Abstract
The purpose of the study in this paper is based on a detailed analysis of various mathematical and statistical methods for determining the similarity and uniformity of daily schedules of electrical load to form an approach to solving several important tasks. Namely, method of forming a static sample of complex load measurement data for the same included composition of single electrical receivers, selection of the optimum declared schedule of electric loading at purchase of the electric power on "Energorinka", as well as solving other similar tasks, all this in general requires effective clustering of graphs.
 Due to the fact that daily schedules of electrical load are divided into working days, pre-weekend, pre-holiday and weekend in this paper it is proposed to express the data of electric load graphs in Cartesian or polar coordinates. This allowed us to record the difference between the daily schedule of electrical load on normal working days from the day before.
 The efficiency of application of these methods of search of identical and similar daily schedules of electric loading was analyzed by allocation of several basic types of similarity of time series, similar: in time, on the form, on structure.
 It is shown that the calculation of the Euclidean distance allows to determine the uniformity of daily load graphs. And phase analysis is their similarity. The most acceptable method is to determine the correlation coefficient, which reveals the similarity and similarity of daily schedules of electrical load.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
