Refined PEMEL operation of hydrogen-integrated islanded microgrid via penalized sensitivity region-based robust optimization

Distributed renewable generators are essential for islanded microgrids (IMGs), yet their intermittency and uncertainty may increase operating costs and induce frequency and voltage issues. Proton exchange membrane electrolyzers (PEMELs) as a controllable demand-side resource can consume surplus renewables and produce green hydrogen. However, existing PEMEL models rely on the assumption of fixed conversion rate or lack physics-grounded characterization of device-level behaviors in power grid operation. Meanwhile, conventional IMG operation focuses on keeping operating constraints and ignores robustness on objectives against uncertainties, while hydrogen demand is not treated as an explicit uncertainty source. To address these gaps, this paper proposes a refined PEMEL operation method for a hydrogen-integrated IMG via new penalized sensitivity region based robust optimization (PSRRO). First, a power-responsive PEMEL model is developed to capture nonlinear device-level characteristics driven by current density. Second, a PSRRO method is proposed to optimize IMG operation under multi-dimensional uncertainties of renewable generation, electrical load, and hydrogen demand. Lastly, an efficient solution algorithm is developed to reduce computational burdens and enhance the applicability of the PSRRO method. Numerical results verify that the proposed method can achieve robust performance by keeping both narrow objective variations and operating constraints against multi-dimensional uncertainties, with high computing efficiency for practical use.