Derin Öğrenme ile Kısa Vadeli Rüzgar Hız Tahmini

Year 2025, Volume: 4 Issue: 1, 151 - 162, 18.02.2025

Fatih Karaaslan , Zeynep Mine Alçin , Muzaffer Aslan

https://doi.org/10.62520/fujece.1517615

Abstract

Rüzgar hızı, yenilenebilir rüzgar enerjisine yatırım yapılması ve planlanmasında rüzgar hızının tahmin edilmesi hayati önem taşımaktadır. Ayrıca mevcut rüzgar santrallerinin üretimi ve iletim hatlarının kapasitelerinin artırılmasında da rüzgar hızının tahmin edilmesi oldukça önem arz etmektedir. Fakat rüzgar hızının aralıklı ve stokastik dalgalanmaları, yüksek kaliteli rüzgar hızı tahmini için önemli bir sorun oluşturmaktadır. Bu çalışmada, rüzgar santrallerinin planlanması ve uygulanabilirlik çalışmaları için rüzgar hız tahminini daha kolay sağlayabilecek derin öğrenme temelli bir yaklaşım önerilmiştir. Bu yaklaşımda öncelikle rüzgar hız zaman verileri sürekli dalgacık dönüşümü ile renkli görüntüye dönüştürüldü. Elde edilen görüntüler, önceden eğitilmiş AlexNet CNN modeline uygulanarak rüzgar hız tahmini gerçekleştirilmektedir. Çalışma, Elazığ meteoroloji bölge müdürlüğünden alınan 2018-2019 yılları arasındaki saatlik hız verileri kullanılmıştır. Yapılan deneysel çalışmalarda, 1-saat, 2-saat ve 3-saat olmak üzere üç farklı ileri ufuk tahmini yapılmıştır. Önerilen tahmin modelinin modelin performans değerlendirilmesi için ortalama mutlak hata (MAE), ortalama karekök hatası (RMSE) ve korelasyon katsayısı (R) metrikleri kullanılmıştır. Deneysel çalışmalarda, tüm veri seti görüntüleri transfer öğrenimi için rastgele bir şekilde sırasıyla %70, %10 ve %20 oranlarında eğitim, doğrulama ve test olmak üzere üç bölüme ayrılmıştır. 1-saat ileri tahminde RMSE, MAE ve R metrikleri için sırasıyla 0,0335, 0,0275 ve 0,9517 deneysel sonuçlar ile en iyi rüzgar hız tahmini gerçekleştirilmiştir. Bu bakımdan önerilen AlexNet modelinde, 1 saat ileri tahmininde, daha güvenilir ve doğru tahmin gerçekleştirdiğinden rüzgar hız tahmininde etkili bir model olduğunu göstermektedir.

Keywords

Rüzgar hız tahmini, Sürekli dalgacık dönüşümü, Evrişimli sinir ağları, Derin öğrenme

Short-Term Wind Speed Forecasting With Deep Learning

Abstract

Wind speed forecasting is crucial for planning and investing in renewable wind energy. In addition, Wind speed forecasting is essential for planning renewable wind energy, optimizing wind power production, and enhancing transmission line capacities. However, intermittent and stochastic fluctuations of wind speed pose a significant problem for high quality wind speed forecasting. In this study, a deep learning-based approach is proposed for wind power plant planning and feasibility studies that can provide wind speed prediction more easily. In this approach, firstly, wind speed time data were converted into color images using continuous wavelet transform. The obtained images were applied to the pre-trained AlexNet CNN model and wind speed prediction was performed. In the study, hourly speed data from the Elazig meteorology regional directorate between 2018-2019 were used. In the experimental studies, three different horizon forecasts were made: 1-hour, 2-hour and 3-hour. Metrics like correlation coefficient (R), mean absolute error (MAE), and root means square error (RMSE) were utilized to assess the proposed forecasting models performance. In the experimental studies, the whole dataset images were randomly divided into three parts as training, validation and test at the rates of 70%, 10% and 20% respectively for transfer learning. In the 1-hour horizon forecast, the best wind speed prediction was achieved with experimental results of 0.0335, 0.0275 and 0.9517 for RMSE, MAE and R metrics, respectively. In this respect, the proposed AlexNet model shows that it is an effective model in wind speed forecast since the 1-hour horizon forecast is more reliable and accurate.

Keywords

Wind speed forecasting, Continuous wavelet transform, Convolutional neural networks, Deep learning

Ethical Statement

Ethics committee permission is not required for this study. Additionally, there is no conflict of interest with any person or institution.

