RAVIPATI KRISHNA REDDY, M. SRIMURALI AND V. SAMPATH KUMAR REDDY
Abstract
This work deals specically with the use of a neural network for Nitrogen Dioxide and ozone modeling. The development of a neural network model is presented to predict the Nitrogen Dioxide and ozone concentrations as a function of meteorological conditions and various air quality parameters. The development of the model was based on the realization that the prediction of Nitrogen Dioxide and ozone from a theoretical basis (i.e. detailed atmospheric diffusion model) is difficult. In contrast, neural networks are useful for modeling because of their ability to be trained using historical data and because of their capability for modeling highly non-linear relationships. The network was trained using four years (2009-2013) meteorological and air quality data. The data were collected from an urban atmosphere. The site was selected to represent a typical residential area with high traffic inuences. Three architecture models were developed. Architecture 1 for the prediction of NO2 by using meteorological parameters as inputs. Architecture 2 for the prediction of O3 by using meteorological parameters including NO2 as inputs. Architecture 3 for the prediction of NO2 and O3 by using meteorological parameters as inputs. The generalization ability of the model is confirmed by correlation and regression between measured and predicted concentrations. The results of this study indicate that the articial neural network (ANN) is a promising method for air pollution modeling.