Multi-level network hardening for resilient distribution systems under extreme rainfall considering decision-dependent uncertainty

With the sharp growth of extreme rainfall events, it is imperative to enhance the resilience of distribution systems under extreme rainfall scenarios. To address the limitations of conventional hardening models, this paper proposes a multi-level network hardening strategy for distribution systems to reduce load loss and total operational cost while considering flood prevention measures and decision-dependent uncertainty. Firstly, through a mechanism analysis of extreme rainfall impacts on distribution systems, and by combining the digital elevation model and the Chicago hyetograph method, an urban waterlogging model based on the Integrated Terrestrial Fluxes Flood model is established, incorporating surface infiltration, storm-drain drainage, and land-use types. Secondly, a time-varying decision-dependent failure probability model for system equipment is developed based on water depth calculation results from the urban waterlogging model. Finally, a multi-level distribution network hardening model integrating flood prevention measures considering decision-dependent uncertainty is constructed, which is formulated as a mixed-integer second-order cone programming problem. Case studies on the IEEE 33-node distribution system demonstrate that, under a 100-year return-period rainstorm scenario, the proposed model can effectively reduce load loss while significantly enhancing the resilience of distribution systems against extreme rainfall events.