Thanks

Elazığ Meteoroloji Müdürlüğü

References

• S. H. Yoon, S. Y. Kim, G. H. Park, Y. K. Kim, C. H. Cho, and B. H. Park, "Multiple power-based building energy management system for efficient management of building energy," Sustain. Cities Soc., vol. 42, pp. 462–470, August 2018.
• J. Zhang, C. Cheng, and S. Yu, "Recognizing the mapping relationship between wind power output and meteorological information at a province level by coupling GIS and CNN technologies," Appl. Energy, vol. 360, p. 122791, April 2024.
• H. Wang, Z. Tan, Y. Liang, F. Li, Z. Zhang, and L. Ju, "A novel multi-layer stacking ensemble wind power prediction model under Tensorflow deep learning framework considering feature enhancement and data hierarchy processing," Energy, vol. 286, 2024.
• D. Niu, M. Yu, L. Sun, T. Gao, and K. Wang, "Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism," Appl. Energy, vol. 313, 2022.
• C. Yildiz, H. Acikgoz, D. Korkmaz, and U. Budak, "An improved residual-based convolutional neural network for very short-term wind power forecasting," Energy Convers. Manag., vol. 228, 2021.
• F. Demir, and B. Taşcı, "Predicting the power of a wind turbine with machine learning-based approaches from wind direction and speed data," in 2021 International Conference on Technology and Policy in Energy and Electric Power (ICT-PEP), pp. 37–40, 2021.
• R. Azimi, M. Ghofrani, and M. Ghayekhloo, "A hybrid wind power forecasting model based on data mining and wavelets analysis," Energy Convers. Manag., vol. 127, pp. 208–225, 2016.
• K. U. Jaseena and B. C. Kovoor, "Decomposition-based hybrid wind speed forecasting model using deep bidirectional LSTM networks," Energy Convers. Manag., vol. 234, 2021.
• L. Li, Y. Li, B. Zhou, Q.Wu, X. Shen, H. Liu, Z. Gong, "An adaptive time-resolution method for ultra-short-term wind power prediction," Int. J. Electr. Power Energy Syst., vol. 118, 2020.
• F.Noman, G.Alkawsi, A.A. Alkahtani, A.Q. Al-Shetwi, S. K.Tiong, N. Alalwan, J. Ekanayake, A. I. Alzahrani, "Multistep short-term wind speed prediction using nonlinear auto-regressive neural network with exogenous variable selection," Alexandria Eng. J., vol. 60, no. 1, pp. 1221–1229, 2021.
• A. Zameer, J. Arshad, A. Khan, and M. A. Z. Raja, "Intelligent and robust prediction of short-term wind power using genetic programming based ensemble of neural networks," Energy Convers. Manag., vol. 134, pp. 361–372, 2017.
• A. Altan, S. Karasu, and E. Zio, "A new hybrid model for wind speed forecasting combining long short-term memory neural network, decomposition methods and grey wolf optimizer," Appl. Soft Comput., vol. 100, 2021.
• C. Emeksiz and M. Tan, "Wind speed estimation using novelty hybrid adaptive estimation model based on decomposition and deep learning methods (ICEEMDAN-CNN)," Energy, vol. 249, 2022.
• M. A. Gabry, A. Gharieb, M. Y. Soliman, I. Eltaleb, S. M. Farouq-Ali, and C. Cipolla, "Advanced Deep Learning for microseismic events prediction for hydraulic fracture treatment via Continuous Wavelet Transform," Geoenergy Sci. Eng., vol. 239, 2024.
• Q. Yu et al., "Cement pavement void detection algorithm based on GPR signal and continuous wavelet transform method," Sci. Rep., vol. 13, no. 1, 2023.
• M. Aslan, "CNN based efficient approach for emotion recognition," J. King Saud Univ. - Comput. Inf. Sci., vol. 34, no. 9, pp. 7335–7346, 2022.
• M. S. Jrad, A. E. Oueslati, and Z. Lachiri, "Image segmentation based thresholding technique: Application to DNA sequence scalograms," in 2016 2nd International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), pp. 319–324, 2016.
• B. Ari et al., "Wavelet ELM-AE based data augmentation and deep learning for efficient emotion recognition using EEG recordings," IEEE Access, vol. 10, pp. 72171–72181, 2022.
• R. S. Salles et al., "Visualization of quality performance parameters using wavelet scalograms images for power systems," in Congresso Brasileiro de Automática-CBA, vol. 2, no. 1, 2020.
• A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," in Advances in Neural Information Processing Systems, pp. 1097–1105, 2012.
• A. S. Nandhini, R. M. Aiyer, V. P. Kumar, and E. M. R. Nithin, "Pancreases segmentation and classification based on RCNN and AlexNet," in EAI/Springer Innovations in Communication and Computing, pp. 457–466, 2022.
• H. C. Shin et al., "Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning," IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1285–1298, 2016.
• N. V. Sridharan, S. Vaithiyanathan, and M. Aghaei, "Voting based ensemble for detecting visual faults in photovoltaic modules using AlexNet features," Energy Reports, vol. 11, pp. 3889–3901, 2024.
• T. Lu, B. Han, and F. Yu, "Detection and classification of marine mammal sounds using AlexNet with transfer learning," Ecol. Inform., vol. 62, 2021.
• X. Chen, T. Sun, X. Lai, Y. Zheng, and X. Han, "Transfer learning strategies for lithium-ion battery capacity estimation under domain shift differences," J. Energy Storage, vol. 90, 2024.
• E. N. Zurel, Z. M. Alçin, and M. Aslan, "Konutlardaki elektrikli cihazların evrişimli sinir ağı ile otomatik sınıflandırılması," Gazi Üniversitesi Fen Bilim. Derg. Part C Tasarım ve Teknol., vol. 10, no. 4, pp. 940–952, 2022.
• E. S. Olivas, J. D. M. Guerrero, M. Martinez Sober, J. R. Magdalena Benedito, and A. J. Serrano López, Handbook of research on machine learning applications and trends: Algorithms, methods, and techniques, 2009, pp. 1–703.
• H. Acikgoz, U. Budak, D. Korkmaz, and C. Yildiz, "WSFNet: An efficient wind speed forecasting model using channel attention-based densely connected convolutional neural network," Energy, vol. 233, 2021.