Author : K. Sudha 1
Date of Publication :20th February 2018
Abstract: Atmospheric pollution is currently one of the most important environmental problems at the global scale. Several forecasting models have been developed with the aim of obtaining the concentrations of atmospheric pollutants. Artificial Intelligence models have been widely used for the prediction of air pollutants, especially the Artificial Neural Networks (ANNs) and Adaptive Network Based Fuzzy Inference System (ANFIS). Study area chosen for the present work is Sanathnagar, Punjagutta, Hyderabad Central University and Pashamylaram areas in Hyderabad, Telangana. The present study includes application of Artificial Neural Networks for the prediction of air pollutant concentrations of NO2, SO2 and PM10 of all selected study areas by using meteorological parameters as inputs. The models are developed based on trial and error method by choosing various number of neurons and number of hidden layers. The best performing network was sought by experimenting combinations of number of neurons and number of hidden layers with respect to Coefficient of Correlation and RMSE (Root Mean square Error).In the present study Adaptive Network based Fuzzy Inference System was also developed for the prediction of NO2, SO2 and PM10 of all selected study areas by using meteorological parameters as inputs.