Hourly Air Quality Index (AQI) Forecasting Using Machine Learning Methods

• Jose Antonio Moscoso-López ^[1] ; Daniel Urda ^[2] ; Javier González-Enrique ^[1] ; Juan Jesus Ruiz-Aguilar ^[1] ; Turias, Ignacio J. ^[1]
  1. [1] Universidad de Cádiz

     Universidad de Cádiz

     Cádiz, España

     
  2. [2] Universidad de Burgos

     Universidad de Burgos

     Burgos, España

     
• Localización: 15th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2020): Burgos, Spain ; September 2020 / coord. por Álvaro Herrero Cosío , Carlos Cambra Baseca , Daniel Urda Muñoz , Javier Sedano Franco , Héctor Quintián Pardo , Emilio Santiago Corchado Rodríguez , 2021, ISBN 978-3-030-57802-2, págs. 123-132
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Air Quality Index (AQI) is an index to inform the daily air quality. AQI is a dimensionless quantity to show the state of air pollution simplifying the information of concentrations in μg/m3. Air quality indexes have been established for each of the five pollutants located in an interesting area to study in as Algeciras (Spain). Hourly data of air pollutants, available during 2010–2015, were analysed for the development of the proposed AQI. This work proposes a two-step forecasting approach to obtain future values, eight hours ahead, of AQI using Machine Learning methods. ANN, SVR and LSTM are capable of modelling nonlinear time series and can be trained to accurately generalize when a new database is presented.