Tri-level optimal hardening plan for a resilient distribution system considering reconfiguration and DG islanding

There is an increasing awareness of resilience in the face of extreme natural disasters and man-made attacks. Power system resilience includes hardening measures and operational restoration measures. These two aspects of resilience measures are innovatively combined in this paper to improve power system resilience. A tri-level defender-attacker-defender (DAD) model is built to find the best hardening plan under malicious attacks given the available defending resources and operational restoration measures for a distribution system. On the first level, the system defender makes hardening decisions. On the second level, the attacker finds the attack scenario with maximum damage. Given the hardening and attack plans, on the third level, the system operator undertakes resilient operational measures including optimal DG islanding formation and topology reconfiguration. This new model overcomes the limitations of existing DAD models by integrating these operational measures into the mathematical model; the produced hardening strategy is more practical for today's fast-developing smart distribution system. The DAD problem is formed as a tri-level mixed integer optimization problem and the Column Constraint Generation (CCG) method is adopted to accurately achieve the optimal solution. The proposed methodology is applied on an IEEE 33-bus system and a real 94-bus system. Its effectiveness is verified by numerical case studies.