A general topology identification framework for distribution systems using smart meter and μ-PMU measurements

To enhance topology observability of the power distribution network (PDN), a general topology identification framework using multiple measurement data records of nodal voltages and currents is constructed in this paper. The framework includes three topology identification approaches to handle different situations. The first approach describes a modified three-stage heuristic procedure for the single topology identification while leveraging the sparsity feature of the nodal admittance matrix, which can lead to more accurate identification results than the base model in the literature; The second approach presents a multi-topology identification model to process the measurement data set with a given number of topologies, which classifies records into a fixed number of topology categories and then determines the nodal admittance matrix for each topology category; The third approach designs a two-stage procedure for the measurement data set containing an unknown number of topologies, which determines the number of topologies, the record classification, and the nodal admittance matrices of individual topologies. Effectiveness of the proposed models in efficiently identifying the PDN's topologies is illustrated via several cases, with measurement data sets containing a given or an unknown number of topologies. In addition, the sensitivity analysis quantifies the influence of the number of records, the standard deviation of measurement errors, and the pre-specified maximum number of nonzero elements in each column of the nodal admittance matrix on the accuracy of the identification results.