EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-scale dynamic graph neural network for PM2.5 concentration prediction in regional station cluster

Xin Lu, Juyang Liao, Huihua Huang, Jiazhen Li, Hao Huang, Yingfang Zhu iD

PLOS One · 04.12.2025 · 20 (12) · e0338392

DOI: 10.1371/journal.pone.0338392 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 04.12.2025. Verfasst von Xin Lu und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer Science and Mathematics, Central South University of Forestry and Technology, Changsha, China. Hauptthemen laut MEDLINE: Particulate Matter, Neural Networks, Computer, Environmental Monitoring, Air Pollutants. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Hunan Province, Scientific Research Foundation of Hunan Provincial Education Department.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of PM2.5 concentrations is crucial for public health and environmental management. However, effectively capturing complex spatiotemporal dependencies across multiple time scales remains a persistent challenge for existing methods, particularly in regions with sparse monitoring stations. This study proposes a Multi-Scale Dynamic Graph Neural Network (MSDGNN) for PM2.5 forecasting in station clusters. The model incorporates multi-scale temporal modeling (hourly, daily, weekly) to capture both short- and long-term dependencies. A learnable mapping matrix dynamically groups stations to strengthen spatial correlation learning. Furthermore, MSDGNN employs multi-head attention and spatiotemporal graph attention mechanisms to construct dynamic graphs, utilizing adaptive adjacency matrices and Chebyshev graph convolutions for effective feature propagation. We evaluated MSDGNN on 22 air quality monitoring stations in the Chang-Zhu-Tan region. Results demonstrated that our model reduces MAE by 6.77% and RMSE by 8.67% compared to the best baseline, validating its capability to learn complex dependencies and deliver accurate predictions under diverse spatiotemporal conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Particulate MatterNeural Networks, ComputerEnvironmental MonitoringAir PollutantsAir PollutionSpatio-Temporal AnalysisGraph Neural Networks

Substanzen

Particulate MatterAir Pollutants

Autor:innen und Institutionen

  • College of Computer Science and Mathematics, Central South University of Forestry and Technology, Changsha, China.
  • Hunan Changsha-Zhuzhou-Xiangtan City Cluster National Research Station of Ecosystem, Hunan Botanical Garden, Changsha, China.

ORCID-Profile: Yingfang Zhu

Förderung

  • Natural Science Foundation of Hunan Province
  • Scientific Research Foundation of Hunan Provincial Education Department

Zitieren

APA

Lu, X., Liao, J., Huang, H., Li, J., Huang, H., & Zhu, Y. (2025). Multi-scale dynamic graph neural network for PM2.5 concentration prediction in regional station cluster. PLOS One, 20(12), e0338392. https://doi.org/10.1371/journal.pone.0338392

Vancouver

Lu X, Liao J, Huang H, Li J, Huang H, Zhu Y. Multi-scale dynamic graph neural network for PM2.5 concentration prediction in regional station cluster. PLOS One. 2025;20(12):e0338392. doi:10.1371/journal.pone.0338392

BibTeX
@article{Lu2025,
  author = {Lu, Xin and Liao, Juyang and Huang, Huihua and Li, Jiazhen and Huang, Hao and Zhu, Yingfang},
  title = {Multi-scale dynamic graph neural network for PM2.5 concentration prediction in regional station cluster},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {12},
  pages = {e0338392},
  doi = {10.1371/journal.pone.0338392}
}
Kostenlos starten →