EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Investigating the Methodology of Multisource Data Fusion in Wind Speed Forecasting Correction

Julong Chen, Bin Wang, Xuepeng Mou, Chen Luo, Zhen Li, Jingming Xia

Advances in Meteorology · 2025-01 · 2025 (1)

DOI: 10.1155/adme/8892096 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Meteorology, online erschienen am 14.10.2025. Verfasst von Julong Chen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 14 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate numerical weather prediction (NWP) is the foundation of high‐quality meteorological services. Although European Centre for Medium‐Range Weather Forecasts (ECMWF) products are widely used, they still exhibit systematic forecast errors. Multisource data fusion, as one of the correction methods, can significantly reduce forecast errors. Given the heterogeneity of geographical data, ground truth data, and satellite data, it is essential to establish a mechanism to effectively extract useful information from these multisource datasets while avoiding interference from noise and redundant information. Therefore, this article proposes a multisource data numerical forecast correction network (MSD‐NFCNet), which corrects the ECMWF 10‐m wind speed forecast by integrating FY4A satellite data, digital elevation models (DEMs), and ERA5 data. The MSD‐NFCNet incorporates a spatial resolution alignment module that uses the correction upsampling network (CUPS‐Net) and multiscale residual network (MSR‐Net) submodules to match the spatial grids of multisource heterogeneous data. Additionally, a multisource feature fusion module is designed to fuse the spatiotemporal features of the heterogeneous data. In this article, the MSD‐NFCNet is used to correct the 10‐m wind speed in the ECMWF product, and the results are compared with the original ECMWF forecast, ERA5 ground truth, and other correction methods. Experimental results show that the root mean square error (RMSE) of the corrected 10‐m wind speed using MSD‐NFCNet decreases by 61.03% compared to the ECMWF forecast, demonstrating that the proposed model significantly improves forecast accuracy.

Förderung

  • National Key Research and Development Program of China

Zitieren

APA

Chen, J., Wang, B., Mou, X., Luo, C., Li, Z., & Xia, J. (2025). Investigating the Methodology of Multisource Data Fusion in Wind Speed Forecasting Correction. Advances in Meteorology, 2025(1). https://doi.org/10.1155/adme/8892096

Vancouver

Chen J, Wang B, Mou X, Luo C, Li Z, Xia J. Investigating the Methodology of Multisource Data Fusion in Wind Speed Forecasting Correction. Advances in Meteorology. 2025;2025(1). doi:10.1155/adme/8892096

BibTeX
@article{Chen2025,
  author = {Chen, Julong and Wang, Bin and Mou, Xuepeng and Luo, Chen and Li, Zhen and Xia, Jingming},
  title = {Investigating the Methodology of Multisource Data Fusion in Wind Speed Forecasting Correction},
  journal = {Advances in Meteorology},
  year = {2025},
  volume = {2025},
  number = {1},
  doi = {10.1155/adme/8892096}
}
Kostenlos starten →