Determining Effective Parameters on CO Concentration in Tehran Air by Sensitivity Analysis based on Neural Network Prediction

Abstract

One of the most toxic pollutant gases produced by fossil fuels is carbon monoxide. Hence, the accurate and regular estimation and control of CO in the cities such as Tehran is inevitable. In this research, for the first time, CO concentration in ambient air was predicted based on 12 important urban and meteorological parameters by neural network. Also, the sensitivity analysis of the factors that effect on the concentration of carbon monoxide in Tehran was investigated based on the pollutant concentration predictive model. In this research, the daily statistical data of Tehran metropolis over the course of five consecutive years from 12 factors affecting the amount of carbon monoxide in Tehran, such as population, density, precipitation, temperature, urban traffic, wind speed, gasoil consumption, moisture, air flow, effective vision and air pressure was used. Based on this database, the artificial neural network with the best possible algorithm had been trained to predict this contaminant and root mean square error of model was equal to 2.54. Then, sensitivity analysis was done to find the most effective factor on the concentration of carbon monoxide, urban density and air pressure. In order to control this hazardous contaminant in urban management, these parameters should be taken into account. Based on the result, by preventing the construction of high towers in Tehran, wind speed average will increase and increasing in wind speed (25%) caused to reducing in carbon monoxide concentration (about 12%). Also, prevention of urban density (25%) will cause to prevention of increasing CO concentration (about 10%).

Prediction of Dissolved Oxygen Concentration in Sewage Treatment Process Based on Data Recognition Algorithm

“…Qaderi and others proposed a hybrid model for the anaerobic digestion process. e model is established based on the material balance equation, in which the biological growth rate is expressed by a neural network [3]. Zhang and others studied the dynamic simulation of the activated sludge process, used a neural network to improve the prediction model, and developed a program to improve the accuracy of the existing mechanical model of activated sludge process [4].…”