Author(s)

E. Agirre, G. Ibarra, A. Anta & L. J.R. Barron

Abstract

The goal of this work was to build and evaluate a multilayer perceptron (MLP) based model to forecast O3 and NO2 hourly levels in real-time up to eight hours ahead in the Bilbao area. Two MLP models were designed: the MLP1 model and the MLP2 model. Current and historical hourly values of the meteorological variables, traffic variables, O3 and NO2 concentrations and seasonal input variables were used to determine the input variables of the models. These data were measured in several stations in Bilbao, from 1993 to 1994. The performance of each model was evaluated using the statistics of the Model Validation Kit and it was compared with persistance of levels and observed values. This work showed that the MLP2 model, a multilayer perceptron based model with seasonal inp

Keywords