Short-Term Forecasts of the Electrical Energy Consumption in Republic of Bulgaria

In this paper, a short-term forecast of the electrical energy load on the territory of the Republic of Bulgaria is made. Two classical regression analysis methods ("a moving window" and ARIMA), and six cluster analysis methods (K-Means, Time Series K-Means, Mini Batch K-Means, Agglomerative clustering, and OPTICS) were used for the forecasts. The following libraries have been used during the research: for processing and manipulation of the input data; to present the input data in graphical form and for mathematical calculations. Data are taken from the website of the Energy System Operator (ESO) in Bulgaria for the period 25.03 - 30.06.2021. For all of the methods mentioned above, short-term forecasts for the period 01 - 07 July 2021 are made. These forecasts are compared with the actual load for the same period published on the ESO website. The results of the comparison show a higher accuracy of the methods used by the authors.