A Transformer-based multimodal-learning framework using sky images for ultra-short-term solar irradiance forecasting

被引:61
作者
Liu, Jingxuan [1 ]
Zang, Haixiang [1 ]
Cheng, Lilin [1 ]
Ding, Tao [2 ]
Wei, Zhinong [1 ]
Sun, Guoqiang [1 ]
机构
[1] Hohai Univ, Coll Energy & Elect Engn, Nanjing 210098, Peoples R China
[2] Xi An Jiao Tong Univ, Dept Elect Engn, Xian 710049, Peoples R China
基金
中国国家自然科学基金;
关键词
Solar irradiance forecasting; Multimodal-learning; Transformer; Ground-based sky image; RADIATION;
D O I
10.1016/j.apenergy.2023.121160
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
The development of solar energy is crucial to combat the global climate change and fossil energy crisis. However, the inherent uncertainty of solar power prevents its large-scale integration into power grids. Although various sky-image-derived modeling methods exist to forecast the variations of solar irradiance, few focus on fully uti-lizing the coupling correlations between sky images and historical data to improve the forecasting performance. Therefore, a novel multimodal-learning framework is proposed for forecasting global horizontal irradiance (GHI) in the ultra-short-term. First, the historical and empirically estimated clear-sky GHI are encoded by Informer. Then, the ground-based sky images are transformed into optical flow maps, which can be handled by Vision Transformer. Subsequently, a cross-modality attention method is proposed to explore the coupling correlations between the two modalities. Last, a generative decoder is used to implement multi-step forecasting. The experimental results show that the proposed method achieves a normalized root mean square error (NRMSE) of 4.28% in 10-min-ahead forecasting. Several state-of-the-art methods are also used for comparisons. The exper-imental results show that the proposed method outperforms the benchmark methods and exhibits higher ac-curacy and robustness in ultra-short-term GHI forecasting.
引用
收藏
页数:19
相关论文
共 50 条
[31]   Proactively selection of input variables based on information gain factors for deep learning models in short-term solar irradiance forecasting [J].
Chen, Yunxiao ;
Bai, Mingliang ;
Zhang, Yilan ;
Liu, Jinfu ;
Yu, Daren .
ENERGY, 2023, 284
[32]   Transformer-based short-term traffic forecasting model considering traffic spatiotemporal correlation [J].
Chang, Ande ;
Ji, Yuting ;
Bie, Yiming .
FRONTIERS IN NEUROROBOTICS, 2025, 19
[33]   Distributed Photovoltaic Ultra-Short-Term Power Forecasting Utilizing OSAPSO-Optimized Spatio Temporal Dual Transformer Model [J].
Shi, Lei ;
Lu, Tianguang ;
Guo, Yu ;
Yi, Xinning ;
Xu, Xiaotong ;
Liu, Hao .
2024 THE 7TH INTERNATIONAL CONFERENCE ON ENERGY, ELECTRICAL AND POWER ENGINEERING, CEEPE 2024, 2024, :1399-1404
[34]   Sky-image-based sun-blocking index and PredRNN plus plus for accurate short-term solar irradiance forecasting [J].
Rajagukguk, Rial A. ;
Lee, Hyunjin .
BUILDING AND ENVIRONMENT, 2025, 274
[35]   Short-term solar irradiance forecasting using convolutional neural networks and cloud imagery [J].
Choi, Minsoo ;
Rachunok, Benjamin ;
Nateghi, Roshanak .
ENVIRONMENTAL RESEARCH LETTERS, 2021, 16 (04)
[36]   Univariate short term forecasting of solar irradiance using modified online backpropagation through time [J].
Sharma, Animesh Anant .
2016 20TH INTERNATIONAL COMPUTER SCIENCE AND ENGINEERING CONFERENCE (ICSEC), 2016,
[37]   Multi-Branch ResNet-Transformer for Short-Term Spatio-Temporal Solar Irradiance Forecasting [J].
Ziyabari, Saeedeh ;
Zhao, Zhenyu ;
Du, Liang ;
Biswas, Saroj K. .
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, 2023, 59 (05) :5293-5303
[38]   Deep learning and statistical methods for short- and long-term solar irradiance forecasting for Islamabad [J].
Haider, Syed Altan ;
Sajid, Muhammad ;
Sajid, Hassan ;
Uddin, Emad ;
Ayaz, Yasar .
RENEWABLE ENERGY, 2022, 198 :51-60
[39]   Short-term Solar Irradiance Forecasting Based on Multi-Branch Residual Network [J].
Ziyabari, Saeedeh ;
Du, Liang ;
Biswas, Saroj .
2020 IEEE ENERGY CONVERSION CONGRESS AND EXPOSITION (ECCE), 2020, :2000-2005
[40]   Solar irradiance forecasting using a ground-based sky imager developed at UC San Diego [J].
Yang, Handa ;
Kurtz, Ben ;
Dung Nguyen ;
Urquhart, Bryan ;
Chow, Chi Wai ;
Ghonima, Mohamed ;
Kleissl, Jan .
SOLAR ENERGY, 2014, 103 :502-524