Scholarlink

<p>High-precision prediction of near-surface PM<sub>2.5</sub> concentration is a significant theoretical prerequisite for effective monitoring and prevention of air pollution, and also provides guiding suggestions for the prevention and control of PM<sub>2.5</sub>-related health risks. It has been acknowledged that existing PM<sub>2.5</sub> prediction models predominantly rely on variables influenced by near-surface factors. This inherent limitation could hinder the comprehensive exploration of the continuous spatio-temporal characteristics associated with PM<sub>2.5</sub>. In this study, an optimal 7-day prediction model for PM<sub>2.5</sub> concentration based on the Stacking algorithm was constructed based on multi-source data mainly including atmospheric environment ground monitoring station data, MODIS remote sensing-derived aerosol optical depth (AOD) daily data and meteorological factors. The findings indicated that the PM<sub>2.5</sub> forecasting outcomes derived from this integrated RF-LSTM-Stacking model exhibited a superior fit, with R², RMSE, and MAE values of 0.95, 7.74&#xa0;µg/m³, and 6.08&#xa0;µg/m³, correspondingly. This approach enhanced the accuracy of prediction to a degree of approximately 17% in comparison with a solitary machine learning model. The findings of this study demonstrated that the integration of the LSTM-RF model with the fusion-based Stacking algorithm led to a substantial enhancement in the accuracy of PM<sub>2.5</sub> predictions. This model was found to serve as an effective reference for the monitoring of PM<sub>2.5</sub> prediction and early warning systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

PM2.5 concentration 7-day prediction in the Beijing–Tianjin–Hebei region using a novel stacking framework

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月1日

查看全文
  • Xintong Gao,
  • Xiaohong Wang,
  • Fuping Li,
  • Wenhao Jiang,
  • Meng Zhe,
  • Jiaxing Sun,
  • Ao Zhang,
  • Linlin Jiao

摘要

High-precision prediction of near-surface PM2.5 concentration is a significant theoretical prerequisite for effective monitoring and prevention of air pollution, and also provides guiding suggestions for the prevention and control of PM2.5-related health risks. It has been acknowledged that existing PM2.5 prediction models predominantly rely on variables influenced by near-surface factors. This inherent limitation could hinder the comprehensive exploration of the continuous spatio-temporal characteristics associated with PM2.5. In this study, an optimal 7-day prediction model for PM2.5 concentration based on the Stacking algorithm was constructed based on multi-source data mainly including atmospheric environment ground monitoring station data, MODIS remote sensing-derived aerosol optical depth (AOD) daily data and meteorological factors. The findings indicated that the PM2.5 forecasting outcomes derived from this integrated RF-LSTM-Stacking model exhibited a superior fit, with R², RMSE, and MAE values of 0.95, 7.74 µg/m³, and 6.08 µg/m³, correspondingly. This approach enhanced the accuracy of prediction to a degree of approximately 17% in comparison with a solitary machine learning model. The findings of this study demonstrated that the integration of the LSTM-RF model with the fusion-based Stacking algorithm led to a substantial enhancement in the accuracy of PM2.5 predictions. This model was found to serve as an effective reference for the monitoring of PM2.5 prediction and early warning systems.