EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

4344. Electrical energy demand forecasting for Türkiye

Invited abstract in session WD-28: Advancements of OR-analytics in statistics, machine learning and data science 11, stream Advancements of OR-analytics in statistics, machine learning and data science.

Wednesday, 14:30-16:00
Room: 065 (building: 208)

Authors (first author is the speaker)

1. M. Mücahit DENK
Graduate School of Sciences and Engineering, Koç Üniversitesi
2. Ceyda Oguz
Department of Industrial Engineering, Koc University

Abstract

This study aims to compare the performances of different forecasting techniques to predict electrical energy consumption of Türkiye. The literature on energy demand forecasting generally focuses on multiple regression analysis and time series methods such as ARMA, ARIMA, SARIMAX methods. More recently, machine learning methods have become popular and artificial neural networks are particularly successful in predictions made using time series data.
In this study, the energy demand for Türkiye will be predicted by using the data set produced by the local authorities such as Load Dispatch Information System and Load Dispatch Directorates. In addition, economic and environmental factors affecting electricity use will be determined and included in the analysis. For example, economic indicators such as the number of companies, population, low income per capita, and indicators created from air temperature, business days, holidays, festivals, special days of the Hijri and Gregorian calendars will be included in the analysis.
As the forecasting method, multiple regression, time series methods, SARIMAX and machine learning methods will be used. Then the performance of these methods will be evaluated, and the best one will be selected.

Keywords

Status: accepted

Back to the list of papers