Scholarlink

<p>In order to reduce the harmful effects of pollution on the environment and human health, air pollution forecasting plays a crucial role in air quality management. The Yangtze River Delta (YRD) region, one of the largest urban agglomerations in northern China, has observed poor air quality and atmospheric pollution due to the recent rise of the industrial sector and automobile emissions. Research methodology in this study uses a machine learning approach, namely random forest (RF) model to forecast the YRD region’s ambient PM<sub>2.5</sub> levels and meteorological parameters, using data spanning from January 1, 2018, to December 31, 2023. To assess the accuracy of the model, the cross-validation (CV), mean absolute error (MAE), root mean squared error (RMSE), determination coefficient R<sup>2</sup> were applied. The ambient PM<sub>2.5</sub> concentrations were well-predicted by the RF model, with R<sup>2</sup>, RMSE, and MAE values of 0.78, 9.21&#xa0;µg/m<sup>3</sup>, and 7.12&#xa0;µg/m<sup>3</sup>, respectively. Seasonal analysis revealed that the summer had the lowest RMSE and MAE values, indicating the least polluted and cleanest air, while the winter had the highest RMSE and MAE values, indicating the air was most polluted. The results are useful for air quality management and can be applied to similar research in other areas, which proves the need for environmentally friendly approaches.</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

Predicting ambient PM2.5 levels in the Yangtze River Delta with random forest algorithms: A machine learning approach

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Ahmad Hasnain,
  • Muhammad Zaffar Hashmi,
  • Uzair Aslam Bhatti,
  • Basit Nadeem,
  • Geng Wei,
  • Waqas Akram Cheema,
  • Muhammad Asif

摘要

In order to reduce the harmful effects of pollution on the environment and human health, air pollution forecasting plays a crucial role in air quality management. The Yangtze River Delta (YRD) region, one of the largest urban agglomerations in northern China, has observed poor air quality and atmospheric pollution due to the recent rise of the industrial sector and automobile emissions. Research methodology in this study uses a machine learning approach, namely random forest (RF) model to forecast the YRD region’s ambient PM2.5 levels and meteorological parameters, using data spanning from January 1, 2018, to December 31, 2023. To assess the accuracy of the model, the cross-validation (CV), mean absolute error (MAE), root mean squared error (RMSE), determination coefficient R2 were applied. The ambient PM2.5 concentrations were well-predicted by the RF model, with R2, RMSE, and MAE values of 0.78, 9.21 µg/m3, and 7.12 µg/m3, respectively. Seasonal analysis revealed that the summer had the lowest RMSE and MAE values, indicating the least polluted and cleanest air, while the winter had the highest RMSE and MAE values, indicating the air was most polluted. The results are useful for air quality management and can be applied to similar research in other areas, which proves the need for environmentally friendly approaches.