Distribution grid topology reconstruction: An information theoretic approach

Recently, a rapidly penetration of distributed generation raises various issues. One of the key issues is frequent distribution grid re-configuration, which is hard to detect based on traditional approaches. Wrong topology information causes wrong control signal, making fast changing smart grid prone to go over stability boundaries and to collapse. To ensure system robustness, we propose a new data-driven re-configuration approach, thanks to recently progressively deployed larger sensor networks in distribution systems by utilities. Specifically, an Information Theory based algorithm, Chow-Liu algorithm, is used based on a proof for assumption and verification in distribution systems. Simulation results show highly accurate re-configuration estimation in IEEE standard distribution test systems.