ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

MODELLING OF SO2 POLLUTION IN SELÇUKLU DISTRICT OF KONYA WITH ARTIFICIAL NEURAL NETWORKS

Journal: International journal of ecosystems and ecology science (IJEES) (Vol.6, No. 4)

Publication Date:

Authors : ; ;

Page : 543-550

Keywords : Artificial neural network; modelling; air pollution; SO2; meteorological factors;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Some of the air pollutants cause significant effects on health of human and other livings. SO2 is one of these air pollutants that may harm the tissues and mucous membranes of the eyes, disturb throat due to the irritating odour and cause problems in the respiratory system and bronchi. Also, SO2 gas may be lethal for patients who have significant illnesses like lung failure and asthma. In addition, animal life and vegetation are affected negatively from SO2 gas. Depending on the exposure period, chronic injuries occur in the plants such as decrease in growth and yield, increase in senescence, and colour problems. When all problems originating from SO2 are considered, prediction of the future concentration of SO2 is very significant so as to take precautions. In this study SO2 pollution in Selçuklu district which is one of the biggest districts of Konya was tried to be predicted with artificial neural networks using meteorological factors and air pollutants emitted to the area. Artificial neural networks consist of interconnected structures for making parallel computations. The working principle of human brain is used in artificial neural networks. Measurements of pollutants which are O3, NOx, PM10 and meteorological factors such as wind speed, temperature, and humidity made in winter period of 2016 were used as a parameter in this study. These air pollutants and meteorological factors were integrated to the model as input parameters and SO2 concentration of one day and five day later was predicted in order to compare efficiency of the models.

Last modified: 2016-07-18 08:50:31