Line–Household Relationship Identification Method for a Low-Voltage Distribution Network Based on Voltage Clustering and Electricity Consumption Characteristics

Abstract

To address the issue of inconspicuous electricity consumption characteristics among vacant users in low-voltage distribution networks (LVDNs), which hinders effective line–household relationship identification (LHRI), a method for identifying line–household relationship based on voltage clustering and electricity consumption characteristics is proposed. Initially, the paper employs Dynamic Time Warping (DTW) to analyze the similarity of user voltage profiles and utilizes the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to cluster users. This approach identifies the topological relationship between vacant users and regular users to obtain multiple user categories. Subsequently, by analyzing the electricity consumption characteristic, the connection relationships between different user categories and phase lines are clarified based on the correlation between the electricity consumption characteristic vector of phase lines and the electricity consumption characteristic vector of user categories, thereby revealing the line–household relationship for all users. On the test dataset, the LHRI algorithm proposed in this article achieved 100% accuracy, within an allowable error range of 0.2%, and improved the accuracy by 20% compared to the traditional identification method. Finally, the LVDN simulation model established by OpenDSS 9.4.0.3 was used to verify the effectiveness of the proposed method, confirming its potential and advantages in practical applications.