Predicting Total Hydro Carbons Amount of Air Using Artificial Neural Network

Abstract

In this article, parameters affecting on formation and elimination of hydrocarbons using artificial neural network are considered and a model to predict THC (total hydrocarbon) amount in air using neural network is earned. Also using neural network model and surveying effect of each parameters on THC amount, optimization of offered model is done. The database to get mentioned model consists 1500 samples of current information in two stations of quality control of Tehran city air. Results of using artificial neural network in prediction of THC amount indicate that neural network model is suitable for predicting THC amount. Also to compare improvement of implementing THC prediction model using artificial neural network, a multivariable regression model is used to predict THC amount and its results indicate that MSE is very low when we use artificial neural network.