Modeling of suspended particles concentrations in the urban area using artificial intelligence methods

Abstract

The paper develops unique and reliable models for predicting PM2,5 for the City of Tuzla based on the existing monitoring results of PM2,5 and meteorological data (pressure, temperature, wind and humidity) using statistical methods, neural network modeling and genetic programming methods. A correlation between the concentration of pollutants and the influence factors such as temperature and wind has been demonstrated. The developed models can be used for the prediction of PM2,5 concentrations for the early warnings and public protection from the harmful effects of polluted air on human health. The obtained results can be used in the process of making strategic decisions and activities related to air quality control and management. Designing of suspended materials concentration in urban areas is very significant when regular measurements are performed, but the measurements of polluting materials are often lacking. In case of the interruption of the pollutants concentration measurements in Tuzla City for a short or longer time, appliance of the model that is resulting from this work can predict the concentration of pollutants and plan actions based on them.