K mean value-based electric power user clustering method and device

Abstract

The invention discloses a K mean value-based electric power user clustering method and device. Electric power consumption loads of all electric power users at each preset time point are ranked, corresponding electric power consumption load sets of the electric power users at all preset time points can be obtained, rankings of a preset clustering number are obtained, electric power consumption loads corresponding to the rankings are respectively chosen from all the electric power consumption load sets, electric power consumption loads corresponding to each ranking are obtained and are initialized as clustering centers of the preset clustering number, and a K mean value clustering algorithm is adopted for clustering the electric power users. Instead of randomly choosing the electric power consumption loads of the electric power users at all preset time points as initialization clustering centers, the method comprises a step of building new electric power consumption loads of the preset clustering numbers at all the preset time points according to all power consumption loads corresponding to all rankings, and a step of initializing the clustering centers; the clustering centers can be reasonably initialized; clustering operation can be performed after the clustering center initialization; accurate clustering results can be obtained, and clustering accuracy can be improved.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular to a K-means-based power user clustering method and device. Background technique [0002] In the process of power system providing power to users, different power users may have different power consumption conditions, and the classification of power users has an important impact on the economic analysis, operation and planning of power systems. At present, the commonly used clustering method for power users is the K-means (K-means) clustering algorithm, which mainly selects K data from the data set S as the center of the initial clustering, and connects each data in the data set to the center closest to the center The central clustering of . First randomly select K data as the initial center, calculate the distance from each data to each selected center, assign the data to the nearest center to form a class, calculate the mean value of each class, and execute the loop repea...

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to address the problem of inaccurate clustering of power users and provide a method and device for clustering power users based on K-means to improve the accuracy of clustering of power users

Images