A wind power prediction method based on CEEMDAN- LightGBM-Transformer

Authors

DOI:

https://doi.org/10.4108/ew.14761

Keywords:

Power prediction, Mode decomposition, Gradient boosting, Self attention mechanism

Abstract

In the development of green energy, wind power, as the core clean energy, has the inherent intermittency, volatility and randomness of its output, which is very easy to cause the power imbalance of the energy system and the surge of grid accommodation pressure, seriously restricting the safe and stable operation of the energy system.  This study proposes the CEEMDAN-LightGBM-Transformer model. First, Complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) decomposes the data into multiple intrinsic mode function(IMF), and uses the "drop_k" strategy to select IMF components with prominent feature information and low redundancy, effectively stripping the noise and invalid fluctuations in wind power output data, and consolidating the foundation for accurate analysis of energy data. Then, Light gradient boosting machine(LightGBM) is used to construct nonlinear features, which are combined with strong correlation linear features to rank the importance of features, and accurately explore the correlation between wind power output and multidimensional factors. Finally, the multi-dimensional features are fused using a Transformer with an Encoder-Decoder architecture to generate a low-error power prediction sequence. The experimental results show that compared with multiple classical baseline models such as random forest, the mean value of Root Mean Squared Error (mRMSE) and mean value of Mean Absolute Error(mMAE) of this model are reduced by 61%, and the R² value is more than 98%, which has significant advantages in prediction performance. In addition, CEEMDAN, LightGBM and Transformer have shown good synergy in enhancing the performance of the model, which can effectively cope with the complex dynamic operating characteristics of wind power systems. The accurate prediction model can effectively stabilize the fluctuation risk of energy system caused by wind power integration, and provide reliable technical support for the grid dispatching of new power system.

Downloads

Download data is not yet available.

References

[1] Tuncar, E.A.; Sağlam, Ş.; Oral, B. A review of short-term wind power generation forecasting methods in recent technological trends[J]. Energy Reports, 2024, 12: 197-209.

[2] Wang, G.; Zhang, Z.; Lin, J. Multi-energy complementary power systems based on solar energy: A review[J]. Renewable and Sustainable Energy Reviews, 2024, 199: 114464.

[3] Chaudhary, K.A.;Guha, D.; Negi, R. Dynamic event-triggered recurrent neural network framework for synchronized frequency control of delayed power system[J].Journal of the Franklin Institute,2026,363(14): 108961-108961.

[4] Alsamamra, H.R.; Salah, S.; Shoqeir, J.H. Performance analysis of ARIMA Model for wind speed forecasting in Jerusalem, Palestine[J]. Energy Exploration & Exploitation, 2024, 42(5): 1727-1746.

[5] Shrivastava, R.; Iyer, I.S.; Batra, A. Short term wind prediction with an ARIMA model[J]. Wind Energy and Engineering Research, 2025: 100020.

[6] Abedinia, O.; Ghasemi-Marzbali, A.; Shafiei; M.; Sobhani,B.;Gharehpetian,B.G. A multi-level model for hybrid short term wind forecasting based on SVM, wavelet transform and feature selection[C]//2022 IEEE international conference on environment and electrical engineering and 2022 IEEE industrial and commercial power systems europe (EEEIC/I&CPS europe). IEEE, 2022: 1-6.

[7] Benbouhenni, H.; Colak, I.; Bizon, N. Backstepping control of multilevel modified SVM inverter in variable speed DFIG-based dual-rotor wind power system[J].Transactions of the Institute of Measurement and Control,2025,47(10): 1917-1930.

[8] Sias, Q.A.; Gantassi, R.; Choi, Y.; Bae, J.H. Recurrence Multilinear Regression Technique for Improving Accuracy of Energy Prediction in Power Systems. Energies 2024, 17, 5186.

[9] Mustaffa, Z.; Sulaiman, M.H. Random forest based wind power prediction method for sustainable energy system[J]. Cleaner Energy Systems, 2025: 100210.

[10] Jamii, J.; Trabelsi, M.; Mansouri, M.; Kouadri, A.; Mimouni,F.M.; Nounou, M. Medium-term wind power forecasting using reduced principal component analysis based random forest model[J]. Wind Engineering, 2024, 48(4): 597-616.

[11] Zhang, Y.; Kong, X.; Wang, J.; Wang, H.;Cheng, X. Wind power forecasting system with data enhancement and algorithm improvement[J]. Renewable and Sustainable Energy Reviews, 2024, 196: 114349.

[12] Mo, S.; Wang, H.; Li, B.; Xue, Z.; Fan, S.; Liu, X. Powerformer: A temporal-based transformer model for wind power forecasting[J]. Energy Reports, 2024, 11: 736-744.

[13] Karijadi, I.; Chou, S.Y.; Dewabharata, A. Wind power forecasting based on hybrid CEEMDAN-EWT deep learning method[J]. Renewable Energy, 2023, 218: 119357.

[14] Li, Q.; Ren, X.; Zhang, F.; Guo, L.; Hao, B. A novel ultra-short-term wind power forecasting method based on TCN and Informer models[J]. Computers and Electrical Engineering, 2024, 120: 109632.

