考虑数据分解和Gish-BiTCN-MHSA的短期光伏功率预测

刘海鹏; 何艳苹; 金怀平; 方奇文; 吴洪

doi:10.19912/j.0254-0096.tynxb.2024-0603

PDF(3822 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (8) : 430-438. DOI: 10.19912/j.0254-0096.tynxb.2024-0603

考虑数据分解和Gish-BiTCN-MHSA的短期光伏功率预测

刘海鹏¹, 何艳苹¹, 金怀平¹, 方奇文², 吴洪¹

作者信息 +

SHORT-TERM PV POWER PREDICTION CONSIDERING DATA DECOMPOSITION AND GISH-BITCN-MHSA

Liu Haipeng¹, He Yanping¹, Jin Huaiping¹, Fang Qiwen², Wu Hong¹

Author information +

文章历史 +

摘要

为有效应对分布式光伏电站输出功率的波动对电网稳定性的挑战,提出一个新的短期光伏功率预测框架。首先,使用最优变分模态分解（OVMD）技术将原始光伏功率数据分解成多个模态分量,并将其与相关特征融合,生成一系列子序列。然后,采用结合Gish激活函数的双向时间卷积网络（Gish-BiTCN）对每个子序列进行预测,引入多头注意力机制（MHSA）使模型更加关注和捕捉时间相关特征。最后,通过对所有子序列的预测值进行重构得到最终的预测结果。通过实验验证其在光伏发电预测方面的优越性。

Abstract

In order to effectively cope with the challenge of grid in stability caused by output power fluctuations of distributed PV power plants, a new short-term PV power prediction framework is proposed in this study. Firstly, the raw PV power data are decomposed into multiple modal components using the optimal variational modal decomposition （OVMD） technique, which is fused with relevant features to generate a series of subsequences. Then, a bidirectional time convolutional network （Gish-BiTCN） incorporating the Gish activation function is used to predict each subsequence, and the mechanism of multiple heads’attention （MHSA） is introduced to enable the model to pay more attention to and capture the time-related features. Finally, the final prediction results are obtained by reconstructing the predicted values of all subsequences. Its superiority in PV power generation prediction is verified through experiments.

导出引用

刘海鹏, 何艳苹, 金怀平, 方奇文, 吴洪. 考虑数据分解和Gish-BiTCN-MHSA的短期光伏功率预测[J]. 太阳能学报. 2025, 46(8): 430-438 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0603

Liu Haipeng, He Yanping, Jin Huaiping, Fang Qiwen, Wu Hong. SHORT-TERM PV POWER PREDICTION CONSIDERING DATA DECOMPOSITION AND GISH-BITCN-MHSA[J]. Acta Energiae Solaris Sinica. 2025, 46(8): 430-438 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0603

