Simulation of Tehran Air Pollution Using Artificial Neural Networks

Abstract

Air pollution is one of the most important environmental issues of the populated cities, which is seriously threatening human health. This is caused by several factors, such as climo-physiologic characteristics of the region, deficiency of perennial plants, vehicles-mobile pollution sources- and also stationary sources including factories and power plants. Vehicles' gas emission is the major source of air pollution which is related to the type and magnitude of fuel they consume. In comparison, the stationary sources only contribute one fifth of the total pollutant emission. In this study, the effect of fuel consumption increase on Tehran (Iran's capital) air pollution is evaluated. Obviously, weather patterns have also considerable role on air pollution which are considered in the proposed simulation model. For this purpose, the most effective parameters on Tehran air pollution are identified and selected as the input of simulation model. Then, Artificial Neural Network (ANN) models are chosen due to their efficient power and robustness in identifying the nature of complicated phenomena which air pollution is one of them. Tehran is selected as a case study because of its higher violation to the air pollution standards and being one of the five polluted cities around the world. Results show that the proposed model can be implemented successfully to monitor the air pollution changes and consequently makes it possible for associated managers to develop appropriate policies against the air pollution.