Critical Measurement Identification Based on Observability and Computability for Distribution Networks

Modern distribution systems, characterized by high penetration of distributed energy resources (DERs), flexible loads, and reconfigurable topologies, present increasing challenges for accurate system observation and state estimation (SE). However, traditional SE methods, especially those based on weighted least squares (WLS), are inadequate due to limited measurements and incomplete topology information, and they often require extensive sensor deployment to achieve better observability and computability. To address this gap, this paper proposes a novel critical measurement identification method that integrates topological observability, numerical observability, and a newly designed entropy-based computability index. The method not only identifies unobservable branches but also evaluates the effectiveness of additional measurements in reducing state uncertainty, guiding cost-effective measurement deployment. The proposed approach is validated on an IEEE 33-node test system. Simulation results demonstrate its ability to improve observability and SE accuracy with minimal additional measurement investment, making it well-suited for practical active distribution systems.