Wind Power Prediction Correction and Climbing Characteristics Based on the Combination of XGBoost and LSTM
DOI:
https://doi.org/10.13052/dgaej2156-3306.4044Keywords:
Wind power forecasting, XGBoost, LSTM, error correction, climbing event detectionAbstract
In wind power forecasting, Existing methods face dual challenges, specifically manifested as insufficient modeling capabilities for complex time series features, low prediction accuracy of climbing events, and limited model generalization capabilities. Facing the problem of insufficient accuracy of traditional prediction methods when facing nonlinear and abrupt power changes, a hybrid prediction correction model combining XGBoost and LSTM is proposed, considering the capabilities of XGBoost in static feature extraction and LSTM in dynamic time series modeling. Construct a multi-modal feature fusion architecture, quantitatively evaluate the importance of input features through XGBoost, assist LSTM in capturing key inputs, and improve the model’s ability to perceive wind power changing trends. A power sudden change point detection algorithm based on phase space reconstruction is designed to identify wind power climbing and downclimbing events accurately. By establishing a coupling model between power change rate and grid frequency response, the system’s response prediction ability to power sudden change caused by grid disturbance is strengthened. Combining feature importance-driven error pattern recognition with LSTM memory cell state compensation strategy, the model output is dynamically adjusted, and the cumulative error is significantly reduced. The traditional LSTM model does not capture the long-term dependence characteristics of wind power series enough, resulting in the root mean square error (RMSE) of ultra-short-term prediction of 13.5 kW⋅h and the false report rate of climbing events of 42.7%, Although the single XGBoost model can deal with nonlinear features, there is a lag in time series dynamic modeling, the prediction accuracy rate is only 76.2%, and the recognition rate of extreme climbing events with slope exceeding 24.8 kW⋅h/10 min is less than 59.3%. In contrast, the measured data in a certain area shows that the prediction error rate of the model without the correction strategy soared to 22.3% in strong turbulent weather, while the model initially combining XGBoost and LSTM reduced the RMSE to 7.2 kW.h. The climbing event capture rate increased to 88%, and the extreme scene prediction accuracy reached 91.6%, showing the potential for collaborative processing of complex features.
Downloads
References
T. Ahilan, G. Sujesh, and K. Yarrapragada, “Wind turbine power prediction via deep neural network using hybrid approach,” Proceedings of the Institution of Mechanical Engineers Part a-Journal of Power and Energy, vol. 237, no. 3, pp. 484–494, 2023.
J. Q. An, F. Yin, M. Wu, J. H. She, and X. Chen, “Multisource Wind Speed Fusion Method for Short-Term Wind Power Prediction,” Ieee Transactions on Industrial Informatics, vol. 17, no. 9, pp. 5927–5937, 2021.
W. J. Chen et al., “Ultra-Short-Term Wind Power Prediction Based on Bidirectional Gated Recurrent Unit and Transfer Learning,” Frontiers in Energy Research, vol. 9, 2021.
X. J. Chen, X. Q. Zhang, M. Dong, L. S. Huang, Y. Guo, and S. Y. He, “Deep Learning-Based Prediction of Wind Power for Multi-turbines in a Wind Farm,” Frontiers in Energy Research, vol. 9, 2021.
F. X. Dong, S. Y. Ju, J. F. Liu, D. R. Yu, and H. Li, “An ultra-short-term wind power robust prediction method considering the periodic impact of wind direction,” Renewable Energy, vol. 247, 2025.
Y. Y. He and Y. Wang, “Short-term wind power prediction based on EEMD-LASSO-QRNN model,” Applied Soft Computing, vol. 105, 2021.
Y. Huang, X. X. Li, D. Li, Z. S. Zhang, T. W. Yin, and H. T. Chen, “Probabilistic prediction of wind farm power generation using non-crossing quantile regression,” Control Engineering Practice, vol. 156, 2025.
Y. Y. Jia, B. X. Ren, Q. Li, C. G. Wang, D. J. Wang, and X. M. Zou, “An Integrated Scheme for Forecasting and Controlling Ramps in Offshore Wind Farms Considering Wind Power Uncertainties during Extreme Storms,” Electronics, vol. 12, no. 21, 2023.
T. Kari, S. Guoliang, L. Kesong, M. Xiaojing, and W. Xian, “Short-Term Wind Power Prediction Based on Combinatorial Neural Networks,” Intelligent Automation and Soft Computing, vol. 37, no. 2, pp. 1437–1452, 2023.
C. D. Li, M. H. Zhang, Y. Zhang, Z. Y. Yi, and H. Q. Niu, “Ultra-Short-Term Wind Farm Power Prediction Considering Correlation of Wind Power Fluctuation,” Sensors, vol. 24, no. 20, 2024.
