EUVIMEDEUVIMED entdecken
Crossref

PM2.5 concentration prediction using a whale optimization algorithm based hybrid deep learning model in Beijing, China

Qing Wei, Huijin Zhang, Ju Yang, Bin Niu, Zuxin Xu

Environmental Pollution · 2025-04 · 371 · 125953

DOI: 10.1016/j.envpol.2025.125953 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Pollution, 2025-04. Verfasst von Qing Wei und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China; Key Laboratory of Urban Water Supply, Water Saving and …. Hauptthemen laut MEDLINE: Particulate Matter, Air Pollutants, Environmental Monitoring, Deep Learning. Die Arbeit wurde bisher 31-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: Tongji University, National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PM2.5 is a significant global atmospheric pollutant impacting visibility, climate, and public health. Accurate prediction of PM2.5 concentrations is critical for assessing air pollution risks and providing early warnings for effective management. This study proposes a novel hybrid machine learning model that combines the whale optimization algorithm (WOA) with a convolutional neural network (CNN), long short-term memory (LSTM), and an attention mechanism (AM) to predict daily PM2.5 concentrations. Tested with meteorological and air pollution daily data from 2014 to 2018, the WOA-CNN-LSTM-AM model demonstrates substantial improvements. It achieves MAE, RMSE, MBE, and R2 values of 14.29, 21.96, -0.23, and 0.93, respectively, showing a reduction in prediction errors by 39% compared to CNN and 34% compared to LSTM models. In the medium-term forecast, the accuracy of the hybrid model is 30%-54% over WOA-CNN-LSTM and 26%-39% over CNN-LSTM-AM. The R2 value decreases by 2.5% from the 1-day to 5-day forecast, maintaining high accuracy. SHAP analysis reveals that NO2 and CO are the primary drivers for PM2.5 predictions. This study provides a reliable tool for short and medium-term PM2.5 prediction and air pollution control.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Particulate MatterAir PollutantsEnvironmental MonitoringDeep LearningAir PollutionBeijingAlgorithmsWhalesAnimals

Stichwörter der Autor:innen

Convolutional neural networkLong short-term memoryPM(2.5) predictionShapley additive explanationWhale optimization algorithm

Substanzen

Particulate MatterAir Pollutants

Autor:innen und Institutionen

  • College of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China; Key Laboratory of Urban Water Supply, Water Saving and Water Environment Governance in the Yangtze River Delta of Ministry of…
  • Guangdong Institute of Water Resources and Hydropower Research, Guangzhou, 510000, China.
  • PowerChina East China Survey, Design and Research Institute Co. Ltd., Hangzhou, 310000, China.

Förderung

  • Tongji University
  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Wei, Q., Zhang, H., Yang, J., Niu, B., & Xu, Z. (2025). PM2.5 concentration prediction using a whale optimization algorithm based hybrid deep learning model in Beijing, China. Environmental Pollution, 371, 125953. https://doi.org/10.1016/j.envpol.2025.125953

Vancouver

Wei Q, Zhang H, Yang J, Niu B, Xu Z. PM2.5 concentration prediction using a whale optimization algorithm based hybrid deep learning model in Beijing, China. Environmental Pollution. 2025;371:125953. doi:10.1016/j.envpol.2025.125953

BibTeX
@article{Wei2025,
  author = {Wei, Qing and Zhang, Huijin and Yang, Ju and Niu, Bin and Xu, Zuxin},
  title = {PM2.5 concentration prediction using a whale optimization algorithm based hybrid deep learning model in Beijing, China},
  journal = {Environmental Pollution},
  year = {2025},
  volume = {371},
  pages = {125953},
  doi = {10.1016/j.envpol.2025.125953}
}
Kostenlos starten →