Publicação
A deep learning-based model to predict nitrogen dioxide in urban environments
| Resumo: | Nowadays, our society faces several problems regarding the environmental sustainability of our planet. One of these problems, which severely impacts human lives, such as climate change, is air pollution. Air pollution in urban environments is derived from road transport and different economic activities and directly impacts human health. Then, air quality monitoring stations are essential to determine potentially dangerous situations. In this context, this work proposes the conception, tunning and evaluation of three Deep Learning Models, namely LSTM, GRU and CNN, to predict the concentration of NO2 in the city of Porto for the next two days, achieving satisfactory results, especially with GRU models, with a RMSE of 8.89 μg/m3. |
|---|---|
| Autores principais: | Oliveira, P. |
| Outros Autores: | Díaz-Longueira, Antonio; Marcondes, Francisco Supino; Durães, Dalila; Calvo-Rolle, José Luis; Jove, Esteban; Novais, Paulo |
| Assunto: | Air pollution Deep learning Nitrogen dioxide Time series |
| Ano: | 2025 |
| País: | Portugal |
| Tipo de documento: | comunicação em conferência |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Nowadays, our society faces several problems regarding the environmental sustainability of our planet. One of these problems, which severely impacts human lives, such as climate change, is air pollution. Air pollution in urban environments is derived from road transport and different economic activities and directly impacts human health. Then, air quality monitoring stations are essential to determine potentially dangerous situations. In this context, this work proposes the conception, tunning and evaluation of three Deep Learning Models, namely LSTM, GRU and CNN, to predict the concentration of NO2 in the city of Porto for the next two days, achieving satisfactory results, especially with GRU models, with a RMSE of 8.89 μg/m3. |
|---|