Student:	Neslihan KAYA
Supervisor:	Doç. Dr. Şükrü ÖZŞAHİN
Department:	Endüstri Mühendisliği
Institution:	Graduate School of Natural and Applied Sciences
University:	Karadeniz Technical University Turkey

Title of the Thesis:	FORECASTING OF TURKEY’S RENEWABLE ENERGY RESOURCES DEMAND WITH TİME SERİES ESTİMATİON METOD
Level:	M.Sc.
Acceptance Date:	13/6/2017
Number of Pages:	83
Registration Number:	i3208

Summary:

The damage caused by the use of fossil fuels has become a requirement for renewable

energy because of its limited reserves and the obligation to import. This demand, which is

energy, will increase the validity of planning studies to be predicted correctly. In this study

to avoid the harmful effects of nonrenewable energy resources on the environment and to

provide economical advantages, renewable energy resources which is an alternative energy

that we can go towards is (solar, wind, hydropower, geotermal, biomass) introduced,

whose potential and cost are analyzed across Turkey and across the World and avantages

and disavantage are emphasised. When the demand for energy is estimated correctly, it

contributes to investment decisions and import and export rates. With the work done,

estimates of YSA and ARIMA models of each energy source for the next 14 years and the

demand for the next 12 years with gray forecasting model were obtained by using

consumption data between 1990-2014. Arıma and Gray Estimation models established for

evaluate the ANN models and comparisons were made. When the results of these methods

are evaluated, it is seen that the YSA method is much more successful than the other two

methods.

Key Words: Artificial neural network, Time series, Grey forecast, Renewable energy