ESTIMATION OF AIR POLLUTANT CONCENTRATIONS FROM METEOROLOGICAL PARAMETERS USING ARTIFICIAL NEURAL NETWORK

Hamdy K. Elminir - Hala Abdel-Galil

   The lack of environmental data is a common feature of many developing countries. This is fact in Egypt, where air quality is beginning to be systematically monitored in some places of the country. To overcome these problems, the need for accurate estimates of air quality levels becomes evermore important. To achieve such prediction tasks, the use of artificial neural network (ANN) is regarded as a cost effective technique superior to traditional statistical methods. In this paper, ANN trained with a back propagation algorithm is used to estimate the well known pollutants, from readily observable local meteorological data. The results indicate that the ANN model predicted air pollutant concentrations with good accuracy of approximately 96 %.

Keywords: air quality, pollutants, artificial neural network

[full-paper]