[15] Woldegiyorgis, A. T.;Chaka, D.M.;Asmare, E.;Asmare, E.; Admassu, C.;Anose, A.F. Assessment of artificial neural network and Weibull distribution methods for wind power density prediction in Ethiopia[J].Environmental Research Communications,2026,8(8):085006-085006.

[16] Li, N.; Dong, J.; Liu, L.; Li, H.; Yan, J. A novel EMD and causal convolutional network integrated with Transformer for ultra short-term wind power forecasting[J]. International Journal of Electrical Power & Energy Systems, 2023, 154: 109470.

[17] Qazi, A. M.;Chiang, D. H.Gated recurrent neural network enhanced wind power prediction accuracy using TPE Bayesian optimization[J].Results in Engineering,2026,PP: 32112136-112136.

[18] Hussain, R.M.S.; Baig, A.M.M; Yousuf, U.M .A modular based framework for multi-step spatio-temporal wind power forecasting [J]. Energy Strategy Reviews,2026,65:102203.

[19] Ahsan, M.; Dao, B.P.Multi-Channel SCADA-Based Image-Driven Power Prediction for Wind Turbines Using Optimized LeNet-5-LSTM Hybrid Neural Architecture[J]. Energies,2026,19(5):1169-1169.

[20] Venkateswaran, R.; Venkateswaran, D.; Choi,W.;Jeong, J.H.Hybrid CEEMDAN–Transformer–LSTM Framework for enhanced multi-horizon wind speed forecasting[J].Journal of Wind Engineering and Industrial Aerodynamics, 2026, 272:106408-106408.

[21] Su, Y.; Wang, Z.; Dong, Z.; Hua, X.; Ye, T.; Song, Z.; Shao, Y. Frequency-aware ultra-short-term wind power forecasting using CEEMDAN–VMD–SE and Transformer–GRU networks[J].Energy,2025,338138715-138715.

[22] Lu, X.; Shao, J.; Lai, K. K.; Shi,Y.;.Chen, Q.; Cui, H. Bitcoin price forecasting with a deep hybrid EMD-CNN-GRU model[J].Soft Computing, 2026, 30(1):561-580.

[23] Randrianirina, S. F. M. J.; Rakotomalala, F. F. L.; Andriamahitasoa, A. B.;Bernard, A.A.;Zely, A.R.;Liva,G.R.Optimizing Wind Power Forecasting Using Machine Learning: A Comparative Study with Emphasis on LightGBM for Predictive Maintenance[J].International Journal of Advanced Engineering Research and Science, 2025,12(6):69-80.

[24] Kopyt, M.; Piotrowski, P.; Baczyński, D. Short-Term Energy Generation Forecasts at a Wind Farm—A Multi-Variant Comparison of the Effectiveness and Performance of Various Gradient-Boosted Decision Tree Models.Energies,2024,17, 6194.

[25] Oriol, T.; Cifuentes, J.;Marulanda, G. Enhancing Wind Power Forecasting in the Spanish Market Through Transformer Neural Networks and Temporal Optimization[J]. Sustainability,2025,17(19):8655-8655.

[26] Wang, Q.; Wang, Y.; Zhang, K.; Liu, Y.; Qiang, W.; Han Wen, Q. Artificial Intelligent Power Forecasting for Wind Farm Based on Multi-Source Data Fusion[J].Processes, 2023,11, 1429.

[27] Sun, G.; Zhao, Y.; Li, J.; Cui, F.; Qi, X. DG-LSTM-SA model: A deep gated LSTM network with self-attention mechanism for power generation and load forecasting[J]. PloS one, 2026, 21(6): e0350071.

[28] Jing, S.; Xi, X.; Su, D.; Han, Z.; Wang, D. Spatio-Temporal Photovoltaic Power Prediction with Fourier Graph Neural Network[J]. Electronics, 2024, 13, 4988.

[29] Zi, X.; Liu, F.; Liu, M.; Wang, Y. Transformer with Adaptive Sparse Self-Attention for Short-Term Photovoltaic Power Generation Forecasting[J]. Electronics, 2025, 14(20): 3981.

[30] Pan, C.; Liu, Y.; Oh, Y.; Lim, C. Short-Term Photovoltaic Power Forecasting Using PV Data and Sky Images in an Auto Cross Modal Correlation Attention Multimodal Framework[J]. Energies, 2024,17, 6378.

[31] Taheri, N.;Tucci, M. Enhancing Regional Wind Power Forecasting through Advanced Machine-Learning and Feature-Selection Techniques[J].Energies,2024,17(21): 5431-5431.

Downloads

Published

16-11-2026

Issue

Section

Digital Twin Technologies for Smart Energy Systems

How to Cite

1.
Zhang J, Zhou Y, Chu X, Yang Y, Xu Y. A wind power prediction method based on CEEMDAN- LightGBM-Transformer. EAI Endorsed Trans Energy Web [Internet]. 2026 Nov. 16 [cited 2026 Nov. 16];13. Available from: https://publications.eai.eu/index.php/ew/article/view/14761