摘要
This paper presents a data-driven distributionally robust framework for post-disaster networked microgrid (NMG) formation that accounts for renewable generation uncertainty. The NMG formation model jointly determines time-varying MG topologies, the placement and dispatch of mobile distributed generators (DGs) and energy storage (ES), and the scheduling of dispatchable loads to maximize weighted critical load restoration. Line losses from a linearized DistFlow model are included in the objective function, and a resilience index is used to ensure sufficient DG reserve for assigned critical loads. Uncertainty in RES outputs is handled by distributionally robust chance constraints (DRCC) defined on a Wasserstein ambiguity set, and the DRCC is reformulated through a CVaR-based tractable approximation. The efficiency of the proposed model is verified by numerical simulations on a modified IEEE 33-node system.