Application of Time Series Data Clustering in Intelligent Identification of Household Transformer Relationship

Abstract

In order to accurately clarify the household transformer relationship of the voltage distribution network, on the basis of analyzing the shortcomings of existing verification methods in engineering applications, the rapid development of new business in the power grid puts forward the intelligence level of the low-voltage distribution network. The topological relationship of the traditional low-voltage distribution network mainly relies on manual file maintenance, resulting in low efficiency of voltage problem governance and emergency repair. It has been verified by experiments that the proposed algorithm can achieve accurate matching of the affiliation between low-voltage users and the station area, and has engineering practicability. Although the phase table has been calibrated for the phase sequence during data acquisition, due to problems such as non-standard wiring, it is inconsistent with the real phase sequence. Based on the determination of the relationship between the branch and the household meter, a household meter phase identification method that comprehensively utilizes the voltage and current data of the user meter is proposed to improve the accuracy of the household meter phase identification.