Vulnerability Assessment Tool for Large Radial Distribution System with High Integration of Distributed Energy Resources

The threat of cyber and physical attack-based power grid disruption has increased significantly with the growing penetration of renewable energy-based distributed energy resources (DERs), sensors, and communication devices. This research develops a simulation-based framework using open-source tools to characterize power system vulnerability to the performance of DER assets. The vulnerability assessment framework involves a pre-processing step where geographically closer DERs are grouped using machine learning approaches. The clustering ensures that similar DERs are grouped and reduces the computational complexity of the assessment. Two types of vulnerability assessment methods are proposed - first, where DER clusters are disconnected in sequence, and the second, where all combinations of DER clusters are disconnected. The algorithms identify the node cluster with maximum absolute average voltage deviation. The benefits of the vulnerability assessment tool are demonstrated on a modified IEEE 8500 distribution test feeder with 142 DERs. The combination analysis shows that the top five maximum voltage deviation locations are being predicted with higher than 95% confidence in the random cluster attack analysis case. This study also shows that the DER clusters with the highest capacity need not always contribute to the maximum system vulnerability. The developed tool can be used by the system operator to develop countermeasures for cyber or physical attacks.