Power consumer information tagging method and device, electronic equipment and storage medium

Abstract

The invention provides a power consumer information tagging method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining three-phase electric meter data ofa user; clustering the three-phase electric meter data to obtain a typical load curve picture; classifying the typical load curve pictures by using a pre-trained VGG model to obtain power utilizationcharacteristic feature tags of the users; combining K-means clustering and a deep convolutional neural network to be used for power consumer information tagging, and obtaining a power consumer behavior portrait. The implementation difficulty is small, the sample size is large, all typical users can be covered, and the accuracy is high.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular to a method and device for labeling electric power user information. Background technique [0002] At present, the competition in the energy market is becoming increasingly fierce, and understanding the needs of massive users is the key for power companies to improve service levels and enhance competitiveness. The development of the Internet of Things and the popularization of smart meters have enabled efficient collection of user data. [0003] User portrait (that is, user information label) is a labeled user model mined based on user data (such as social attributes, living habits, consumption behavior, etc.). The construction of user portraits can make the hidden characteristics of users explicit, help enterprises understand users, gain insight into user needs, discover target users, tap user control potential, finely locate the characteristics of groups of people, and pr...

AI Technical Summary

Problems solved by technology

[0004] At present, most of the methods for analyzing the behavior of power users only directly cluster, classify users with similar power consumption patterns, and further analyze the power consumption behavior and regulation potential of users, for example: based on the annual / monthly load curve, extract the segmented average power, According to the fluctuation rate and other characteristics, analyze the characteristics of users' annual electricity consumption; with the goal of peak shaving and valley filling, analyze and extract the characteristic quantities that characterize the peak and seasonal changes of users' electricity consumption, and study the demand response projects suitable for various users to participate in; Based on the electricity consumption forecast as the main analysis target, based on the economic, climate and electricity price and other related factors, the cluster analysis of user electricity consumption patterns is studied; using the experimental data of electricity consumption behavior of residents and small and medium-sized businesses, the incentive mechanism such as time-of-use electricity price is used to study the influence of user behavior. However, the behavior data used in existing methods are generally obtained through questionnaire surveys and project experiments, which is difficult to implement and the sample size is small, it is difficult to cover all typical users, and the accuracy rate is low

Images