|Autors: FilipovaPetrakieva, S. K., Dochev, V.|
Title: Short-term forecasting of hourly electricity power demand, Reggresion and clusters methods for making short-term prognosis
Keywords: short-term prognosis, hourly electricity power demand, reggresion analysis, clusters’ methods
Abstract: In this paper estimates of hourly electrical power demands for 7 days are performed. Data for the period 01.01.2015 – 24.12.2020 is processed. The models are tested with data for the period 25–31.12.2020 and prognosis is made during the same period. Two groups of methods are used to make the prognosis – classical regression methods and clustering algorithms. The first group test "а moving window" and ARIMA. The second group tests K-Means, Time Series K-Means, Mini Batch K-Means, Agglomerative clustering, and OPTICS. The forecasts are compared with the actual electrical power load for the period 25–12.2020 and are retrieved from the website of the Energy System Operator of Bulgaria. The results of the comparison show high accuracy of the forecasts made with the used methods compared with respect to the actual energy load for the prognosis period.
Copyright Engineering, Technology & Applied Science Research Journal (ETASR), Web of Science
1. V. Atanasov, D. Stoilov, Determining the Places for Installation of Remote Power Disconnectors in Overhead Power Lines, 22nd International Symposium on Electrical Apparatus and Technologies, SIELA’2022’.Bourgas, Bulgaria, 1-4 June 2022, Code: 182112, ISBN: 978-166541139-4, DOI: 10.1109/SIELA54794.2022.9845755 https://www.scopus.com/record/display.uri?eid=2-s2.0-85137612309&origin=resultslist&sort=plf-f - 2022 - в издания, индексирани в Scopus или Web of Science
Вид: статия в списание, публикация в издание с импакт фактор, публикация в реферирано издание, индексирана в Web of Science