ULTRA-SHORT-TERM WIND SPEED FORECASTING BASED ON SPATIO-TEMPORAL FEATURE FUSION OF HYBRID DEEP LEARNING MODEL

Liu Zeqi; Wang Weiqing; Li Xiaozhu; Zheng Jirong

doi:10.19912/j.0254-0096.tynxb.2025-1037

PDF(1351 KB)

Acta Energiae Solaris Sinica ›› 2026, Vol. 47 ›› Issue (12) : 357-369. DOI: 10.19912/j.0254-0096.tynxb.2025-1037

ULTRA-SHORT-TERM WIND SPEED FORECASTING BASED ON SPATIO-TEMPORAL FEATURE FUSION OF HYBRID DEEP LEARNING MODEL

Liu Zeqi¹, Wang Weiqing¹, Li Xiaozhu¹, Zheng Jirong²

Author information +

History +

Abstract

This paper proposes a hybrid deep learning model integrating convolutional neural network （CNN）, bidirectional gated recurrent unit （BiGRU）, and attention mechanism for ultra-short-term wind speed forecasting. Firstly, the sliding window algorithm is adopted to reorganize the wind speed time series into multiple small-sample datasets, and a CNN is used to extract spatial features from the samples. Secondly, a BiGRU is adopted to extract temporal features from the preprocessed data. Finally, an attention mechanism is introduced to fuse and screen the spatial and temporal features of each time step, so as to realize accurate wind speed prediction. This paper conducts comparative experiments on datasets with different prediction steps and input feature combinations, as well as ablation experiments with different network layers and model structures. Furthermore, the proposed model is compared with multiple models of varying complexity to verify its effectiveness. The results demonstrate that the proposed data processing method and prediction model can achieve high-precision ultra-short-term wind speed forecasting, with low model complexity, excellent robustness and superior generalization performance.

Key words

wind forecasting / feature extraction / time series analysis / neural network models / attention mechanism

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Liu Zeqi, Wang Weiqing, Li Xiaozhu, Zheng Jirong. ULTRA-SHORT-TERM WIND SPEED FORECASTING BASED ON SPATIO-TEMPORAL FEATURE FUSION OF HYBRID DEEP LEARNING MODEL[J]. Acta Energiae Solaris Sinica. 2026, 47(12): 357-369 https://doi.org/10.19912/j.0254-0096.tynxb.2025-1037

