Topology Identification Method of Low-Voltage Distribution Network Based on Regression Analysis of Voltage Characteristic Parameters

At present, there are problems of inaccurate records or untimely update of user-transformer relation and users' phase identification in the distribution network, which affects the operation and management of the low-voltage distribution network. In this paper, a topology identification method of low-voltage distribution network based on regression analysis of voltage characteristic parameters is proposed. Firstly, the slope, peak valley degree and other characteristic parameters of the voltage curve of the transformer and the user are extracted to characterize the voltage curve from the global and local aspects. Then, the linear regression equations of the characteristic parameters of transformer phase voltage and user voltage are established, and user-transformer relation and users' phase relation are identified according to the goodness of fit coefficient of the regression equation. Compared with the regression analysis using voltage data directly, the proposed method is not affected by the line parameters and length, which has a wide application range, and recognizes user-transformer relation and users' phase relation accurately.