Modeling and Prediction of Ozone Concentration in City of Mashhad, Using Neuro-Fuzzy ANFIS Systems

Growth of population and over consumption of fossil fuel has caused an increase in the degree of pollution in the last decade. Owing to the environmental awareness, it seems necessary to find a way to predict the amount of pollutants such as ozone which is toxic. Neuro-fuzzy systems are a powerful tool for modeling and simulation of such processes. In this work, we have adopted ANFIS algorithm for predicting the level of ozone concentration in the city of Mashhad. The inputs of the algorithm are the concentration of hydrocarbons (except methane), mono and dioxide of nitrogen, temperature, wind speed and direction. For the output, the concentration of ozone has been chosen. The percentage of error for this work are 1.8 and 2.4 for the training and trial data, respectively. The result of this research reveals that the error resulted from LSE are lower compared to the combined algorithm