network for wind turbine ultra-short-term power prediction” L. Yi et al. The Journal of Supercomputing, 81(8), 906.
  14. Deng, Y. C., Tang, X. H., Zhou, Z. Y., Yang, Y., & Niu, F. (2025). “Application of machine learning algorithms in wind power: a review”. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 47(1), 4451-4471.
  15. Ghanbari, E., & Avar, A. (2025). “Short-term wind power forecasting using the hybrid model of multivariate variational mode decomposition (MVMD) and long short-term memory (LSTM) neural networks”. Electrical Engineering, 107(3), 2903-2933.
  16. Rosca, C. M., & Stancu, A. (2025). “A Comprehensive Review of Machine Learning Models for Optimizing Wind Power Processes”. Applied Sciences, 15(7), 3758.
  17. Elsaraiti, M., & Merabet, A. (2021). “A comparative analysis of the arima and lstm predictive models and their effectiveness for predicting wind speed”. Energies, 14(20), 6782.
  18. Feng, C., Cui, M., Hodge, B. M., & Zhang, J. (2017). “A data-driven multi-model methodology with deep feature selection for short-term wind forecasting”. Applied Energy, 190, 1245-1257
  19. Paik, C., Chung, Y., & Kim, Y. J. (2023). “Power Curve Modeling of Wind Turbines through Clustering-Based Outlier Elimination”. Applied System Innovation, 6(2), 41. https://doi.org/10.3390/asi6020041
  20. Zhao, Y., Ye, L., Wang, W., Sun, H., Ju, Y., and Tang, Y. (2018). “Data-driven correction approach to refine power curve of wind farm under wind curtailment”. IEEE Transactions on Sustainable Energy, 9(1), 95-105. https://doi.org/10.1109/TSTE.2017.2717021
  21. Veljanovski, G., Popovski, P., Atanasovski, M., & Kostov, M. (2022). “Implementation of neural networks and feature selection for short term load forecast”. 2022 57th International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST), 229–232. IEEE. https://doi.org/10.1109/ICEST55464.2022.9828610. Last Access Date: 10.02.2025
  22. Afrasiabi, M., Mohammadi, M., Rastegar, M., & Afrasiabi, S. (2020). Advanced deep learning approach for probabilistic wind speed forecasting. IEEE Transactions on Industrial Informatics, 17(1), 720-727.
  23. Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., Manzagol, P. A., & Bottou, L. (2010). “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion”. Journal of machine learning research, 11(12).
  24. Liu, X., Zhang, L., Zhang, Z., Zhao, T., ve Zou, L. (2021b). “Ultra Short-Term wind power prediction model based on wrf wind speed prediction and CatBoost”. IOP Conference Series: Earth and Environmental Science, 838, 11-22. https://doi.org/10.1088/1755-1315/838/1/012001
  25. Tyass, I., Khalili, T., Mohamed, R., Abdelouahed, B., Raihani, A., & Mansouri, K. (2023). Wind speed prediction based on statistical and deep learning models. International Journal of Renewable Energy Development, 12(2), 288.
  26. Yu, C., Yan, G., Yu, C., & Mi, X. (2023). Attention mechanism is useful in spatio-temporal wind speed prediction: Evidence from China. Applied Soft Computing, 148, 110864.
  27. Yin, D., Zhao, L., Zhai, K., & Zheng, J. (2024). “S. Term Wind Power Prediction Based on Feature-Weighted and Combined Models”. Applied Sciences, 14(17), 7698. https://doi.org/10.3390/app14177698
  28. Alijoyo, F. A., Gongada, T. N., Kaur, C., Mageswari, N., Sekhar, J. C., Ramesh, J. V. N., El-Ebiary, Y. A. B. and Ulmas, Z. (2024). “Advanced Hybrid CNN-Bi-LSTM Model Augmented with GA And FFO For Enhanced Cyclone İntensity Forecasting”. Alexandria Engineering Journal, 92, 346-357. https://doi.org/https://doi.org/10.1016/j.aej.2024.02.062
  29. Dai, X., Liu, G. P., & Hu, W. (2023). “An online-learning-enabled self-attention-based model for ultra-short-term wind power forecasting”. Energy, 272, 127173.
