Air Pollution Forecasting using Neural Networks in an Industrialized Area

G. Ibarra-Berastegi, J. Saenz, A. Ezcurra, J.D. Argandoa, A. Elias, and A. Barona (Spain)


Air pollution forecasting, neural networks, fluid mechanics, chemical engineering, applied physics.


Bilbao and surroundings are an industrialized area located in north central Spain. The objective of this paper is to obtain short-term predictions of air pollution for this area. The predictions are forthcoming hourly levels up to 8 hours ahead for five pollutants (SO2, CO, NO2, NO and O3) in the area of Bilbao (Spain). Traffic, meteorological and air pollution network data corresponding to years 2000 and 2001 have been used. 216 specific models based on different types of neural networks (NN), have been built using data corresponding to year 2000. The candidate types of networks were linear, Multilayer Perceptron (MLP), Radial Basis Functions (RBF) and Generalized Regression Neural Networks (GRNN). For each of the 216 cases, the choice of the best model has been made under the criteria of simultaneously having at a 95% confidence level the best values of R2 , d, FA2 and RMSE when applied to data of year 2001. Depending on the pollutant, location and number of hours ahead the prediction is made, different architectures have been selected. In spite of the different architectures and also the different explanatory mechanisms involved the performance of the selected models is very similar.

Important Links:

Go Back