Modelling of SO2 concentrations using artificial neural networks
[ X ]
Tarih
2006
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
SGEM
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Modelling of air pollution parameters, according to the meteorological data is a necessary for preventing the repetition of same problems. During recent years, neural network-based models have been shown to be powerful tools in the simulation of variations in air quality and provide better alternative to statistical models because of their computational efficiency and generalization ability. In this study, prediction of future daily SO2 concentrations in Konya (Turkey) using MLP (Multilayer Perceptron) artificial neural networks trained with the back-propagation algorithm, which uses gradient descent optimization for error reduction was employed by taking into account meteorological parameters and SO2 (sulphur dioxide) concentrations obtained for two years period from 2003 to 2004. The appropriate architecture of the neural network models was determined through several steps of trainings and testing of the models. The results illustrated that artificial neural networks offer a valuable method for air pollution management.
...
...
Açıklama
Çelebi, Fatma ( Aksaray, Yazar )
Anahtar Kelimeler
Modelling of Air Pollution Parameters, Network-based Models, Hava Kirliliği Parametrelerinin Modellenmesi, Ağ Tabanlı Modeller
Kaynak
International Multidisciplinary Scientific GeoConference
WoS Q Değeri
Scopus Q Değeri
N/A
