Application of Deep Learning Models in Wind Farm Power Output Forecasting

Authors

https://doi.org/10.48314/tsc.v1i2.49

Abstract

Wind energy stands out as one of the fastest-growing clean energy sources worldwide, driven by global commitments to decarbonization, enhanced energy security, and sustainable development. Accurate forecasting of wind farm power output thus becomes a key prerequisite for better harnessing renewables and optimizing grid management. The non-stationary, stochastic, and multi-scale nature of wind speed and direction poses serious limitations for traditional forecasting models. In recent years, the rise of deep learning has sparked a major revolution in time series prediction, particularly outperforming classical statistical and machine learning methods in wind energy forecasting. This paper analyzes recent research on wind power prediction, focusing on deep learning architectures, spatial-temporal models, physics-informed approaches, and hybrid designs. It also reviews data sources such as Supervisory Control And Data Acquisition (SCADA), Meteorological Data (Meteo), Numerical Weather Prediction (NWP), remote sensing, and multi-source structures. Beyond offering a conceptual classification, quantitative and qualitative comparisons of methods, analysis of research challenges, computational complexity reviews, and identification of knowledge gaps, the study outlines future research directions. Findings reveal that attention-based models excel at capturing inter-turbine relationships, while decomposition-driven hybrids like Wavelet–LSTM, VMD–BiLSTM, and EMD–GRU deliver the highest accuracy and stability. Persistent challenges include data heterogeneity, spatial-temporal wind variations, the need for probabilistic forecasts, lack of interpretability, and demand for lightweight architectures.      

Keywords:

Wind power forecasting, Deep learning, Decomposition methods, Numerical weather prediction

