Estimation of daily particle matter concentration (PM2.5) in Ankara Kecioren

Abstract

The particle matter concentration is an important parameter of air quality and used to assess the impact of air on the health and welfare of every living being. The aim of this study is to predict the PM2.5 in Kecioren/Ankara and reveal the influences of meteorological and third parties effect including regional transport through the NOX, temperature and wind velocity by developing a generalized estimating model, that is to say, an ANN model with acceptable accuracy, applicable for air pollution. ANN is a promising modeling technique, especially for highly complex nonlinear problems, using massively parallel-distributed information processing system that simulates the functions of neurons using artificial neurons, inspired by the studies of the brain and the nervous system. The results indicate that the artificial intelligence approach offers a high degree of correlation and its robustness and capability to compute PM 2.5 by using network inputs. The proposed model can be performed to improve air quality.