CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration
PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> is a type of air pollutant that can cause respiratory and cardiovascular problems. Precise PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math>&...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10319682/ |
| _version_ | 1797400993697103872 |
|---|---|
| author | Prasanjit Dey Soumyabrata Dev Bianca Schoen Phelan |
| author_facet | Prasanjit Dey Soumyabrata Dev Bianca Schoen Phelan |
| author_sort | Prasanjit Dey |
| collection | DOAJ |
| description | PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> is a type of air pollutant that can cause respiratory and cardiovascular problems. Precise PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> (<inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula>) concentration prediction may help reduce health concerns and provide early warnings. To better understand air pollution, a number of approaches have been presented for predicting PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentrations. Previous research used deep learning models for hourly predictions of air pollutants due to their success in pattern recognition, however, these models were unsuitable for multisite, long-term predictions, particularly in regard to the correlation between pollutants and meteorological data. This article proposes the combine deep network (CombineDeepNet), which combines multiple deep networks, including a bidirectional long short-term memory, bidirectional gated recurrent units, and a shallow model represented by fully connected layers, to create a hybrid forecasting system. It can effectively capture the complex relationships between air pollutants and various influencing factors to predict hourly PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentrations in multiple monitoring sites based in China. The best root mean square error achieved was 22.0 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (long-term) and 6.2 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (short-term), with mean absolute error values of 3.4 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (long-term) and 2.2 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (short-term). In addition, the correlation coefficient (R<inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula>) reached 0.96 (long-term) and 0.83 (short-term) across six monitoring sites. These results demonstrate that CombineDeepNet enhances prediction accuracy compared with popular deep learning methods. Therefore, CombineDeepNet proves to be a important framework for predicting PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentration. |
| first_indexed | 2024-03-09T02:03:29Z |
| format | Article |
| id | doaj.art-21b82183bb8644ffb4337ee93db58944 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-09T02:03:29Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-21b82183bb8644ffb4337ee93db589442023-12-08T00:02:02ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-011778880710.1109/JSTARS.2023.333326910319682CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> ConcentrationPrasanjit Dey0https://orcid.org/0000-0002-2248-1701Soumyabrata Dev1https://orcid.org/0000-0002-0153-1095Bianca Schoen Phelan2https://orcid.org/0000-0001-8487-3690School of Computer Science, Technological University Dublin, Dublin, IrelandADAPT SFI Research Centre, School of Computer Science, University College Dublin, Belfield, IrelandSchool of Computer Science, Technological University Dublin, Dublin, IrelandPM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> is a type of air pollutant that can cause respiratory and cardiovascular problems. Precise PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> (<inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula>) concentration prediction may help reduce health concerns and provide early warnings. To better understand air pollution, a number of approaches have been presented for predicting PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentrations. Previous research used deep learning models for hourly predictions of air pollutants due to their success in pattern recognition, however, these models were unsuitable for multisite, long-term predictions, particularly in regard to the correlation between pollutants and meteorological data. This article proposes the combine deep network (CombineDeepNet), which combines multiple deep networks, including a bidirectional long short-term memory, bidirectional gated recurrent units, and a shallow model represented by fully connected layers, to create a hybrid forecasting system. It can effectively capture the complex relationships between air pollutants and various influencing factors to predict hourly PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentrations in multiple monitoring sites based in China. The best root mean square error achieved was 22.0 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (long-term) and 6.2 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (short-term), with mean absolute error values of 3.4 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (long-term) and 2.2 <inline-formula><tex-math notation="LaTeX">$\mu {\text {g/m}} ^{3}$</tex-math></inline-formula> (short-term). In addition, the correlation coefficient (R<inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula>) reached 0.96 (long-term) and 0.83 (short-term) across six monitoring sites. These results demonstrate that CombineDeepNet enhances prediction accuracy compared with popular deep learning methods. Therefore, CombineDeepNet proves to be a important framework for predicting PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> concentration.https://ieeexplore.ieee.org/document/10319682/Air pollutionbidirectional gated recurrent units (BiGRU)bidirectional long short-term memory (BiLSTM)prediction |
| spellingShingle | Prasanjit Dey Soumyabrata Dev Bianca Schoen Phelan CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Air pollution bidirectional gated recurrent units (BiGRU) bidirectional long short-term memory (BiLSTM) prediction |
| title | CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration |
| title_full | CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration |
| title_fullStr | CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration |
| title_full_unstemmed | CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration |
| title_short | CombineDeepNet: A Deep Network for Multistep Prediction of Near-Surface PM<inline-formula><tex-math notation="LaTeX">$_{2.5}$</tex-math></inline-formula> Concentration |
| title_sort | combinedeepnet a deep network for multistep prediction of near surface pm inline formula tex math notation latex 2 5 tex math inline formula concentration |
| topic | Air pollution bidirectional gated recurrent units (BiGRU) bidirectional long short-term memory (BiLSTM) prediction |
| url | https://ieeexplore.ieee.org/document/10319682/ |
| work_keys_str_mv | AT prasanjitdey combinedeepnetadeepnetworkformultisteppredictionofnearsurfacepminlineformulatexmathnotationlatex25texmathinlineformulaconcentration AT soumyabratadev combinedeepnetadeepnetworkformultisteppredictionofnearsurfacepminlineformulatexmathnotationlatex25texmathinlineformulaconcentration AT biancaschoenphelan combinedeepnetadeepnetworkformultisteppredictionofnearsurfacepminlineformulatexmathnotationlatex25texmathinlineformulaconcentration |