Abstract:Air pollution is a complex process and is affected by meteorological conditions and other chemical components. Numerous studies have demonstrated that data-driven spatio-temporal prediction models of PM2.5 concentration are comparable with the model-driven model. However, data-driven models are usually depending on the statistical correlation between PM2.5 and other factors and have challenges in dealing with causality in complex systems. In this paper, we argue that domain knowledge should be incorporated into data-driven models to enhance prediction accuracy and make the model more physically realistic. We focus on the influence of dynamic wind-field on PM2.5 concentration distribution and fuse the pollution diffusion distance with the deep learning model based on a wind-field surface. In order to model spatial dependence between monitoring stations, which is dynamic and anisotropic because of the wind-field, we proposed a hybrid deep learning framework, dynamic directed spatio-temporal graph convolution networks (DD-STGCN). It expanded the ability to deal with space-time prediction in the continuous and dynamic wind-field. We used a directed graph time-series to describe the vertex state and topological relationship between vertices and replaced traditional Euclidean distance with wind-field diffusion distance to describe the proximity relationship between vertices. Our experiment results demonstrated that the DD-STGCN model achieved a better prediction ability than LSTM, GC-LSTM, and STGCN models. Compared to the best comparison model, MAPE, MAE, and RMSE were improved by 10.2%, 9.7%, and 9.6% in 12 h on an average, respectively. The performance of our model was further tested during a haze period. In the case that two models both considered the effect of wind, compared with the pure data-driven model, our model performed better in prediction distribution and showed the benefit of spatial interpretability provided by domain knowledge. (C) 2021 Elsevier Ltd. All rights reserved.

MGC-LSTM: a Deep Learning Model Based on Graph Convolution of Multiple Graphs for PM 2.5 Prediction

Forecasting PM2.5 Using Hybrid Graph Convolution-Based Model Considering Dynamic Wind-Field to Offer the Benefit of Spatial Interpretability.

A Graph-Based LSTM Model for PM2.5 Forecasting

A Hybrid CNN-LSTM Model for Forecasting Particulate Matter (PM2.5)

Adaptive Scalable Spatio-Temporal Graph Convolutional Network for PM2.5 Prediction

Long short-term memory - Fully connected (LSTM-FC) neural network for PM 2.5 concentration prediction.

A Theory-Guided Graph Networks Based PM2.5 Forecasting Method.

Fine particulate matter concentration prediction based on hybrid convolutional network with aggregated local and global spatiotemporal information: A case study in Beijing and Chongqing

Prediction of PM2.5 Concentration Based on Deep Learning, Multi-Objective Optimization, and Ensemble Forecast

Long Short-Term Memory based PM2.5 Concentration Prediction Method

Enhancing Air Quality Forecasting: A Novel Spatio-Temporal Model Integrating Graph Convolution and Multi-Head Attention Mechanism

Spatio-temporal fusion of meteorological factors for multi-site PM2.5 prediction: A deep learning and time-variant graph approach

A Hybrid Integrated Deep Learning Model for Predicting Various Air Pollutants

A Hybrid CNN-LSTM Model for Predicting PM2.5 in Beijing Based on Spatiotemporal Correlation

A Deep Learning Model for PM2.5 Concentration Prediction

The Air Quality Prediction Based on a Convolutional LSTM Network.

PM2.5Forecast Based on a Multiple Attention Long Short-Term Memory (MAT-LSTM) Neural Networks

Multi-Source PM2.5 Prediction Model Based on Fusion of Graph Attention Networks and Multiple Time Periods

A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities

Urban PM2.5 Concentration Prediction Via Attention-Based CNN–LSTM

Near-surface PM2.5 prediction combining the complex network characterization and graph convolution neural network