Ghorbani M, Naghipour L, Karimi V, Farhoudi R. Sensitivity Analysis of the Effective Input Parameters upon the Ozone Concentration using Artificial Neural Networks. ijhe 2013; 6 (1) :11-22
URL:
http://ijhe.tums.ac.ir/article-1-5104-en.html
1- , Ghorbani@tabrizu.ac.ir
Abstract: (11310 Views)
Background and Objectives: Weather pollution, caused by Ozone (O3) in metropolitans, is one of the major components of pollutants, which damage the environment and hurt all living organisms. Therefore, this study attempts to provide a model for the estimation of O3 concentration in Tabriz at two pollution monitoring stations: Abresan and Rastekuche.
Materials and Methods: In this research, Artificial neural networks (ANNs) were used to consider the impact of the meteorological and weather pollution parameters upon O3 concentration, and weight matrix of ANNs with Garson equation were used for sensitivity analysis of the input parameters to ANNs.
Results: The results indicate that the O3 concentration is simultaneously affected by the meteorological and the weather pollution parameters. Among the meteorological parameters used by ANNs, maximum temperature and among the air pollution parameters, carbon monoxide had the maximum effect.
Conclusion: The results are representative of the acceptable performance of ANNs to predict O3 concentration. In addition, the parameters used in the modeling process could assess variations of the ozone concentration at the investigated stations.
Type of Study:
Research |
Subject:
Air Received: 2012/10/28 | Accepted: 2013/01/24 | Published: 2013/10/29