Model-based optimization of energy efficiency in alkaline water electrolysis under current fluctuations

Alkaline water electrolysis (AWE) is regarded as a promising pathway for large-scale green hydrogen production, yet its performance is constrained by the intermittency of renewable energy and the insufficient optimization of current-temperature-flow interactions. To address this issue, an electro-thermal empirical model was established using 78 h of performance data, with parameters identified via gradient descent, and subsequently applied to simulation-based analysis and operating condition optimization. Under fluctuating current conditions, coordinated regulation of lye temperature and flow rate was enabled within the proposed framework, allowing system dynamics to be captured accurately while ensuring practical feasibility. Validation experiments conducted under various fluctuation scenarios revealed that optimized control reduced outlet temperature fluctuations by 81.8 % and lowered specific energy consumption by 1.82 %, with more pronounced effects observed at medium and low loads and under random fluctuations, thereby demonstrating strong adaptability. Close agreement between experimental results and model predictions was achieved, confirming the reliability of the approach and highlighting the potential of data-driven coupling optimization to enhance the efficiency and stability of AWE systems, providing guidance for large-scale renewable‑hydrogen integration.