EUVIMEDEUVIMED entdecken
Crossref

SolarFusionNet: Enhanced Solar Irradiance Forecasting via Automated Multi-Modal Feature Selection and Cross-Modal Fusion

Tao Jing, Shanlin Chen, David Navarro-Alarcon, Yinghao Chu, Mengying Li

IEEE Transactions on Sustainable Energy · 2025-04 · 16 (2) · 761-773

DOI: 10.1109/tste.2024.3482360 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Sustainable Energy, 2025-04. Verfasst von Tao Jing und 4 weiteren Autor:innen. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 44 Literaturangaben. Gefördert von: Research Grants Council of the Hong Kong, Basic and Applied Basic Research Foundation of Guangdong Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Solar forecasting has emerged as a cost-effective technology to mitigate the negative impacts of intermittent solar power on the power grid. Despite the multitude of deep learning methodologies available for forecasting solar irradiance, there is a notable gap in research concerning the automated selection and holistic utilization of multi-modal features for ultra-short-term regional irradiance forecasting. Our study introduces SolarFusionNet, a novel deep learning architecture that effectively integrates automatic multi-modal feature selection and cross-modal data fusion. SolarFusionNet utilizes two distinct types of automatic variable feature selection units to extract relevant features from multichannel satellite images and multivariate meteorological data, respectively. Long-term dependencies are then captured using three types of recurrent layers, each tailored to the corresponding data modal. In particular, a novel Gaussian kernel-injected convolutional long short-term memory network is specifically designed to isolate the sparse features present in the cloud motion field derived from optical flow. Subsequently, a hierarchical multi-head cross-modal self-attention mechanism is proposed based on the physical-logical dependencies among the three modalities to investigate the coupling correlations among the modalities. The experimental results indicate that SolarFusionNet exhibits robust performance in predicting regional solar irradiance, achieving higher accuracy than other state-of-the-art models and a forecast skill ranging from 37.4% to 47.6% against the smart persistence model for the 4-hour-ahead forecast.

Förderung

  • Research Grants Council of the Hong Kong
  • Basic and Applied Basic Research Foundation of Guangdong Province

Zitieren

APA

Jing, T., Chen, S., Navarro-Alarcon, D., Chu, Y., & Li, M. (2025). SolarFusionNet: Enhanced Solar Irradiance Forecasting via Automated Multi-Modal Feature Selection and Cross-Modal Fusion. IEEE Transactions on Sustainable Energy, 16(2), 761-773. https://doi.org/10.1109/tste.2024.3482360

Vancouver

Jing T, Chen S, Navarro-Alarcon D, Chu Y, Li M. SolarFusionNet: Enhanced Solar Irradiance Forecasting via Automated Multi-Modal Feature Selection and Cross-Modal Fusion. IEEE Transactions on Sustainable Energy. 2025;16(2):761-773. doi:10.1109/tste.2024.3482360

BibTeX
@article{Jing2025,
  author = {Jing, Tao and Chen, Shanlin and Navarro-Alarcon, David and Chu, Yinghao and Li, Mengying},
  title = {SolarFusionNet: Enhanced Solar Irradiance Forecasting via Automated Multi-Modal Feature Selection and Cross-Modal Fusion},
  journal = {IEEE Transactions on Sustainable Energy},
  year = {2025},
  volume = {16},
  number = {2},
  pages = {761--773},
  doi = {10.1109/tste.2024.3482360}
}
Kostenlos starten →