Short-Term Wind Power Load Forecasting Based on VMD-SVM
Vol. 28 (2026): 2026 2nd International Conference on Agricultural Sciences, Economics, Biomedical and Environmental Sciences (SEMBE 2026)
Received: 2026-07-18
Accepted: 2026-07-18
Published: 2026-07-18
Abstract
With the continuous global expansion of renewable wind power capacity, the accurate short-term forecasting of wind power load plays an increasingly crucial role in ensuring efficient power system dispatching, grid stability, and overall operational safety. However, due to the inherent non-stationarity, multi-scale fluctuations, and severe noise interference commonly found in real-world load time series data, traditional forecasting methods frequently fall short in terms of both predictive accuracy and systemic robustness. To effectively address this critical issue, this paper proposes a novel short-term forecasting approach based on the synergistic integration of Variational Mode Decomposition (VMD) and Support Vector Machine (SVM) algorithms. The proposed method first employs the VMD technique to systematically decompose the original complex load time series, successfully extracting essential underlying features across different frequency bands. Subsequently, dedicated SVM regression models are established for each independently decomposed mode to generate precise localized predictions. The final comprehensive load values are then obtained by carefully reconstructing the predicted modal components. Extensive experimental results based on the 2026 Kaggle Hourly Energy Consumption dataset indicate that the proposed VMD-SVM model achieves a Mean Absolute Error (MAE) of 101.43, a Root Mean Square Error (RMSE) of 126.26, and an impressive R2 of 0.9961 on the testing set. These metrics demonstrate that the hybrid approach significantly outperforms standalone SVM, Random Forest, and XGBoost models. Ultimately, this study proves that the proposed methodology substantially enhances model robustness against non-stationary load fluctuations, providing highly reliable technical support for practical operational applications in modern wind farms.
Keywords
References
[1] L. Ø. Bentsen, N. D. Warakagoda, R. Stenbro, and P. Engelstad, "Relative evaluation of probabilistic methods for spatio-temporal wind forecasting," Journal of Cleaner Production, vol. 434, p. 139944, 2024.
[2] S. Zhang, C. Zhu, and X. Guo, "Wind-speed multi-step forecasting based on variational mode decomposition, temporal convolutional network, and transformer model," Energies, vol. 17, no. 9, p. 1996, 2024.
[3] D. Li, Y. Hu, S. Miao, Z. Fang, Y. Liang, and S. He, "Adaptive temporal transformer method for short-term wind power forecasting considering shift in time series distribution," AIP Advances, vol. 14, no. 2, 2024.
[4] J. Jonkers, D. N. Avendano, G. Van Wallendael, and S. Van Hoecke, "A novel day-ahead regional and probabilistic wind power forecasting framework using deep CNNs and conformalized regression forests," Applied Energy, vol. 361, p. 122900, 2024.
[5] M. Yang, Y. Jiang, J. Che, Z. Han, and Q. Lv, "Short-Term forecasting of wind power based on error traceability and numerical weather prediction wind speed correction," Electronics, vol. 13, no. 8, p. 1559, 2024.
[6] S. Lin, J. Wang, X. Xu, H. Tan, P. Shi, and R. Li, "SWSA transformer: A forecasting method of ultra-short-term wind speed from an offshore wind farm using global attention mechanism," Journal of Renewable and Sustainable Energy, vol. 15, no. 4, 2023.
[7] Y. Sun, Y. Huang, and M. Yang, "Ultra-short-term wind power interval prediction based on fluctuating process partitioning and quantile regression forest," Frontiers in Energy Research, vol. 10, p. 867719, 2022.
[8] S. Wang, H. Liu, and G. Yu, "Short-term wind power combination forecasting method based on wind speed correction of numerical weather prediction," Frontiers in Energy Research, vol. 12, p. 1391692, 2024.
[9] T. Oriol, J. Cifuentes, and G. Marulanda, "Enhancing Wind Power Forecasting in the Spanish Market Through Transformer Neural Networks and Temporal Optimization," Sustainability, vol. 17, no. 19, p. 8655, 2025.
[10] C. Pan, S. Wen, M. Zhu, H. Ye, J. Ma, and S. Jiang, "Hedge backpropagation based online LSTM architecture for ultra-short-term wind power forecasting," IEEE Transactions on Power Systems, vol. 39, no. 2, pp. 4179–4192, 2023.
[11] Y. Zhou, Y. Sun, S. Wang, R. J. Mahfoud, H. H. Alhelou, N. Hatziargyriou, and P. Siano, "Performance improvement of very short-term prediction intervals for regional wind power based on composite conditional nonlinear quantile regression," Journal of Modern Power Systems and Clean Energy, vol. 10, no. 1, pp. 60–70, 2021.
[12] J. Huan, L. Deng, Y. Zhu, S. Jiang, and F. Qi, "Short-to-medium-term wind power forecasting through enhanced transformer and improved EMD integration," Energies, vol. 17, no. 10, p. 2395, 2024.
[13] X. Dai, G. P. Liu, and W. Hu, "An online-learning-enabled self-attention-based model for ultra-short-term wind power forecasting," Energy, vol. 272, p. 127173, 2023.
Copyright and License
Published in2026-07-18 16:13:14
DOI https://doi.org/10.70088/yvm55r92
Creative Commons
Copyright: © 2026 by the authors.
Submitted for possible open access
publication under the terms and
conditions of the Creative Commons
Attribution (CC BY) license
(https://creativecommons.org/license
s/by/4.0/).
Copyright
Copyright © The Author(s), 2026. Published by SEMBE 2026
Journal Information
- Vol. 28 (2026): 2026 2nd International Conference on Agricultural Sciences, Economics, Biomedical and Environmental Sciences (SEMBE 2026)
- 2026-07-18
- ISSN: (Print) 3078-770X/ (Online) 3078-7718
- Journal Homepage