BiLSTM-KAN SHORT-TERM WIND SPEED PREDICTION MODEL CONSIDERING SPATIO-TEMPORAL CORRELATION

Ren Yichao; Huang Guoyong; Deng Weiquan; Kong Xianzheng

doi:10.19912/j.0254-0096.tynxb.2025-0684

PDF(1551 KB)

Acta Energiae Solaris Sinica ›› 2026, Vol. 47 ›› Issue (9) : 738-745. DOI: 10.19912/j.0254-0096.tynxb.2025-0684

BiLSTM-KAN SHORT-TERM WIND SPEED PREDICTION MODEL CONSIDERING SPATIO-TEMPORAL CORRELATION

Ren Yichao, Huang Guoyong, Deng Weiquan, Kong Xianzheng

Author information +

History +

Abstract

To comprehensively capture the spatiotemporal characteristics of wind speed, this paper proposes a short-term wind speed forecasting model based on graph convolutional network （GCN）, successive variational mode decomposition （SVMD）, bidirectional long short-term memory （BiLSTM） network, and the Kolmogorov-Arnold Network （KAN）. First, GCN is employed to extract spatial features among wind speeds at different horizontal distances. Then, SVMD is used to decompose the target wind speed series, thereby extracting modal features across various temporal scales. Finally, the extracted spatial features and the multi-scale temporal components are fused and input into a BiLSTM-KAN model to perform forecasting. Experimental results show that for 1-, 3-, 6-, 9-, and 12-step ahead predictions, the proposed model reduces the mean absolute error by 38.36%, 55.45%, 58.21%, 60.53%, and 59.76%, respectively, compared with the BiLSTM model.

Key words

wind power / wind speed / forecasting / graph convolutional network / bidirectional long short-term memory network / Kolmogorov-Arnold Network

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Ren Yichao, Huang Guoyong, Deng Weiquan, Kong Xianzheng. BiLSTM-KAN SHORT-TERM WIND SPEED PREDICTION MODEL CONSIDERING SPATIO-TEMPORAL CORRELATION[J]. Acta Energiae Solaris Sinica. 2026, 47(9): 738-745 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0684

