Topology-Aware Propagation-Based Assessment of Extreme-Weather Impacts on Distribution System Resilience

Junjie Yin, Xinyu Feng, Jian Han, Fangxing Li

Extreme weather events and the increasing integration of distributed energy resources (DERs) introduce growing uncertainty and resilience challenges for distribution systems. Unlike conventional deterministic contingencies, weather-driven disruptions exhibit probabilistic and spatial-temporal characteristics, where outage consequences depend on both geographic exposure and feeder topology. Existing approaches commonly focus on deterministic outage analysis, while topology-aware operational impact assessment under forecast uncertainty remains limited. This paper proposes an event-conditioned uncertainty modeling and topology-aware impact propagation framework for distribution systems under torrential rain events. The proposed framework integrates probabilistic event-track modeling, branch-level fault screening, downstream impact propagation analysis, and operational impact assessment within a unified workflow. Finally, case studies on the IEEE 33-bus distribution feeder demonstrate that the proposed framework can distinguish geographic exposure from topology-dependent operational impacts and support progressive early-warning impact assessment under uncertain scenarios. Furthermore, the impact zones are visualized on the CURENT Large-scale Testbed (LTB)-AGVis platform.