Physical vulnerability-oriented reconfiguration of microgrids: centralized vs. decentralized control

The increasing penetration of distributed energy resources and networked microgrids (MGs) has amplified the vulnerability of cyber-physical energy systems to operational disturbances and control-dependent interactions. While microgrid reconfiguration (MR) is widely used to enhance system resilience, existing approaches rarely quantify how different control architectures influence physical vulnerability under identical network conditions. This paper proposes a unified cyber-physical vulnerability assessment framework for MR based on a composite integrated system vulnerability (ISV) index. The ISV combines dynamic instability, operational limit violations, and frequency deviation into a single physically interpretable metric. MR is formulated as an optimization problem that minimizes ISV, subject to AC power flow feasibility, radiality, operational constraints, and a bounded number of switching actions. The framework is applied to a benchmark multi-microgrid system operating under both centralized and decentralized control architectures. For each candidate topology, tie switch positions are evaluated by recomputing ISV under the corresponding control mode, enabling a fair and architecture-aware comparison. Simulation results demonstrate that identical reconfiguration actions can yield markedly different vulnerability levels depending on the control structure. Additional stress scenarios with increased loading further reveal how vulnerability escalates as operational margins shrink. The results highlight the critical role of control-topology interaction in MG vulnerability and provide a systematic methodology for control-aware reconfiguration of cyber-physical energy systems.