References

[1] Deshmukh M K G, Sameeroddin M, Abdul D, et al. Renewable energy in the 21st century: a review[J]. Materials Today: Proceedings, 2023, 80: 1756-1759.
[2] 樊肖杰, 迟永宁, 马士聪, 等. 大规模海上风电接入电网关键技术与技术标准的研究及应用[J]. 电网技术, 2022, 46(8): 2859-2870.
Fan X J, Chi Y N, Ma S C, et al.Research and application of key technologies and technical standards for large-scale offshore wind farms connecting to power grid[J]. Power System Technology, 2022, 46(8): 2859-2870.
[3] 陈志永, 张倩茅, 秦梁栋, 等. 考虑大规模区外来电接入的输电网扩展规划方法[J]. 电网与清洁能源, 2022, 38(10): 87-97.
Chen Z Y, Zhang Q M, Qin L D, et al.A transmission expansion planning method considering of large-scale external power access[J]. Power System and Clean Energy, 2022, 38(10): 87-97.
[4] 马林, 戴彬婷, 郭亚娟, 等. 基于多因素条件下风能利用系数Cp输出特性动态性能分析[J]. 中国高新科技, 2022(16): 33-35.
Ma L, Dai B T, Guo Y J, et al.Dynamic performance analysis of the output characteristics of wind energy utilization coefficient Cp under multi-factor conditions[J]. China High-Tech, 2022(16): 33-35.
[5] 霍涛. 风速风向对风机塔筒结构动力响应和疲劳寿命的影响[J]. 建筑结构, 2020, 50(18): 26-33.
Huo T.Influence of wind speed and wind direction on the dynamic response and fatigue life of wind turbine tower-silo structures[J]. Building Structure, 2020, 50(18): 26-33.
[6] Chen W M, Wang P D, Huang S, et al.MPC-based fatigue load suppression of waked wind farm with 2Dof WT control strategy[J]. IEEE Transactions on Sustainable Energy, 2024, 15(4): 2219-2233.
[7] 徐鑫裕, 边晓燕, 张骞, 等. 基于数据驱动的双馈风电场经VSC-HVDC并网次同步振荡影响因素分析[J]. 电力系统保护与控制, 2021, 49(21): 80-87.
Xu X Y, Bian X Y, Zhang Q, et al.Analysis of influencing factors of subsynchronous oscillation caused by a DFIG-based wind farm via the VSC-HVDC grid-connected system based on a data driven method[J]. Power System Protection and Control, 2021, 49(21): 80-87.
[8] Wu H J, Meng K, Fan D, et al.Multistep short-term wind speed forecasting using transformer[J]. Energy, 2022, 261: 125231.
[9] Yuzgec U, Dokur E, Balci M.A novel hybrid model based on Empirical Mode Decomposition and Echo State Network for wind power forecasting[J]. Energy, 2024, 300: 131546.
[10] 兰昆, 吴战波, 赵泽妮, 等. 基于多因子的太阳辐照度预测方法研究进展[J]. 太阳能学报, 2024, 45(5): 593-601.
Lan K, Wu Z B, Zhao Z N, et al.Advance in prediction methods for solar irradiance based on multiple-factors[J]. Acta Energiae Solaris Sinica, 2024, 45(5): 593-601.
[11] Aasim, Singh S N, Mohapatra A. Repeated wavelet transform based ARIMA model for very short-term wind speed forecasting[J]. Renewable Energy, 2019, 136: 758-768.
[12] Pearre N S, Swan L G.Statistical approach for improved wind speed forecasting for wind power production[J]. Sustainable Energy Technologies and Assessments, 2018, 27: 180-191.
[13] 赵泽妮, 云斯宁, 贾凌云, 等. 基于统计模型的短期风能预测方法研究进展[J]. 太阳能学报, 2022, 43(11): 224-234.
Zhao Z N, Yun S N, Jia L Y, et al.Recent progress in short-term forecasting of wind energy based on statistical models[J]. Acta Energiae Solaris Sinica, 2022, 43(11): 224-234.
[14] 贾凌云, 云斯宁, 赵泽妮, 等. 神经网络短期光伏发电预测的应用研究进展[J]. 太阳能学报, 2022, 43(12): 88-97.
Jia L Y, Yun S N, Zhao Z N, et al.Recent progress of short-term forecasting of photovoltaic generation based on artificial neural networks[J]. Acta Energiae Solaris Sinica, 2022, 43(12): 88-97.
[15] 姜明洋, 徐丽, 张开军, 等. 基于季节指数调整的循环神经网络风速时间序列预测[J]. 太阳能学报, 2022, 43(2): 444-450.
Jiang M Y, Xu L, Zhang K J, et al.Recurrent neural network prediction of wind speed time series based on seasonal exponential adjustment[J]. Acta Energiae Solaris Sinica, 2022, 43(2): 444-450.
