Generation and analysis of wind-photovoltaic power output scenarios for off-grid electrolysis systems

Green hydrogen production through water electrolysis is a key solution for achieving carbon neutrality. However, its intermittent and fluctuating nature poses significant challenges to the stable operation of electrolyzers. To enable efficient integration into off-grid hydrogen production systems, a deep understanding of electrolyzer performance under real-world operating conditions is required. This study proposes a systematic approach for the generation and validation of representative wind and photovoltaic power output scenarios. By integrating Principal component analysis and K-means clustering techniques, historical power data from multiple regional stations in China are processed to extract characteristic operating curves, accurately capturing key dynamic parameters such as ramp rates, fluctuation intensity, and low-load duration. Multi-scale experiments on proton exchange membrane (PEM) electrolyzers are conducted to validate the effectiveness of these characteristic curves. Experimental results demonstrate efficient power tracking and safe operation in tests involving both 100 W and 100 kW PEM electrolyzers. Under dynamic renewable energy input conditions, the hydrogen-in-oxygen concentration remained stable about 0.35 % and 0.6 % respectively, while the hydrogen production rate exhibited excellent tracking performance. Based on this, regional operational challenges are further quantified, revealing the impact of different renewable energy characteristics on system design. This framework provides valuable references for optimizing electrolyzer systems and evaluating their performance under various renewable energy conditions, offering critical foundations for the design optimization and control of off-grid hydrogen production systems.