Mass power consumer clustering algorithm based on data-physical feature joint driving

Abstract

The invention discloses a massive power user clustering algorithm based on data-physical characteristic joint driving. The massive power user clustering algorithm comprises the following steps: S1, clustering daily typical load curves of single users; s2, carrying out per-unit processing on the daily typical load data of the user, and calculating the per-unit value of the daily typical load of the user; s3, clustering daily typical load curves of multiple users; s4, automatic iteration is carried out, and the clustering number meeting the contour coefficient and industry concentration degree standard is searched; and S5, outputting a user clustering result corresponding to the optimal clustering number. According to the invention, on the basis of massive user historical power consumption data accumulated in a power consumption information acquisition system, production characteristics and power consumption demands of all industries and users are mined and mastered, so that the load prediction precision and the scheduling management level of a power grid company can be improved; and accurate data support and decision basis can be provided for electricity price formulation, economic dispatching, demand response and the like.

Description

technical field [0001] The present invention relates to a clustering algorithm for massive electric power users, and more specifically, to a clustering algorithm for massive electric power users based on joint drive of data-physical features. Background technique [0002] Based on the massive user historical electricity consumption data accumulated in the electricity consumption information collection system, mining and mastering the production characteristics and electricity demand of various industries and users can not only improve the accuracy of load forecasting and scheduling management level of power grid companies, but also help formulate electricity prices. , economic scheduling, demand response, etc. to provide accurate data support and decision-making basis. [0003] Due to the massive number of users, user clustering is the prerequisite and key influencing factor for user analysis and mining. User clustering methods are divided into two categories: one is the me...

AI Technical Summary

Problems solved by technology

User clustering methods are divided into two categories: one is the method of clustering directly based on user industry attributes. This type of method is simple and intuitive, and has strong physical meaning. Subdivided industries with different characteristics make the clustering results not uniform in electricity consumption characteristics; while the user clustering method based on load characteristics can ensure the consistency of load characteristics within the clustering results, but it lacks A Physical Explanation of the Consistency of Load Characteristics

Images