Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Advances in Computer, Signals and Systems
  4. Vol 9, Issue 3, 2025

Ultra-short-term Wind Power Forecasting Based on ICEEMDAN-Informed BiGRU Network with Multi-head Attention

Download as PDF

DOI: 10.23977/acss.2025.090310 | Downloads: 0 | Views: 387

Author(s)

Daoyuan Li 1, Yutang Ye 1

Affiliation(s)

1 College of Engineering, Northeast Agricultural University, Harbin, Heilongjiang, China

Corresponding Author

Daoyuan Li

ABSTRACT

To improve the ultra-short-term forecasting accuracy of wind power series, this paper proposes a combined forecasting model—ICEEMDAN-Informer-BiGRU-Attention—that integrates improved modal decomposition with deep learning modeling to address its strong non-stationarity and complex temporal structure. This model first decomposes the wind power series using an improved ensemble empirical mode decomposition (ICEEMDAN) algorithm to obtain multiple subsequences with clear frequency domain features and reduced volatility. It then employs an informer architecture to capture long-term dependencies within the series, introduces a bidirectional gated recurrent neural network (BiGRU) to model short-term dynamic features, and incorporates a multi-head self-attention mechanism to further enhance the representation of key time steps. Experimental results on a real wind farm dataset demonstrate that the proposed model outperforms mainstream models such as LSTM, GRU, and informer in terms of RMSE, MAE, and R². Its predictions are closer to real data, with more concentrated residual distributions, resulting in optimal overall performance. This validates the model's effectiveness and engineering feasibility in wind power forecasting scenarios.

KEYWORDS

Wind Power Forecasting; ICEEMDAN; Informer; BiGRU; Multi-head Attention Mechanism; Deep Learning

CITE THIS PAPER

Daoyuan Li, Yutang Ye, Ultra-short-term Wind Power Forecasting Based on ICEEMDAN-Informed BiGRU Network with Multi-head Attention. Advances in Computer, Signals and Systems (2025) Vol. 9: 79-87. DOI: http://dx.doi.org/10.23977/acss.2025.090310.

REFERENCES

[1] J. Wu, S. Li, J. C. Vasquez, and J. M. Guerrero, "A bi-level mode decomposition framework for multi-step wind power forecasting using deep neural network," Energy Convers. Manage.: X, vol. 23, p. 100650, Jul. 2024.
[2] S. Ullah, "A novel hybrid ensemble approach for wind speed forecasting with dual-stage decomposition strategy using optimized GRU and transformer models," Energy, vol. 329, p. 136739, Aug. 2025.
[3] 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, p. 131546, Aug. 2024.
[4] Q. Li, X. Ren, F. Zhang, L. Gao, and B. Hao, "A novel ultra-short-term wind power forecasting method based on TCN and informer models," Comput. Electr. Eng., vol. 120, p. 109632, Dec. 2024.
[5] D. Yang, M. Li, J. Guo, and P. Du, "An attention-based multi-input LSTM with sliding window-based two-stage decomposition for wind speed forecasting," Appl. Energy, vol. 375, p. 124057, Dec. 2024.
[6] Y. Xiao, S. Wu, C. He, Y. Hu, and M. Yi, "An effective hybrid wind power forecasting model based on 'decomposition-reconstruction-ensemble' strategy and wind resource matching," Sustainable Energy Grids Networks, vol. 38, p. 101293, Jun. 2024.
[7] X. Wu, D. Wang, M. Yang, and C. Liang, "CEEMDAN-SE-HDBSCAN-VMD-TCN-BiGRU: a two-stage decomposition-based parallel model for multi-altitude ultra-short-term wind speed forecasting,” Energy, vol. 330, p. 136660, Sep. 2025.
[8] D. Korkmaz, H. Acikgoz, and M. Ustundag, "Cross-scale fusion network and two-stage decomposition for power forecasting of offshore wind turbines," Ocean Eng., vol. 341, p. 122541, Dec. 2025.
[9] B. S. Bommidi, K. Teeparthi, and V. Kosana, "Hybrid wind speed forecasting using ICEEMDAN and transformer model with novel loss function," Energy, vol. 265, p. 126383, Feb. 2023.
[10] Z. J. Wu, Y. Dong, and P. He, "ICEEMDAN-based combined wind power forecasting," Recent Pat. Eng., vol. 19, no. 3, Jan. 2025.
[11] M. Wang, F. Ying, and Q. Nan, "Refined offshore wind speed prediction: leveraging a two-layer decomposition technique, gated recurrent unit, and kernel density estimation for precise point and interval forecasts," Eng. Appl. Artif. Intell., vol. 133, p. 108435, Jul. 2024.
[12] M. Gong, "Short-term wind power forecasting model based on temporal convolutional network and informer," Energy, vol. 283, p. 129171, Nov. 2023.
[13] S. Sun, P. Yu, J. Xing, Y. Cheng, S. Yang, and Q. Ai, "Short-term wind power prediction based on ICEEMDAN-SE-LSTM neural network model with classifying seasonal," Energy Eng., vol. 120, no. 12, pp. 2761–2782, Nov. 2023.
[14] M. Hu, G. Zheng, Z. Su, L. Kong, and G. Wang, "Short-term wind power prediction based on improved variational modal decomposition, least absolute shrinkage and selection operator, and BiGRU networks," Energy, vol. 303, p. 131951, Sep. 2024.
[15] P. Dhaka, M. Sreejeth, and M. M. Tripathi, "Wind power forecasting using multivariate signal decomposition and stacked GRU ensembles with error correction," Future Gener. Comput. Syst., p. 108105, Aug. 2025.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 38552
Visits: 697758

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.