Key points are not available for this paper at this time.
This paper introduces a comprehensive and scalable framework to enhance the resilience of power systems against flooding, with a particular emphasis on substation protection. Unlike traditional approaches that rely on independent probability distributions to model flooding, this study develops a stochastic flood scenario generator that incorporates spatial correlation, hurricane effects, and lognormal distributions, yielding more realistic and decision-relevant scenarios. A Flood Risk Metric (FRM) is proposed to quantify substation criticality based on flood hazard, population impact, and proximity to essential services, enabling the pre-selection of high-priority and low-impact substations to reduce the optimization search space. The framework accounts for practical deployment constraints, including limited installation time, team routing, and travel duration between substations, which renders conventional MILP solvers infeasible. To address this, metaheuristic algorithms—Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Grey Wolf Optimizer (GWO)—are employed. Additionally, k-means clustering is used to form superzones, further improving computational efficiency. A real-world case study in Charleston, SC, validates the proposed method, demonstrating significant cost reductions and effective protection strategies. Sensitivity analyses highlight the framework's robustness to changes in team availability, travel speed, and risk prioritization. This study bridges the gap between theoretical resilience modeling and real-world operational constraints, offering a practical solution for substation protection in flood-prone regions.
Liasi et al. (Thu,) studied this question.