A neural network-based method for short-term predictions of ambient SO2 concentrations in highly polluted industrial areas of complex terrain

Abstract

A new method for short-term air pollution prediction is described, based on the neural network. It was developed for prediction for SO₂ pollution around the biggest Slovenian thermal power plant at Sostanj. Because of the high SO₂ emissions, there is a need for a reliable air pollution prediction method that would enable lowering the peaks of pollutant concentrations in critical meteorological situations. In complex topography, classical methods for air pollution modelling are not reliable enough. The results obtained by this new method are very promising.

The method can also be used, with slight modifications, for other important air pollutants, the concentrations of which can be measured continuously.