  30. Nguyen, T. H. T., and Phan, Q. B. (2022). “Hourly day ahead wind speed forecasting based on a hybrid model of EEMD, CNN-Bi-LSTM embedded with GA optimization”. Energy Reports, 8, 53-60. https://doi.org/https://doi.org/10.1016/j.egyr.2022.05.110
  31. Li, Q., Wang, G., Wu, X., Gao, Z., & Dan, B. (2024). “Arctic short-term wind speed forecasting based on CNN-LSTM model with CEEMDAN”. Energy, 299, 131448. https://doi.org/10.1016/j.energy.2024.131448
  32. Park, R.-J., Kang, S., Lee, J., and Jung, S. (2022). “CNN-LSTM based wind power prediction system to improve accuracy”. Renewable Energy. 4(2), 144-166. https://api.semanticscholar.org/CorpusID:24903056.Last Access Date: 15.02.2025
  33. Bilal, B., Adjallah, K. H., Sava, A., Yetilmezsoy, K., & Ouassaid, M. (2023). “Wind turbine output power prediction and optimization based on a novel adaptive neuro-fuzzy inference system with the moving window”. Energy, 263(E), 126159. https://doi.org/10.1016/j.energy.2022.126159
  34. Zhang, H., Zhao, L., and Du, Z. (2021). “Wind power prediction based on CNN-LSTM”. 2021 IEEE 5th Conference on Energy Internet and Energy System Integration (EI2), 3097-3102. https://api.semanticscholar.org/CorpusID:24710878. Last Access Date: 15.02.2025
  35. Yong, B., Qiao, F., Wang, C., Shen, J., Wei, Y., & Zhou, Q. (2019). “Ensemble neural network method for wind speed forecasting”. In 2019 IEEE International Workshop on Signal Processing Systems (SiPS) (pp. 31–36). IEEE. https://doi.org/10.1109/SiPS47522.2019.9020410. Last Access Date: 15.03.2025
  36. Xin, Z., Liu, X., Zhang, H., Wang, Q., An, Z., & Liu, H. (2024). An enhanced feature extraction based long short-term memory neural network for wind power forecasting via considering the missing data reconstruction. Energy Reports, 11, 97-114.
  37. Kuang, H., Guo, Q., Li, S., and Zhong, H. (2021). “Short-term wind power forecasting model based on multi-feature extraction and CNN-LSTM”. IOP Conference Series: Earth and Environmental Science, 702, 43-49. https://api.semanticscholar.org/CorpusID:23385927.Last Access Date: 15.02.2025
  38. Zhao, Z., Yun, S., Jia, L., Guo, J., Meng, Y., He, N., Li, X., Shi, J., & Yang, L. (2023). “Hybrid VMD-CNN-GRU-based model for short-term forecasting of wind power considering spatio-temporal features”. Engineering Applications of Artificial Intelligence, 121, 105982. https://doi.org/10.1016/j.engappai.2023.105982
  39. Zhang, D., Chen, B., Zhu, H., Goh, H. H., Dong, Y., & Wu, T. (2023). “Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model”. Energy, 285, 128762.https://doi.org/10.1016/j.energy.2023.128762
  40. Joseph, L. P., Deo, R. C., Casillas-Pèrez, D., Prasad, R., Raj, N., & Salcedo-Sanz, S. (2024). “Multi-step-ahead wind speed forecast system: Hybrid multivariate decomposition and feature selection-based gated additive tree ensemble model”. IEEE Access, 12, 58750-58777.
