Evaluation and Prediction of Consumption of Volatile Organic Compounds in Refueling Situations Using Neural Network Model

One of the most important pollutants that cause air pollution in cities is volatile organic compounds that cause many complications in people. Gasoline pumps as well as cars are among the most important sources that cause the accumulation of these pollutants, and due to the increasing number of cars, gasoline pumps have become a dangerous place that should be taken into consideration. Since air pollution is a very complex process that depends on many factors, it is very difficult and expensive to predict such data, which have nonlinear dynamics. In this research, by collecting experimental data from three gas stations in Zanjan and identifying the influencing parameters, modeling has been done using artificial neural network. That, two multilayer perceptron models and radial basis function were investigated. In the statistical part of this research, the correlation coefficient and the sum of squared errors were used as necessary criteria to measure the accuracy of the two mentioned models. The results and analysis conducted in this study showed that the pollution resulting from the consumption of gasoline includes two stages, one at the time of refueling and the other after refueling. Also, in the summer, the concentration of volatile organic compounds is higher than in the winter, and this amount of pollution is observed more in the morning and evening.