References

  1. [1] Schaffarczyk, A. P. (2024). Introduction to wind turbine aerodynamics. Springer Nature. https://doi.org/10.1007/978-3-031-56924-1
  2. [2] Wang, X., Guo, P., & Huang, X. (2011). A review of wind power forecasting models. Energy procedia, 12, 770–778. https://doi.org/10.1016/j.egypro.2011.10.103
  3. [3] Maradin, D. (2021). Advantages and disadvantages of renewable energy sources utilization. International journal of energy economics and policy, 11(3), 176–183. https://doi.org/10.32479/ijeep.11027
  4. [4] Lange, M., & Focken, U. (2006). Physical approach to short-term wind power prediction (Vol. 208). Springer. https://doi.org/10.1007/3-540-31106-8%0A%0A
  5. [5] Shadab, A., Said, S., & Ahmad, S. (2019). Box Jenkins multiplicative ARIMA modeling for prediction of solar radiation: A case study. International journal of energy and water resources, 3(4), 305–318. https://doi.org/10.1007/s42108-019-00037-5
  6. [6] Naderipour, M., Hosseini, H. A.-S., & Jamali, M. B. (2025). A novel model based on the ARIMA model in predicting stock prices of Tehran Stock Exchange companies. System engineering and productivity, 6(3), e732324. https://doi.org/10.22034/sep.2025.2075221.1411
  7. [7] Cassola, F., & Burlando, M. (2012). Wind speed and wind energy forecast through Kalman filtering of numerical weather prediction model output. Applied energy, 99, 154–166. https://doi.org/10.1016/j.apenergy.2012.03.054
  8. [8] Grigonytė, E., & Butkevičiūtė, .E. (2016). Short-term wind speed forecasting using ARIMA model. Energetika, 62(1–2), 45–55. https://doi.org/10.6001/energetika.v62i1-2.3313
  9. [9] Jiao, J. (2018). A hybrid forecasting method for wind speed. MATEC web of conferences (Vol. 232, p. 3013). EDP Sciences. https://doi.org/10.1051/matecconf/201823203013
  10. [10] Xu, P., Zhang, M., Chen, Z., Wang, B., Cheng, C., & Liu, R. (2023). A deep learning framework for day ahead wind power short-term prediction. Applied sciences, 13(6), 4042. https://doi.org/10.3390/app13064042
  11. [11] Vishnutheerth, E. P., Vijay, V., Satheesh, R., & Kolhe, M. L. (2024). A comprehensive approach to wind power forecasting using advanced hybrid neural networks. IEEE access, 12, 124790–124800. https://doi.org/10.1109/ACCESS.2024.3450096
  12. [12] Tang, J., Yue, G., Lv, J., & Yue, J. (2025). An ensemble learning model for ultra-short-term wind power prediction based on variational mode decomposition and adaptive weighting. Natural resource modeling, 38(4), e70015. https://doi.org/10.1111/nrm.70015
  13. [13] Daryabi, S., Srujana, N. D., & Rama Sudha, K. (2025). Partial least squares enhanced long short-term memory models for wind power forecasting. Intelligent strategies for ICT (pp. 325–338). Singapore: Springer Nature Singapore. https://doi.org/10.1007/978-981-96-5607-3_29
  14. [14] Shi, J., Teh, J., & Lai, C. M. (2025). Wind power prediction based on improved self-attention mechanism combined with bi-directional temporal convolutional network. Energy, 322, 135666. https://doi.org/10.1016/j.energy.2025.135666
  15. [15] Peng, H., Sun, H., Luo, S., Zuo, Z., Zhang, S., Wang, Z., & Wang, Y. (2024). Diffusion-based conditional wind power forecasting via channel attention. IET renewable power generation, 18(3), 306–320. 10.1049/rpg2.12825
  16. [16] Liu, R., Song, Y., Yuan, C., Wang, D., Xu, P., & Li, Y. (2023). GAN-based abrupt weather data augmentation for wind turbine power day-ahead predictions. Energies, 16(21). https://doi.org/10.3390/en16217250
  17. [17] Dev, U., Ahmad, S., Uddin, M. R., Dhar, R., Mubarak, H., Hazari, M. R., … & Alshammari, O. (2025). Short-term wind speed forecasting using different lstm deep learning methods. Innovations in electrical and electronics engineering (pp. 199–211). Singapore: Springer Nature Singapore. https://doi.org/10.1007/978-981-97-9112-5_12
  18. [18] Karakan, A. (2024). Predicting energy production in renewable energy power plants using deep learning. Energies, 17(16), 4031. https://doi.org/10.3390/en17164031
  19. [19] Elsaraiti, M., & Merabet, A. (2021). Application of long-short-term-memory recurrent neural networks to forecast wind speed. Applied sciences, 11(5), 2387. https://doi.org/10.3390/app11052387
  20. [20] Huang, S., Yan, C., & Qu, Y. (2023). Deep learning model-transformer based wind power forecasting approach. Frontiers in energy research, 10, 1055683. https://doi.org/10.3389/fenrg.2022.1055683
  21. [21] Wang, W., Yang, J., Li, Y., Ren, G., & Li, K. (2025). Data-driven deep learning model for short-term wind power prediction assisted with WGAN-GP data preprocessing. Expert systems with applications, 275, 127068. https://doi.org/10.1016/j.eswa.2025.127068
  22. [22] Liu, Z., Gao, W., Wan, Y. H., & Muljadi, E. (2012). Wind power plant prediction by using neural networks. 2012 IEEE energy conversion congress and exposition (ECCE) (pp. 3154–3160). IEEE. https://doi.org/10.1109/ECCE.2012.6342351
  23. [23] Wang, R., Qiu, H., Jiang, G., Liu, X., & Cheng, X. (2024). Class-imbalanced spatial--temporal feature learning for blade icing recognition of wind turbine. IEEE transactions on industrial informatics, 20(8), 10249–10258. https://doi.org/10.1109/TII.2024.3393550
  24. [24] Zhang, J., Zhao, Z., Yan, J., & Cheng, P. (2023). Ultra-short-term wind power forecasting Based on CGAN-CNN-LSTM model supported by Lidar. Sensors, 23(9), 4369. https://doi.org/10.3390/s23094369
  25. [25] Al-qaness, M. A. A., Ewees, A. A., Aseeri, A. O., & Abd Elaziz, M. (2024). Wind power forecasting using optimized LSTM by attraction–repulsion optimization algorithm. Ain shams engineering journal, 15(12), 103150. https://doi.org/10.1016/j.asej.2024.103150
  26. [26] Daenens, S., Verstraeten, T., Daems, P. J., Nowe, A., & & Helsen, J. (2024). Spatio-temporal graph neural networks for power prediction in offshore wind farms using scada data. Wind energy science discussions, 2024, 1–19. https://doi.org/10.5194/wes-10-1137-2025
  27. [27] Xue, L., Zhang, X., Jiang, W., Huo, K., & Shen, Q. (2023). A classification performance evaluation measure considering data separability. Artificial neural networks and machine learning ICANN 2023 (pp. 1–13). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-44207-0_1
  28. [28] Peng, S., Guo, L., Huang, H., Liu, X., & Peng, J. (2024). ForecastNet wind power prediction based on spatio-temporal distribution. Applied sciences, 14(2), 937. https://doi.org/10.3390/app14020937
  29. [29] Tang, F. (2025). Short-term wind power prediction based on improved sparrow search algorithm optimized long short-term memory with peephole connections. Wind engineering, 49(1), 71–90. https://doi.org/10.1177/0309524X241257429
  30. [30] Qureshi, S., Shaikh, F., Kumar, L., Ali, F., Awais, M., & Gürel, A. E. (2023). Short-term forecasting of wind power generation using artificial intelligence. Environmental challenges, 11, 100722. https://doi.org/10.1016/j.envc.2023.100722

Downloads

Published

2025-06-27

Issue

Vol. 1 No. 2 (2025): Transactions on Soft Computing

Section

Articles

How to Cite

Heydari Vahed, M., Ghazvini, M., & Ghasemian, F. (2025). Application of Deep Learning Models in Wind Farm Power Output Forecasting. Transactions on Soft Computing , 1(2), 149-167. https://doi.org/10.48314/tsc.v1i2.49

Similar Articles

You may also start an advanced similarity search for this article.