  41. Mao, J., Zhao, J., Zhang, H., & Gu, B. (2025). A Novel Hybrid Deep Learning Model for Day-Ahead Wind Power Interval Forecasting. Sustainability, 17(7), 3239. https://doi.org/10.3390/su17073239
  42. Emu, I. H., Rahman, M. S., Farzana, F., Islam, M. R., Shiam, I. F., & Sultana, A. (2024). “Wind power forecasting using long short-term memory (LSTM), gated recurrent unit (GRU) and hybrid model”. 2024 IEEE International Conference on Electrical, Computer and Communication Engineering (ECCE), 1–6. IEEE. https://doi.org/10.1109/ECCE60783.2024.10533822. Last Access Date: 18.02.2025
  43. Lai, J.P., Lin, Y.L., Lin, H.C., Shih, C.Y., Wang, Y.P., & Pai, P.F. (2023). “Tree-Based Machine Learning Models with Optuna in Predicting Impedance Values for Circuit Analysis”. Micromachines, 14(2), 265. https://doi.org/10.3390/mi14020265
  44. Almarzooq, H., and Bin Waheed, U. (2024). “Automating hyperparameter optimization in geophysics with Optuna: a comparative study.” Geophysical Prospecting, 72. 25-46. https://doi.org/10.1111/1365-2478.13484
  45. Hanifi, S., Cammarono, A., and Zare Behtash, H. (2024a). “Advanced hyperparameter optimization of deep learning models for wind power prediction”. Renewable Energy, 221, 119-131. https://doi.org/https://doi.org/10.1016/j.renene.2023.119700
  46. Sun, T.-H., Wang, C.-C., Wu, Y.-L., Hsu, K.-C., & Lee, T.-H. (2023). Machine learning approaches for biomarker discovery to predict large-artery atherosclerosis. Scientific Reports, 13, 15809. https://doi.org/10.1038/s41598-023-42338-0
  47. Chen, C. and Liu, L.-M. (1993), Forecasting time series with outliers. J. Forecast., 12: 13-35. https://doi.org/10.1002/for.3980120103
  48. Battaglia, F., and Cucina, D. (2020). “Outlier identifiability in time series”. Stat, 9(1), 281-295. https://doi.org/https://doi.org/10.1002/sta4.281
  49. Che, Z., Purushotham, S., Cho, K., Sontag, D., and Liu, Y. (2018). “Recurrent neural networks for multivariate time series with missing values”. Scientific Reports, 8(1), 60-85. https://doi.org/10.1038/s41598-018-24271-9
  50. Lee, K., Park, B., Kim, J., ve Hong, J. (2024). “Day-ahead wind power forecasting based on feature extraction integrating vertical layer wind characteristics in complex terrain”. Energy, 288, 129-143. https://doi.org/https://doi.org/10.1016/j.energy.2023.129713
  51. Lipton, Z. Kale, D. Elkan, C. and Wetzel, R. (2015). “Learning to diagnose with LSTM recurrent neural networks.” arXiv. https://doi.org/10.48550/arXiv.1511.03677
  52. Sriramalakshmi, P., Subhasree, V., Vondivillu, S. T., & Prakash, S. (2022, May). “Time Series Analysis and Forecasting of Wind Turbine Data”. In 2022 International Virtual Conference on Power Engineering Computing and Control: Developments in Electric Vehicles and Energy Sector for Sustainable Future (PECCON) (pp. 1-9). IEEE. Last Access Date: 15.03.2025
  53. Ahsan, M. M., Mahmud, M. A. P., Saha, P. K., Gupta, K. D., & Siddique, Z. (2021). “Effect of Data Scaling Methods on Machine Learning Algorithms and Model Performance”. Technologies, 9(3), 52. https://doi.org/10.3390/technologies9030052
  54. Zhang, S., Zhang, C., and Yang, Q. (2003). “Data Preparation for data mining”. Applied Artificial Intelligence, 17, 375-381. https://doi.org/10.1080/713827180
  55. What Is Data Mining. (2006). “Introduction to data mining”. In Mining multimedia databases, mining time series.
  56. Zucatelli, P. J., Nascimento, E. G. S., Santos, A. Á. B., Arce, A. M. G., and Moreira, D. M. (2019). “Study of wind speed forecasting applying computational intelligence”. Aerodynamics. 4(2), 73-100. https://doi.org/10.5772/intechopen.89758
  57. Shi, K., Shi, R., Fu, T., Lu, Z., & Zhang, J. (2024). “A Novel Identification Approach Using RFECV–Optuna–XGBoost for Assessing Surrounding Rock Grade of Tunnel Boring Machine Based on Tunneling Parameters”. Applied Sciences, 14(6), 2347. https://doi.org/10.3390/app14062347
  58. Patel, S. K., Panchal, S. R., & Panchal, S. S. (2024, December 19). “Wind Nowcast On Radar Imagery Employing Deep Learning Technique”. Journal of Electrical Systems, 20(3). https://doi.org/10.52783/jes.7493
  59. Bergstra, J., and Bengio, Y. (2012). “Random search for hyper-parameter optimization”. J. Mach. Learn. Res., 13(2), 281-305.
  60. Anande, T., Alsaadi, S. & Leeson, M. Enhanced Modelling Performance with Boosting Ensemble Meta-Learning and Optuna Optimization. SN COMPUT. SCI. 6, 12 (2025). https://doi.org/10.1007/s42979-024-03544-3