About the article
K.B. Korneev, Yu.M. Pavlova, R. Osei-Оvusu
The necessity of power consumption forecasting is indicated, which requires a qualitative and quantitative analysis of existing load curves for different groups of power consumers. It is noted that the use of intelligent accounting systems allows you to accumulate a large amount of statistical data that require further processing. Due to the fact that visual methods for identifying and grouping data are costly and not always accurate, it is proposed to average the collected statistical data and identify typical profiles (patterns) in them, according to which subsequent measurements can be attributed to a separate group (cluster) within the entire sample of load profiles.
power system, load profile, pattern, forecasting, electrical analyzer, cluster, analysis, control, tariff