基于VMD-T2V-Transformer的太阳辐射预测

胡雅彬; 史加荣; 陈应瑞; 雍龙泉

doi:10.19912/j.0254-0096.tynxb.2024-0442

PDF(2280 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (7) : 778-784. DOI: 10.19912/j.0254-0096.tynxb.2024-0442

第二十七届中国科协年会学术论文

基于VMD-T2V-Transformer的太阳辐射预测

胡雅彬¹, 史加荣¹, 陈应瑞¹, 雍龙泉²

作者信息 +

SOLAR RADIATION PREDICTION BASED ON VMD-T2V-TRANSFORMER

Hu Yabin¹, Shi Jiarong¹, Chen Yingrui¹, Yong Longquan²

Author information +

文章历史 +

摘要

太阳辐射的不确定性导致太阳能发电具有明显的随机性和不稳定性。针对此问题,该文结合变分模态分解（VMD）、Time2Vec（T2V）和Transformer,提出一种用于太阳辐射预测的VMD-T2V-Transformer模型。首先利用VMD将太阳辐射序列分解为若干子序列;接着采用T2V对分解后的每个子序列进行时间特征嵌入;然后对嵌入时间特征后的子序列建立Transformer预测模型;最后将各模型的预测结果进行叠加,得到最终预测值。实验结果表明：该文所提模型优于其他主流模型,RMSE和MAE至少降低13.81％和16.44％。

Abstract

The uncertainty in solar radiation leads to obvious randomness and instability in solar power generation. To address this issue, this paper proposes a VMD-T2V-Transformer model for solar radiation prediction by integrating variational mode decomposition （VMD）, Time2Vec （T2V） and Transformer. First, the VMD is used to decompose the solar radiation sequence into several sub-sequences. Next, the T2V is adopted to embed the temporal features of each decomposed sub-sequence. Then, a Transformer prediction model is established for each sub-sequence based on the embedded time features. Finally, the predicted results of all sub-models are superimposed to obtain the final predicted values. The experimental results show that the proposed model in this paper outperforms other mainstream models in terms of RMSE and MAE, which can be reduced by at least 13.81％ and 16.44％ respectively.

导出引用

胡雅彬, 史加荣, 陈应瑞, 雍龙泉. 基于VMD-T2V-Transformer的太阳辐射预测[J]. 太阳能学报. 2025, 46(7): 778-784 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0442

Hu Yabin, Shi Jiarong, Chen Yingrui, Yong Longquan. SOLAR RADIATION PREDICTION BASED ON VMD-T2V-TRANSFORMER[J]. Acta Energiae Solaris Sinica. 2025, 46(7): 778-784 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0442

