Topology Identification Using Graph Theory Informed State Estimation-Based Model Selection for Power Distribution Systems

This article presents a graph theory-informed approach to state estimation-based model selection for identifying the operational topology of a power distribution system. The proposed method is implemented in two stages, where the first stage is a graph-based candidate topology enumeration, and the second stage is a state estimation-based model selection. The first stage shrinks the sample space of possible switching combinations to provide candidate topologies for the second stage. The second stage solves state estimation for each candidate topology to obtain the system variables consistent with the system measurements. A candidate topology with the least objective is the most likely topology for the given measurements. The accuracy of the proposed approach is quantified using IEEE 123-bus and relevant performance metrics. It is demonstrated that the proposed method is robust against the high percentage of measurement errors and is computationally efficient in identifying the topology by reducing the search space by 96.87%.