A Data-Driven Topology Estimation For Distribution Grid

The topological structure of distribution grid is the basis for the realization of various functions of the future intelligent distribution grid. Aiming at the shortcomings of traditional distribution network topology estimation methods such as large amount of calculation, poor real-time performance, relying on the information provided by the AMI(advanced metering infrastructure) system, a data-driven topology estimation method in radial distribution gild is proposed. First, use the kernel density estimation method to calculate the mutual information between the voltage data of each bus to analyze the buses' connection relationship; then generate the distribution grid topology in the form of adjacency matrix according to the maximum spanning tree algorithm; The proposed method is verified through the IEEE 33-bus system.