Stochastic Two-Stage Resilient Scheduling of Hydrogen System in Power Distribution Grids with High Penetration of Solar PV Systems: A Convex Worst Case Analysis

The renewable distribution network holds tremendous potential for economic advantages and sustainability improvements within the power industry. However, resilient operations in post-disaster scenarios present significant challenges for the system. Operational risks arise from uncertain resources, making day-ahead scheduling complex for the distribution grid's operator. Therefore, real-time modeling is considered more suitable to effectively manage these uncertainties. The hydrogen system (HSYS) is also an emerging technology that improves the technical and operational characteristics of distribution networks, particularly in challenging and unpredictable situations. The study presents a comprehensive framework for evaluating HSYSs within a photovoltaic-focused power distribution network, focusing on day-ahead resiliency-oriented operations (RESOP) while considering risk factors associated with uncertain incidents. The effectiveness and feasibility of implementing the proposed HSYS in day-ahead risk-based resilient scheduling of the distribution network are analyzed using a specific distribution system. The model is formulated as a MILP problem, and the impacting uncertainties are modeled using stochastic programming. The primary risk associated with uncertainties is captured using the conditional value-at-risk (CVaR) approach, which involves the possible worst realizations of uncertainties in the model. The study's findings demonstrate that the integration of distributed HSYSs leads to significant cost savings of up to 16 % under normal conditions and 1.25 % when considering risk modeling.