Artificial neural network for the identification of unknown air pollution sources

Artificial neural networks (ANN), whose performances to deal with pattern r ecognition problems is well known, are proposed to identify air pollution s ources. The problem that is addressed is the apportionment of a small numbe r of sources from a data set of ambient concentrations of a given pollutant . Three layers feed-forward ANN trained with a back-propagation algorithm a re selected. A test case is built, based on a Gaussian dispersion model. A subset of hourly meteorological conditions and measured concentrations cons titute the input patterns to the network that is wired to recover relevant emission parameters of unknown sources as outputs. The rest of the model da ta are corrupted adding noise to some meteorological parameters and we test the effectiveness of the method to recover the correct answer. The ANN is applied to a realistic case where 24 h SO2 concentrations were previously m easured. Some of the limitations of the ANN approach, together with its cap abilities, are discussed in this paper. (C) 1999 Published by Elsevier Scie nce Ltd. All rights reserved.