质子交换膜电解池进水温度与流速参数分析及系统能效预测

PEM Electrolyzer Inlet Water Temperature and Flow Velocity Parameter Analysis and System Energy Efficiency Prediction

操作参数的变化对质子交换膜(PEM)电解池的系统能效和耐用性产生重大影响。基于多物理场电热耦合模型,对进水温度和进水流速进行了参数分析,研究了它们对电解池性能的影响。研究表明,提高进水温度能够显著改善电解池的电流密度,进而提高电解制氢效率。然而,这也使得电解池内部的温度可能升至超过膜的热降解临界温度(80℃)。另一方面,提高进水流速加强了电解池内部的热交换,使温度分布更加均匀,避免了电解池尾部局部热点的形成。但提升温度和流速也增加了电解系统中泵和散热器的功耗,对系统能效产生了不利影响。采用机器学习的方法,将多物理场仿真与人工神经网络(ANN)相结合,利用仿真模型的电热特性数据并加入系统能效计算,建立代理模型对其质子膜上最高温度与电解系统能效进行快速预测,以便电解池系统实际运行中可以及时调整进口参数。其中,模型预测的均方根误差仅有0.106%,确保了预测精度,并大大提高了仿真效率。因此,对不同进口参数下电解池的系统能效预测提供了重要的参考依据。

This study investigates the impact of varying operating parameters on the energy efficiency and durability of a proton exchange membrane(PEM)electrolyzer.Using a multi-physics electro-thermal model,we analyze the parametric effects of inlet water temperature and flow velocity on electrolysis performance.Results reveal that increasing inlet water temperature enhances electrolyzer current density,improving efficiency but risking membrane thermal degradation above 80℃.Conversely,higher inlet water flow velocity promotes uniform temperature distribution,yet increases pump and radiator power consumption,reducing system energy efficiency.We employ a machine learning approach combining multiphysics simulations with artificial neural networks(ANN)to predict system energy efficiency and maximum membrane temperature.The model achieves a root-mean-square error of only 0.106%,ensuring accurate predictions and enhancing simulation efficiency.This work serves as a valuable reference for predicting electrolytic cell system energy efficiency under varying inlet parameters.