Back to Search
Start Over
Urban PM2.5 Concentration Prediction via Attention-Based CNN–LSTM.
- Source :
- Applied Sciences (2076-3417); 3/15/2020, Vol. 10 Issue 6, p1953, 17p
- Publication Year :
- 2020
-
Abstract
- Urban particulate matter forecasting is regarded as an essential issue for early warning and control management of air pollution, especially fine particulate matter (PM<subscript>2.5</subscript>). However, existing methods for PM<subscript>2.5</subscript> concentration prediction neglect the effects of featured states at different times in the past on future PM<subscript>2.5</subscript> concentration, and most fail to effectively simulate the temporal and spatial dependencies of PM<subscript>2.5</subscript> concentration at the same time. With this consideration, we propose a deep learning-based method, AC-LSTM, which comprises a one-dimensional convolutional neural network (CNN), long short-term memory (LSTM) network, and attention-based network, for urban PM<subscript>2.5</subscript> concentration prediction. Instead of only using air pollutant concentrations, we also add meteorological data and the PM<subscript>2.5</subscript> concentrations of adjacent air quality monitoring stations as the input to our AC-LSTM. Hence, the spatiotemporal correlation and interdependence of multivariate air quality-related time-series data are learned by the CNN–LSTM network in AC-LSTM. The attention mechanism is applied to capture the importance degrees of the effects of featured states at different times in the past on future PM<subscript>2.5</subscript> concentration. The attention-based layer can automatically weigh the past feature states to improve prediction accuracy. In addition, we predict the PM<subscript>2.5</subscript> concentrations over the next 24 h by using air quality data in Taiyuan city, China, and compare it with six baseline methods. To compare the overall performance of each method, the mean absolute error (MAE), root-mean-square error (RMSE), and coefficient of determination (R<superscript>2</superscript>) are applied to the experiments in this paper. The experimental results indicate that our method is capable of dealing with PM<subscript>2.5</subscript> concentration prediction with the highest performance. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 20763417
- Volume :
- 10
- Issue :
- 6
- Database :
- Complementary Index
- Journal :
- Applied Sciences (2076-3417)
- Publication Type :
- Academic Journal
- Accession number :
- 142616989
- Full Text :
- https://doi.org/10.3390/app10061953