Research On Topology Recognition And Accuracy Evaluation Method Of Low Voltage Distribution Network Containing Source

With the development of Advanced Metering Infrastructure(AMI),big data technology and "smart grid",the research on the topology automatic identification technology of lowvoltage distribution network with source has ushered in new opportunities and challenges.Although AMI technology has brought new means for the topology identification of low voltage distribution network(LVDN),the addition of distributed generation(DG)has changed the correlation between electrical quantities and increased the difficulty of topology intelligent identification with AMI data.This paper focuses on using the advantages of AMI measurement data to solve the problem of automatic identification of the topology of the low-voltage distribution network with source.Main contents of the work are as follows:Firstly,a basic overview of the three-phase four-wire power flow model and algorithm of the low-voltage distribution network with source,and the correlation mechanism analysis of the temporal and spatial characteristics of the electrical data metered from source-containing low-voltage distribution network are provided,which lays a theoretical foundation for the subsequent study on topology automatic identification algorithm and algorithm of the accuracy evaluation.Specifically,according to the structural characteristics of three-phase four-wire,multi-point grounding(repeated grounding),and DG access in the low-voltage distribution network,a suitable power flow model and algorithm are constructed;and the temporal and spatial characteristics of the three electrical quantities is analyzed which includes voltage,current and power.Secondly,it explains the topological structure of LVDN,which is divided into "line-user relationship" and "phase-user relationship",and combining with the temporal and spatial characteristics of the electrical quantity,the internal connection between the electrical quantity and the physical topology of the LVDN is analyzed;Analyzing the correlation of electrical quantity,the vacant users are classified based on the correlation of voltage,and load users and photovoltaic DG are distinguished according to the correlation of power;Based on Kirchhoff’s current law(KCL law),the topology identification model of the low-voltage distribution point network with source is built,and transformed into a quadratic programming problem for solution,and at the same time,the problem is relaxed into a quadratic continuous programming model for easy solution.The simulation results prove that the proposed topology identification method can more accurately identify the topological relationship of the source-containing lowvoltage distribution network compared with the method of using the voltage correlation to identify the topology.Finally,aiming at the problem that the accuracy cannot be confirmed without prior topology,the evaluation method of the accuracy of the source low-voltage distribution network topology recognition mentioned is studied,which considering AMI measurement data quality.The influence degree of the integrity,timing length,similarity,and dissimilarity of metered data on the proposed topology recognition method is explored,and the influence mechanism is analyzed;Based on this influence mechanism,a measurement data quality evaluation model covering five indicators is proposed;Using the positive correlation between the quality score of the measurement data and the recognition accuracy rate,a corresponding topology recognition result accuracy evaluation model is built,which can obtain the range of the recognition accuracy rate based on the quality score of the AMI measurement data.Through simulation experiments,it is proved that the proposed evaluation model is practical and effective.