Analyzing Transmission Lines Robustness and Resilience Prediction Accuracy with Line Graph Based Model

We examine the resilience prediction accuracy of differential centrality and spectral graph measurements, applied to a power systems line-graph based model (LGBM). We test LGBM robustness and related metrics using IEEE test cases. We consider predictive resilience for the LGBM of different IEEE systems. and analyze the accuracy of Sum of Flow Robustness when facing Degree-based attacks (SFRD) and we introduce the Sum of Flow Robustness facing Eigenvalue-based attacks (SFRE). Results show that number of links, average path length, and radius, provide accurate predictions for resilience. We use the weighted networks based on admittance matrix (electrical characteristics) to predict the resilience based on BBM and LGBM. Results show that clustering coefficient, provides accurate predictions for resilience in BBM weighted network, and average path length provides accurate predictions for resilience in LGBM weighted network.