Dynamic scheduling of renewable-powered multi-electrolyzer systems for hydrogen production with simulated annealing optimization: a load balancing and efficiency enhancement approach

Achieving stable and efficient hydrogen production via large-scale alkaline electrolysis under fluctuating renewable energy remains a major challenge, primarily due to overlooked electrolyzer startup thresholds and runtime imbalance among stacks in conventional scheduling strategies. These issues often lead to energy curtailment, equipment degradation, and operational risks. The authors’ previous work combined segmented power allocation with periodic cycling, improving performance but lacking real-time adaptability and optimal coordination between hydrogen production and load balancing. Building on this, this study proposes a dynamic scheduling strategy (CS4) that integrates a segmented dispatch algorithm—enhancing energy capture by aligning power input with electrolyzer startup thresholds—with a simulated annealing (SA)-based optimization approach. A binary-variable, state-aware model represents electrolyzer transitions among production, standby, and shutdown states, enabling responsive control under variable wind input. CS4 establishes a closed-loop framework to balance load, extend equipment life, and improve efficiency. Simulations over 250 days using real wind data show that CS4 increases hydrogen output by 5.36 %, reduces runtime deviation by 99.82 %, and improves lifecycle total profit by 10.48 % compared to baseline strategies. Its verification of the robustness and scalability across system sizes confirms CS4 as a robust, adaptive, and lifecycle-conscious solution for intelligent hydrogen production under intermittent renewables.