D. Q. Li, X. D. Yu, S. L. Liu, X. Dong, H. Z. Zang, and R. Xu, “Wind power prediction based on PSO-Kalman,” Energy Reports, vol. 8, pp. 958–968, 2022.
F. Li, M. G. Zhang, Y. Yu, and S. Q. Li, “Deep Belief Network-Based Hammerstein Nonlinear System for Wind Power Prediction,” Ieee Transactions on Instrumentation and Measurement, vol. 73, 2024.
S. Li, L. L. Huang, Y. Liu, and M. Y. Zhang, “Modeling of Ultra-Short Term Offshore Wind Power Prediction Based on Condition-Assessment of Wind Turbines,” Energies, vol. 14, no. 4, 2021.
X. Y. Li, Y. N. Qiu, Y. H. Feng, and Z. Wang, “Wind turbine power prediction considering wake effects with dual laser beam LiDAR measured yaw misalignment,” Applied Energy, vol. 299, 2021.
H. Liu and Z. J. Zhang, “Development and trending of deep learning methods for wind power predictions,” Artificial Intelligence Review, vol. 57, no. 5, 2024.
Y. K. Liu, Y. Gu, Y. W. Long, Q. Y. Zhang, Y. G. Zhang, and X. Zhou, “Research on Physically Constrained VMD-CNN-BiLSTM Wind Power Prediction,” Sustainability, vol. 17, no. 3, 2025.
H. F. Luo, X. Dou, R. Sun, and S. J. Wu, “A Multi-Step Prediction Method for Wind Power Based on Improved TCN to Correct Cumulative Error,” Frontiers in Energy Research, vol. 9, 2021.
X. S. Peng, K. Cheng, J. X. Lang, Z. W. Zhang, T. Cai, and S. X. Duan, “Short-Term Wind Power Prediction for Wind Farm Clusters Based on SFFS Feature Selection and BLSTM Deep Learning,” Energies, vol. 14, no. 7, 2021.
X. S. Peng, C. Li, S. Y. Jia, L. S. Zhou, B. Wang, and J. F. Che, “A short-term wind power prediction method based on deep learning and multistage ensemble algorithm,” Wind Energy, vol. 25, no. 9, pp. 1610–1625, 2022.
X. S. Peng et al., “A Novel Efficient DLUBE Model Constructed by Error Interval Coefficients for Clustered Wind Power Prediction,” Ieee Access, vol. 9, pp. 61739–61751, 2021.
X. S. Peng, Z. M. Yang, Y. H. Li, B. Wang, and J. F. Che, “Short-term wind power prediction based on stacked denoised auto-encoder deep learning and multi-level transfer learning,” Wind Energy, vol. 26, no. 10, pp. 1066–1081, 2023.
X. Ran, C. Xu, L. Ma, and F. F. Xue, “Wind Power Interval Prediction with Adaptive Rolling Error Correction Based on PSR-BLS-QR,” Energies, vol. 15, no. 11, 2022.
H. Rezaie, C. H. Chung, and N. Safari, “Ensemble Wind Power Prediction Interval with Optimal Reserve Requirement,” Journal of Modern Power Systems and Clean Energy, vol. 12, no. 1, pp. 65–76, 2024.
J. N. Shan, H. Z. Wang, G. Pei, S. Zhang, and W. H. Zhou, “Research on short-term power prediction of wind power generation based on WT-CABC-KELM,” Energy Reports, vol. 8, pp. 800–809, 2022.
Z. Tang et al., “Power Prediction of Wind Farm Considering the Wake Effect and its Boundary Layer Compensation,” Protection and Control of Modern Power Systems, vol. 9, no. 6, pp. 19–29, 2024.
Y. Tominaga, “CFD Prediction for Wind Power Generation by a Small Vertical Axis Wind Turbine: A Case Study for a University Campus,” Energies, vol. 16, no. 13, 2023.
H. H. Tsao, Y. G. Leu, and L. F. Chou, “A center-of-concentrated-based prediction interval for wind power forecasting,” Energy, vol. 237, 2021.
B. Wang, T. C. Wang, M. Yang, C. Han, D. W. Huang, and D. K. Gu, “Ultra-Short-Term Prediction Method of Wind Power for Massive Wind Power Clusters Based on Feature Mining of Spatiotemporal Correlation,” Energies, vol. 16, no. 6, 2023.
D. Wang, M. Yang, and W. Zhang, “Wind Power Group Prediction Model Based on Multi-Task Learning,” Electronics, vol. 12, no. 17, 2023.
D. Wang, M. Yang, W. Zhang, C. L. Ma, and X. Su, “Short-term power prediction method of wind farm cluster based on deep spatiotemporal correlation mining,” Applied Energy, vol. 380, 2025.
