Electrical energy consumption forecasting using regression method considering temperature effect for distribution network
Yükleniyor...
Dosyalar
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Springer
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Load profile coefficients (LPCs) represent the pattern of electricity usage daily and yearly for electrical energy consumers. It is important to determine the LPCs accurately and reliably, in order to minimize the imbalance costs in the Electricity Energy Market. Reliable methods and sufficient measurement data are required to make accurate forecasts. The local distribution company (TLDC) already calculates the profile coefficients by taking the average of the consumptions without meteorological measurements in Turkey. TLDC determines the LPC by receiving hourly consumption data directly from the consumers. In this paper, the mathematical forecasting models (MFMs) have been produced for determining LPC Duzce in Turkey using the multiple regression analysis method for the first time. Firstly, hourly electrical energy consumption and meteorological temperatures were measured in some predetermined residential subscribers. The MFMs have been produced by using the measured data, and then, LPCs have been determined by using the MFMs. The electrical energy consumptions have been estimated using the determined LPCs, and the estimation results have been compared with the measurement data. The MFMs have been subjected to suitability tests accepted in the literature, and the performances of the models have been verified. According to the results obtained, it has been seen that the MFMs can estimate loads with an accuracy of up to 96% depending on the future changing meteorological conditions, and it has been proposed as a quick and practical method for LPCs calculation. The paper shows that the produced MFMs provide obtaining satisfactory results for energy consumption forecasting for Duzce in Turkey.
Açıklama
Anahtar Kelimeler
Time Series Method; Load Forecasting; Multiple Regression Analysis, Support Vector Regression; Quantile Regression; Neural-Network; Load; Model; Arima
Kaynak
Electrical Engineering
WoS Q Değeri
Q3
Scopus Q Değeri
Q2
Cilt
104
Sayı
5
