الگوسازی و پیش‌بینی مصرف انرژی بخش حمل و نقل ایران: کاربردی از الگوهای هوش مصنوعی

Ever-increasing dependence of human life on energy has made this factor play a critically crucial role either potentially or actively in the functions of various economic sectors of countries. Therefore, the people in charge of any country should try to make exact forecasting of energy consumption and make correct planning about its consumption in a way to optimally control supply-demand parameters. This study tries to make models and forecasts about energy consumption in Iran’s transportation sector through ‘Fuzzy Neural Network’, ‘Genetic Neural Network’ and ‘Neural Network’ models. Hence, annual data of Iran’s transportation sector energy consumption was taken as output variable of forecasting models while the annual data of population of the whole country, gross domestic product and the number of automobiles were used as the input variables of the forecasting model. Finally, the accuracy of models was assessed by using evaluation indicators. The assessment results indicated that ‘Fuzzy Neural Network’ model was more accurate than other ones considering energy forecasts in the country’s transportation sector. Meanwhile, according to the results of the inputs sensitivity analysis by neural network, Iran’s population input was identified as the input with most influence in energy consumption.