A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea
Both long- and short-term exposure to high concentrations of airborne particulate matter (PM) severely affect human health. Many countries now regulate PM concentrations. Early-warning systems based on PM concentration levels are urgently required to allow countermeasures to reduce harm and loss. Pr...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-03-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/11/4/348 |
| _version_ | 1797571787392811008 |
|---|---|
| author | Guang Yang HwaMin Lee Giyeol Lee |
| author_facet | Guang Yang HwaMin Lee Giyeol Lee |
| author_sort | Guang Yang |
| collection | DOAJ |
| description | Both long- and short-term exposure to high concentrations of airborne particulate matter (PM) severely affect human health. Many countries now regulate PM concentrations. Early-warning systems based on PM concentration levels are urgently required to allow countermeasures to reduce harm and loss. Previous studies sought to establish accurate, efficient predictive models. Many machine-learning methods are used for air pollution forecasting. The long short-term memory and gated recurrent unit methods, typical deep-learning methods, reliably predict PM levels with some limitations. In this paper, the authors proposed novel hybrid models to combine the strength of two types of deep learning methods. Moreover, the authors compare hybrid deep-learning methods (convolutional neural network (CNN)—long short-term memory (LSTM) and CNN—gated recurrent unit (GRU)) with several stand-alone methods (LSTM, GRU) in terms of predicting PM concentrations in 39 stations in Seoul. Hourly air pollution data and meteorological data from January 2015 to December 2018 was used for these training models. The results of the experiment confirmed that the proposed prediction model could predict the PM concentrations for the next 7 days. Hybrid models outperformed single models in five areas selected randomly with the lowest root mean square error (RMSE) and mean absolute error (MAE) values for both PM<sub>10</sub> and PM<sub>2.5</sub>. The error rate for PM<sub>10</sub> prediction in Gangnam with RMSE is 1.688, and MAE is 1.161. For hybrid models, the CNN–GRU better-predicted PM<sub>10</sub> for all stations selected, while the CNN–LSTM model performed better on predicting PM<sub>2.5</sub>. |
| first_indexed | 2024-03-10T20:45:34Z |
| format | Article |
| id | doaj.art-0b53b857d49445da882d694ed3bd093a |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-10T20:45:34Z |
| publishDate | 2020-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-0b53b857d49445da882d694ed3bd093a2023-11-19T20:17:33ZengMDPI AGAtmosphere2073-44332020-03-0111434810.3390/atmos11040348A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South KoreaGuang Yang0HwaMin Lee1Giyeol Lee2Department of Computer Science, Soonchunhyang University, Asan 31538, KoreaDepartment of Computer Software & Engineering, Soonchunhyang University, Asan 31538, KoreaDepartment of Landscape Architecture, Chonnam National University, Gwangju 61186, KoreaBoth long- and short-term exposure to high concentrations of airborne particulate matter (PM) severely affect human health. Many countries now regulate PM concentrations. Early-warning systems based on PM concentration levels are urgently required to allow countermeasures to reduce harm and loss. Previous studies sought to establish accurate, efficient predictive models. Many machine-learning methods are used for air pollution forecasting. The long short-term memory and gated recurrent unit methods, typical deep-learning methods, reliably predict PM levels with some limitations. In this paper, the authors proposed novel hybrid models to combine the strength of two types of deep learning methods. Moreover, the authors compare hybrid deep-learning methods (convolutional neural network (CNN)—long short-term memory (LSTM) and CNN—gated recurrent unit (GRU)) with several stand-alone methods (LSTM, GRU) in terms of predicting PM concentrations in 39 stations in Seoul. Hourly air pollution data and meteorological data from January 2015 to December 2018 was used for these training models. The results of the experiment confirmed that the proposed prediction model could predict the PM concentrations for the next 7 days. Hybrid models outperformed single models in five areas selected randomly with the lowest root mean square error (RMSE) and mean absolute error (MAE) values for both PM<sub>10</sub> and PM<sub>2.5</sub>. The error rate for PM<sub>10</sub> prediction in Gangnam with RMSE is 1.688, and MAE is 1.161. For hybrid models, the CNN–GRU better-predicted PM<sub>10</sub> for all stations selected, while the CNN–LSTM model performed better on predicting PM<sub>2.5</sub>.https://www.mdpi.com/2073-4433/11/4/348air qualityparticulate matterlong short-term memorygated recurrent unithybrid models |
| spellingShingle | Guang Yang HwaMin Lee Giyeol Lee A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea Atmosphere air quality particulate matter long short-term memory gated recurrent unit hybrid models |
| title | A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea |
| title_full | A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea |
| title_fullStr | A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea |
| title_full_unstemmed | A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea |
| title_short | A Hybrid Deep Learning Model to Forecast Particulate Matter Concentration Levels in Seoul, South Korea |
| title_sort | hybrid deep learning model to forecast particulate matter concentration levels in seoul south korea |
| topic | air quality particulate matter long short-term memory gated recurrent unit hybrid models |
| url | https://www.mdpi.com/2073-4433/11/4/348 |
| work_keys_str_mv | AT guangyang ahybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea AT hwaminlee ahybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea AT giyeollee ahybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea AT guangyang hybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea AT hwaminlee hybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea AT giyeollee hybriddeeplearningmodeltoforecastparticulatematterconcentrationlevelsinseoulsouthkorea |