[16] Blfgeh A, Alkhudhayr H.A machine learning-based sustainable energy management of wind farms using Bayesian recurrent neural network[J]. Sustainability, 2024, 16(19): 8426.
[17] 修春波, 苏欢, 苏雪苗. 多通道长短期记忆卷积网络的风速预测[J]. 电力工程技术, 2022, 41(1): 64-69.
Xiu C B, Su H, Su X M.Wind speed prediction based on multi-channel long short-term memory convolution neural network[J]. Electric Power Engineering Technology, 2022, 41(1): 64-69.
[18] 梁超, 刘永前, 周家慷, 等. 基于卷积循环神经网络的风电场内多点位风速预测方法[J]. 电网技术, 2021, 45(2): 534-541.
Liang C, Liu Y Q, Zhou J K, et al.Wind speed prediction at multi-locations based on combination of recurrent and convolutional neural networks[J]. Power System Technology, 2021, 45(2): 534-541.
[19] 毕贵红, 赵鑫, 李璐, 等. 双模式分解CNN-LSTM集成的短期风速预测模型[J]. 太阳能学报, 2023, 44(3): 191-197.
Bi G H, Zhao X, Li L, et al.Dual-mode decomposition CNN-LSTM integrated short-term wind speed forecasting model[J]. Acta Energiae Solaris Sinica, 2023, 44(3): 191-197.
[20] 任欢, 王旭光. 注意力机制综述[J]. 计算机应用, 2021, 41(S1): 1-6.
Ren H, Wang X G.Review of attention mechanism[J]. Journal of Computer Applications, 2021, 41(S1): 1-6.
[21] Chung J, Gulcehre C, Cho K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. ArXiv Preprint ArXiv:1412. 3555, 2014.
[22] LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[23] 吐松江•卡日, 雷柯松, 马小晶, 等. 基于LSTM-Attention和CNN-BiGRU误差修正的光伏功率预测[J]. 太阳能学报, 2024, 45(12): 85-93.
Kari•Tusongjiang, Lei K S, Ma X J, et al. Photovoltaic power prediction based on LSTM-attention and CNN-BiGRU error correction[J]. Acta Energiae Solaris Sinica, 2024, 45(12): 85-93.
[24] 王新宇. 基于深度学习的航拍图像小目标检测算法研究[D]. 沈阳: 沈阳理工大学, 2023.
Wang X Y.Research on small target detection algorithm for aerial photography images based on deep learning[D]. Shenyang: Shenyang Ligong University, 2023.
[25] Jordan M I.An introduction to linear algebra in parallel distributed processing[M]. Parallel Distributed Processing, 1986, 1: 365-422.
[26] 刘建伟, 宋志妍. 循环神经网络研究综述[J]. 控制与决策, 2022, 37(11): 2753-2768.
Liu J W, Song Z Y.Overview of recurrent neural networks[J]. Control and Decision, 2022, 37(11): 2753-2768.
[27] 张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021, 57(11): 57-69.
Zhang C, Guo Y, Li M.Review of development and application of artificial neural network models[J]. Computer Engineering and Applications, 2021, 57(11): 57-69.
[28] Vaswani A, Shazeer N, Parmar N, et al.Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017: 6000-6010.
[29] 陈殿昊, 臧海祥, 蒋雨楠, 等. 基于多层次云图特征与宽度学习的超短期光伏功率预测[J]. 电力系统自动化, 2024, 48(22): 131-139.
Chen D H, Zang H X, Jiang Y N, et al.Ultra-short-term photovoltaic power forecasting based on multi-level sky image features and broad learning[J]. Automation of Electric Power Systems, 2024, 48(22): 131-139.
[30] Zhao Z N, Yun S N, Jia L Y, et al.Hybrid VMD-CNN-GRU-based model for short-term forecasting of wind power considering spatio-temporal features[J]. Engineering Applications of Artificial Intelligence, 2023, 121: 105982.
[31] Guo J X, Yun S N, Meng Y, et al.Prediction of heating and cooling loads based on light gradient boosting machine algorithms[J]. Building and Environment, 2023, 236: 110252.
[32] Meng Y, Yun S N, Zhao Z N, et al.Short-term electricity load forecasting based on a novel data preprocessing system and data reconstruction strategy[J]. Journal of Building Engineering, 2023, 77: 107432.
[33] Jia L Y, Yun S N, Zhao Z N, et al.Improving short-term forecasting of solar power generation by using an EEMD-BiGRU model: a comparative study based on seven standalone models and six hybrid models[J]. International Journal of Green Energy, 2024, 21(14): 3135-3158.