中图分类号： TM615

参考文献

[1] VANDEVENTER W, JAMEI E, THIRUNAVUKKAR ASU G S, et al. Short-term PV power forecasting using hybrid GASVM technique[J]. Renewable energy, 2019, 140: 367-379.
[2] LI G Q, XIE S, WANG B Z, et al.Photovoltaic power forecasting with a hybrid deep learning approach[J]. IEEE access, 2020, 8: 175871-175880.
[3] 何威, 苏中元, 史金林, 等. 基于双重注意力GRU与相似修正的光伏功率预测[J]. 太阳能学报, 2024, 45(3): 480-487.
HE W, SU Z Y, SHI J L, et al.Photovoltaic power forecasting based on dual-attention-GRU and similarity modification[J]. Acta energiae solaris sinica, 2024, 45(3): 480-487.
[4] LÓPEZ GÓMEZ J, OGANDO MARTÍNEZ A, TRONCOSO PASTORIZA F, et al. Photovoltaic power prediction using artificial neural networks and numerical weather data[J]. Sustainability, 2020, 12(24): 10295.
[5] 王瑞, 高强, 逯静. 基于CEEMDAN-LSSVM-ARIMA模型的短期光伏功率预测[J]. 传感器与微系统, 2022, 41(5): 118-122.
WANG R, GAO Q, LU J.Short-term photovoltaic power prediction based on CEEMDAN-LSSVM-ARIMA model[J]. Transducer and microsystem technologies, 2022, 41(5): 118-122.
[6] HU W, ZHANG X Y, ZHU L J, et al.Short-term photovoltaic power prediction based on similar days and improved SOA-DBN model[J]. IEEE access, 2020, 9: 1958-1971.
[7] 吴长林, 陈玉, 高文根. 基于VMD-IPSO-RFR模型的光伏发电功率预测[J]. 四川轻化工大学学报(自然科学版), 2020, 33(2): 73-79.
WU C L, CHEN Y, GAO W G.Photovoltaic power prediction based on VMD-IPSO-RFR model[J]. Journal of Sichuan University of Science & Engineering (natural science edition), 2020, 33(2): 73-79.
[8] 龙小慧, 秦际赟, 张青雷, 等. 基于相似日聚类及模态分解的短期光伏发电功率组合预测研究[J]. 电网技术, 2024, 48(7): 2948-2957.
LONG X H, QIN J Y, ZHANG Q L, et al.Short-term photovoltaic power prediction study based on similar day clustering and modal decomposition[J]. Grid technology, 2024, 48(7): 2948-2957.
[9] 陈杨. 基于VMD-SSA-BP方法的光伏发电短期功率预测[J]. 自动化应用, 2023, 64(15): 67-70.
CHEN Y.Short-term power prediction of photovoltaic power generation based on VMD-SSA-BP method[J]. Automation application, 2023, 64(15): 67-70.
[10] 陈蕻峰, 王贺, 李岩, 等. 组合两步分解和ARIMA-LSTM的短期风速预测研究[J]. 太阳能学报, 2024, 45(2): 164-171.
CHEN H F, WANG H, LI Y, et al.Short-term wind speed prediction by combining two-step decomposition and ARIMA-LSTM[J]. Acta energiae solaris sinica, 2024, 45(2): 164-171.
[11] 刘译文, 赵一帆, 张兰兰, 等. 基于小波包分解的长短期记忆网络光伏功率预测[J]. 计算机与数字工程, 2022, 50(9): 2114-2118.
LIU Y W, ZHAO Y F, ZHANG L L, et al.Photovoltaic power forecasting based on wavelet packet decomposition of long short-term memory network[J]. Computer & digital engineering, 2022, 50(9): 2114-2118.
[12] 焦丕华, 蔡旭, 王乐乐, 等. 考虑数据分解和进化捕食策略的BiLSTM短期光伏发电功率预测[J]. 太阳能学报, 2024, 45(2): 435-442.
JIAO P H, CAI X, WANG L L, et al.BiLSTM short-term photovoltaic power prediction considering data decomposition and evolutionary predation strategies[J]. Acta energiae solaris sinica, 2024, 45(2): 435-442.
[13] CEN Y F, LUO M X, CEN G, et al.Financial market correlation analysis and stock selection application based on TCN-deep clustering[J]. Future internet, 2022, 14(11): 331.
[14] 吴珺玥, 赵二刚, 郭增良, 等. 基于Spearman系数和TCN的光伏出力超短期多步预测[J]. 太阳能学报, 2023, 44(9): 180-186.
WU J Y, ZHAO E G, GUO Z L, et al.Ultra-short-term photovoltaic power multi-step prediction based on Spearman coefficient and TCN[J]. Acta energiae solaris sinica, 2023, 44(9): 180-186.
[15] ZHANG D D, CHEN B A, ZHU H Y, et al.Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model[J]. Energy, 2023, 285: 128762.
[16] CAI S H, XU H, LIU M J, et al.A malicious network traffic detection model based on bidirectional temporal convolutional network with multi-head self-attention mechanism[J]. Computers & security, 2024, 136: 103580.
[17] SHENG Y W, WANG H, YAN J, et al.Short-term wind power prediction method based on deep clustering-improved temporal convolutional network[J]. Energy reports, 2023, 9: 2118-2129.
[18] DRAGOMIRETSKIY K, ZOSSO D.Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.
[19] MIRJALILI S, LEWIS A.The whale optimization algorithm[J]. Advances in engineering software, 2016, 95: 51-67.
[20] 何艳苹, 刘海鹏, 方奇文, 等. 基于多模态扰动的集成短期光伏功率预测研究[J]. 陕西理工大学学报(自然科学版), 2025, 41(1): 33-41.
HE Y P, LIU H P, FANG Q W, et al.Ensemble short-term photovoltaic power prediction based on multimodal perturbation[J]. Journal of Shanxi University of Technology(natural science edition), 2025, 41(1): 33-41.
[21] KAYTAN M, AYDILEK İ B, YEROĞLU C. Gish: a novel activation function for image classification[J]. Neural computing and applications, 2023, 35(34): 24259-24281.