References

[1] Bommidi B S, Teeparthi K.A hybrid wind speed prediction model using improved CEEMDAN and Autoformer model with auto-correlation mechanism[J]. Sustainable Energy Technologies and Assessments, 2024, 64: 103687.
[2] 国家能源局. 2024年1—8月份全国电力工业统计数据[R/OL]. https://www.nea.gov.cn/2024-09/23/c_1212399359.htm 2024.
National Energy Bureau.National electric power industry statistics for january-august 2024[R/OL]. https://www.nea.gov.cn/2024-09/23/c_1212399359.htm 2024.
[3] Zhang C, Li Z B, Ge Y D, et al.Enhancing short-term wind speed prediction based on an outlier-robust ensemble deep random vector functional link network with AOA-optimized VMD[J]. Energy, 2024, 296: 131173.
[4] 唐非. 基于互补集成经验模态分解和Stacking融合的短期风速组合预测模型[J]. 太阳能学报, 2024, 45(7): 735-744.
Tang F.Short-term wind speed combination prediction model based on complementary ensemble empirical mode decomposition and Stacking fusion[J]. Acta Energiae Solaris Sinica, 2024, 45(7): 735-744.
[5] 李永刚, 王月, 刘丰瑞, 等. 基于Stacking融合的短期风速预测组合模型[J]. 电网技术, 2020, 44(8): 2875-2882.
Li Y G, Wang Y, Liu F R, et al.Combination model of short-term wind speed prediction based on Stacking fusion[J]. Power System Technology, 2020, 44(8): 2875-2882.
[6] 潘超, 李润宇, 蔡国伟, 等. 基于属性约简重构的自校正卷积记忆风速预测[J]. 中国电机工程学报, 2023, 43(7): 2721-2732.
Pan C, Li R Y, Cai G W, et al.Wind speed prediction with self-tuning convolutional memory based on attribute reduction reconstruction[J]. Proceedings of the CSEE, 2023, 43(7): 2721-2732.
[7] 凡航, 张雪敏, 梅生伟, 等. 基于时空神经网络的风电场超短期风速预测模型[J]. 电力系统自动化, 2021, 45(1): 28-35.
Fan H, Zhang X M, Mei S W, et al.Ultra-short-term wind speed prediction model for wind farms based on spatiotemporal neural network[J]. Automation of Electric Power Systems, 2021, 45(1): 28-35.
[8] Zhu Q M, Chen J F, Shi D Y, et al.Learning temporal and spatial correlations jointly: a unified framework for wind speed prediction[J]. IEEE Transactions on Sustainable Energy, 2020, 11(1): 509-523.
[9] 王旭光, 张可, 李潇, 等. 基于变量相关注意力机制的短期风速预测[J]. 太阳能学报, 2023, 44(8): 467-476.
Wang X G, Zhang K, Li X, et al.Short-term wind speed forecasting based on variable correlation attention mechanism[J]. Acta Energiae Solaris Sinica, 2023, 44(8): 467-476.
[10] Dai J F, Fu L H.A wind speed forecasting model using nonlinear auto-regressive model optimized by the hybrid chaos-cloud salp swarm algorithm[J]. Energy, 2024, 298: 131332.
[11] Nazari M, Sakhaei S M.Successive variational mode decomposition[J]. Signal Processing, 2020, 174: 107610.
[12] 张越, 臧海祥, 韩海腾, 等. 基于FSN-MCCN-SA-BiLSTM的短期风速预测[J]. 太阳能学报, 2024, 45(8): 529-536.
Zhang Y, Zang H X, Han H T, et al.Short-term wind speed prediction based on FSN-MCCN-SA-BiLSTM[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 529-536.
[13] Zhang Y M, Wang H.Multi-head attention-based probabilistic CNN-BiLSTM for day-ahead wind speed forecasting[J]. Energy, 2023, 278: 127865.
[14] ZHANG C, MA H X, HUA L, et al.An evolutionary deep learning model based on TVFEMD, improved sine cosine algorithm, CNN and BiLSTM for wind speed prediction[J]. Energy, 2022, 254: 124250.
[15] 王泰然, 鲍逸非. 基于卷积神经网络和双向长短期记忆网络的微地震记录去噪方法[J]. 北京大学学报(自然科学版), 2025, 61(3): 487-500.
Wang T R, Bao Y F.Micro-earthquake recording denoising method based on convolutional neural network and bidirectional long short-term memory[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2025, 61(3): 487-500.
[16] Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[J]. Arxiv Preprint Arxiv:1609.02907, 2016.
[17] Parri S, Teeparthi K.Svmd-Tf-Qs: an efficient and novel hybrid methodology for the wind speed prediction[J]. Expert Systems with Application, 2024, 249: 123516.
[18] Xiao Y N, Cui H, Abu Khurma R, et al.Artificial lemming algorithm: a novel bionic meta-heuristic technique for solving real-world engineering optimization problems[J]. Artificial Intelligence Review, 2025, 58(3): 84.
[19] 向玲, 陈锦鹏, 付晓孟婷, 等. 基于VIT和LSTM的风速多步预测[J]. 太阳能学报, 2024, 45(9): 525-533.
Xiang L, Chen J P, Fu X M T, et al. Multi-step wind speed forecasting based on VIT and LSTM[J]. Acta Energiae Solaris Sinica, 2024, 45(9): 525-533.
[20] 李承皓, 杨永标, 宋嘉启, 等. 基于IMVMD和BiLSTM-SARIMA组合模型的台区光伏短期发电功率预测[J]. 太阳能学报, 2025, 46(2): 433-440.
Li C H, Yang Y B, Song J Q, et al.Short term photovoltaic power generation prediction based on IMVMD and BiLSTM-SARIMA combination model in station area[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 433-440.
[21] Liu Z M, Wang Y X, Vaidya S, et al. Kan: Kolmogorov-arnold networks[J]. Arxiv Preprint Arxiv:2404. 19756, 2024.
[22] JI T R, HOU Y T, ZHANG D. A comprehensive survey on Kolmogorov Arnold networks (KAN)[J]. Arxiv Preprint Arxiv:2407.11075, 2024.