中图分类号： TM615

参考文献

[1] 金存银, 张淑花, Li Xingong, 等. 地表太阳辐射短期预测方法研究进展[J]. 太阳能学报, 2023, 44(12): 150-161.
JIN C Y, ZHANG S H, LI X G, et al.Research progress on short-term prediction methods of surface solar radiation[J]. Acta energiae solaris sinica, 2023, 44(12): 150-161.
[2] PERERA M, DE HOOG J, BANDARA K, et al.Day-ahead regional solar power forecasting with hierarchical temporal convolutional neural networks using historical power generation and weather data[J]. Applied energy, 2024, 361: 122971.
[3] 齐月, 房世波, 周文佐. 近50年来中国东、西部地面太阳辐射变化及其与大气环境变化的关系[J]. 物理学报, 2015, 64(8): 398-407.
QI Y, FANG S B, ZHOU W Z.Correlative analysis between the changes of surface solar radiation and its relationship with air pollution, as well as meteorological factor in East and West China in recent 50 years[J]. Acta physica sinica, 2015, 64(8): 398-407.
[4] 李津, 史加荣, 张琰妮, 等. 基于最大信息系数的短期太阳辐射协同估计[J]. 太阳能学报, 2023, 44(9): 286-294.
LI J, SHI J R, ZHANG Y N, et al.Short-term solar radiation synergy estimation based on maximum information coefficient[J]. Acta energiae solaris sinica, 2023, 44(9): 286-294.
[5] 陈龙, 张菁, 张昊立, 等. 基于VMD和射箭算法优化改进ELM的短期光伏发电预测[J]. 太阳能学报, 2023, 44(10): 135-141.
CHEN L, ZHANG J, ZHANG H L, et al.Short-term photovoltaic power generation forecast based on VMD-IAA-IHEKLM model[J]. Acta energiae solaris sinica, 2023, 44(10): 135-141.
[6] WANG L N, MAO M X, XIE J L, et al.Accurate solar PV power prediction interval method based on frequency-domain decomposition and LSTM model[J]. Energy, 2023, 262: 125592.
[7] ABOU HOURAN M, SALMAN BUKHARI S M, ZAFAR M H, et al. COA-CNN-LSTM: coati optimization algorithm-based hybrid deep learning model for PV/wind power forecasting in smart grid applications[J]. Applied energy, 2023, 349: 121638.
[8] NESHAT M, NEZHAD M M, MIRJALILI S, et al.Short-term solar radiation forecasting using hybrid deep residual learning and gated LSTM recurrent network with differential covariance matrix adaptation evolution strategy[J]. Energy, 2023, 278: 127701.
[9] 张雲钦, 程起泽, 蒋文杰, 等. 基于EMD-PCA-LSTM的光伏功率预测模型[J]. 太阳能学报, 2021, 42(9): 62-69.
ZHANG Y Q, CHENG Q Z, JIANG W J, et al.Photovoltaic power prediction model based on EMD-PCA-LSTM[J]. Acta energiae solaris sinica, 2021, 42(9): 62-69.
[10] 李倩倩, 严珂. 基于RGA-BiLSTM模型的太阳辐照度预测[J]. 中国计量大学学报, 2023, 34(1): 74-83.
LI Q Q, YAN K.Solar irradiance prediction based on the RGA-BiLSTM model[J]. Journal of China University of Metrology, 2023, 34(1): 74-83.
[11] DAI Y M, WANG Y X, LENG M M, et al.Lowess smoothing and random forest based GRU model: a short-term photovoltaic power generation forecasting method[J]. Energy, 2022, 256: 124661.
[12] 王冉冉, 高慧敏, 张昕宇. 基于GA-GRU神经网络的光伏MPPT算法[J]. 太阳能学报, 2023, 44(9): 212-219.
WANG R R, GAO H M, ZHANG X Y.MPPT algorithm for photovoltaics based on GA-GRU neural network[J]. Acta energiae solaris sinica, 2023, 44(9): 212-219.
[13] RAI A, SHRIVASTAVA A, JANA K C.A robust auto encoder-gated recurrent unit (AE-GRU) based deep learning approach for short term solar power forecasting[J]. Optik, 2022, 252: 168515.
[14] 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, Long Beach, california, USA, 2017: 6000-6010.
[15] KAZEMI S M, GOEL R, EGHBALI S, et al. Time2Vec: learning a vector representation of time[EB/OL]. (2019-7-11)[2023-8-30]. http://arxiv.org/pdf/1907.05321.pdf
[16] DRAGOMIRETSKIY K, ZOSSO D.Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.
[17] HUANG N E, SHEN Z, LONG S R, et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the royal society of London series A, 1998, 454(1971): 903-998.
[18] 官松泽, 唐钰本, 蔡争, 等. 基于K-means++-Bi-LSTM的太阳辐照度超短期预测[J]. 太阳能学报, 2023, 44(12): 170-174.
GUAN S Z, TANG Y B, CAI Z, et al.Ultra-short-term forecast of solar irradiance based on K-means++-Bi-LSTM[J]. Acta energiae solaris sinica, 2023, 44(12): 170-174.
[19] 汪凯, 叶红, 陈峰, 等. 中国东南部太阳辐射变化特征、影响因素及其对区域气候的影响[J]. 生态环境学报, 2010, 19(5): 1119-1124.
WANG K, YE H, CHEN F, et al.Long-term change of solar radiation in southeastern China: variation, factors, and climate forcing[J]. Ecology and environmental sciences, 2010, 19(5): 1119-1124.