PM2.5 Prediction with a Novel Multi-Step-Ahead Forecasting Model Based on Dynamic Wind Field Distance

Particulate matter with a diameter of less than 2.5 μ m (PM2.5) has damaged public health globally for a decade. Accurate forecasts of PM2.5 concentration can provide early warnings to prevent the public from hazard exposure. However, existing methods have not considered the available spatiotemporal data sufficiently due to their architecture or inadequate input, and most neglected wind impact on spatiotemporal correlation when selecting related sites. To fill this gap, we proposed a long short-term memory-convolutional neural network based on dynamic wind field distance (LSTM-CNN-DWFD) to predict PM2.5 concentration of a specific site for the next 24 h. A KNN method based on dynamic wind field distance was developed and applied to select highly related sites considering wind impact. A local stateful LSTM model was employed to capture temporal correlations in historical air quality and meteorological data for each related site. Then, these temporal features were integrated as a spatiotemporal matrix, and input into CNN for extracting spatiotemporal correlation features. Weather forecasts were also integrated into the model to promote accuracy. Hourly PM2.5 data from 36 monitoring sites in Beijing, China collected from 1 May 2014 to 30 April 2015 were used as experimental dataset. Six-fold rolling origin method was employed to conduct experiments on each site, and the results of 216 experiments validated the performance of the proposed LSTM-CNN-DWFD model. The mean R 2 values of the next 1–6 h prediction were 0.85, 0.81, 0.76, 0.70, 0.64, and 0.59, respectively, showing a decrease trend, indicating that the prediction accuracy decreases as the prediction time increases. Comparisons of LSTM-CNN-DWFD results to results from six other methods show that it delivered higher accuracy PM2.5 predictions, with the mean RMSE (MAE) of 1–6, 7–12, and 13–24 h prediction were 43.90 (29.17), 57.89 (42.16), and 63.14 (47.64), respectively. The results also demonstrate that the sites selected based on dynamic wind field distance are more related to the central site than that based on geographical distance, also contributing to prediction accuracy

Field (mathematics · Geography · Machine learning · Meteorology · Predictive modelling · Air Quality and Health Impacts · Air Quality Monitoring and Forecasting · Computer Science · Environmental Science · Mathematics · Vehicle emissions and performance

• Application of wavelet-packet transform driven deep learning method in PM2.5 concentration prediction

  Open Access•Qinghe Zheng, Xinyu Tian et al.•Sustainable Cities and Society•2023

  Add to list

• Faster R-CNN

  Open Access•Shaoqing Ren, Kaiming He et al.•IEEE Transactions on Pattern…•2017

  Add to list

• Ambient air pollution

  Open Access•World Health Organization•Clean Air Journal•2016

  Add to list

• Gradient-based learning applied to document recognition

  Open Access•Yann LeCun, Léon Bottou et al.•Proceedings of the IEEE•1998

  Add to list

• Nearest neighbor pattern classification

  Open Access•Thomas M Cover, T Cover et al.•IEEE Transactions on Information…•1967

  Add to list

• Long Short-Term Memory

  Sepp Hochreiter, Jurgen Schmidhuber•Neural Computation•1997

  Add to list