Topology Identification For Low Voltage Distribution Network Based On Measurement Data

With the continuous development of distributed energy,higher requirements have been put forward for the stable operation of low-voltage distribution networks.Wide scope and large scale are the main characteristics of low-voltage distribution networks,among which the topological relationship of the lines often changes due to structural modifications and transformer replacement in the substation area.The marketing department also needs accurate topological information to effectively manage the line loss in the substation area.This article conducts research on the clear topology identification between the existing "transformer user" in the low-voltage distribution station area.The main research content of this thesis includes:1.Introduce the technical background and significance of this topic.Firstly,explain the identification problems of the "transformer line phase household" in the low-voltage distribution station area.Secondly,analyze the current research status of three types of topological relationship problems,including the "transformer relationship" between transformers in different substation areas and fuzzy users within the radiation range,and the "line household relationship" between the distribution transformer side outgoing lines and end users in the same substation area,Identify the phase relationship of singlephase users within the power supply range of the same substation,and introduce the research routes and methods for three types of problems at home and abroad.2.Based on the existing structure and measurement system of low-voltage distribution system,this thesis introduces the low-voltage distribution transformer topology and the existing distribution network measurement big data architecture in detail,and comprehensively introduces the structure,function and technical parameters of smart meter in the acquisition architecture,laying the groundwork for the subsequent use of measurement data to identify topology.3.Describe the identification of substation substation transformer relationship,use voltage measurement data for K-means clustering,iterate the clustering center through Euclidean distance,and perform correlation analysis with the three-phase voltage measurement data of the substation substation gateway based on fuzzy user attribution.Effectively identify the substation attribution of fuzzy users in two adjacent substations,and achieve the topological relationship identification between fuzzy users and power supply substations in adjacent substations.4.Analyze the identification problem of different lines and end users at the same substation’s distribution transformer outlet.By combining voltage and active power measurement data within the sampling time series,principal component analysis and Kmeans clustering algorithm are combined to overcome the impact of using a single data for line user relationship identification.Due to the clustering deviation of the substation’s output line side,the overall identification of each level branch line connected to the same substation area is carried out,achieving correct identification of the topological relationship between different lines and end users.5.To address the current issue of phase recognition for single-phase users.Combining the three-phase electricity data collected at the gateway with the single-phase user electricity data,principal component analysis is used to reduce the dimensionality of the measurement data,extract the feature matrix,and combine the concept of graph theory directed tree to obtain the constraint matrix and regression matrix of phase and user through singular value decomposition.After iteration,the constraint matrix is transposed,and the correlation between single-phase users and phase is analyzed.Finally,graph theory visualization is performed on it,Realize phase recognition for single-phase users.