Determination of forecast parameters to be used in short-term (hourly) electricity consumption forecasts

Electricity demand forecasting is a process that closely concerns all phases of electricity (generation, transmission and distribution) from the beginning to the end. The electricity demand in our country is increasing day by day due to various reasons (population growth, technological development, economic change, etc.). Companies and institutions need to meet this increasing demand. Otherwise, consumers will not be able to receive the electricity they need, and it will be inevitable to experience power cuts. As this result will disrupt business life, it returns to the country as an economic loss. It also has a negative impact on people's well-being. In addition, companies and institutions engaged in generation, transmission and distribution, knowing in advance the increasing demand, foresee in which region, in which time period and in what size they should make their investments. Electricity demand forecasting should be handled in three separate periods. These are short-term, medium-term and long-term forecasts. In this thesis, short-term (hourly) demand forecasting was studied. Each week of the year has been handled separately. With this feature, it differs from the literature. Future demands have been estimated in the light of past consumptions. Box-Jenkins method was used. The performance of the study is shown by comparing the data obtained by using the EViews program with the estimation data in the EPİAŞ Transparency Platform with the Microsoft Excel graphics.