The most important scope of this study is to present a new and
different model, which is not given in the literature, for energy consumption forecasting by using ANN technique in Turkey. This
study focuses on the modeling and forecasting of the energy consumption
in Turkey based on basic energy and demographic and
socio-economic indicators (GDP, population, import, export, and
employment) to analyze the energy consumption using the regression
analyses and the ANN technique. Different models are studied
to determine the best possible approach in modeling energy consumption
as a function of various indicators. This study compares
the accuracy in predicting energy consumption in Turkey among
four different models and also examines the relationship between
energy consumption and diverse independent variables: Model 1
aims to develop a modeling of energy consumption forecast based
on GDP and population; Model 2 aims to develop a modeling of energy
consumption forecast based on economic and demographic
indicators which are the GDP, population, and the amount of import
and export; Model 3 aims to develop a modeling of forecast
taking the effect of employment into consideration in addition to
the independent variables of Model 1; and finally, Model 4 aims
to develop a model of the forecast taking the effect of employment
into consideration in addition to the independent variables of
Model 2. Results arising from this study provide important reference
materials for the utility companies in assessing energy consumption
patterns and selecting a more accurate approach to
estimate future energy consumption.