Dynamic hierarchical modeling and control strategy of high temperature proton exchange electrolyzer cell system

High temperature proton exchange membrane electrolyzer cells (HT-PEMECs) show faster reaction kinetics than the low temperature PEMECs (LT-PEMECs) and are suitable for utilizing waste heat from the industry. However, dynamic modeling and control of HT-PEMECs are still lacking, which is critical for integrating the HT-PEMECs with fluctuating renewable power. In this study, hierarchical models are developed to investigate the transient behavior of the HT-PEMEC system with hydrogen recirculation. It is observed that the maximum efficiency point of the reference power can be reached by cooperatively adjusting the current density and anode inlet gas flow rate, and the application of artificial neural networks can accurately predict the operating conditions at the points of maximum efficiency. Moreover, the proposed cooperative model predictive control strategy not only improves the efficiency (about 1.2%) during dynamic processes but also avoids the problem of reactant starvation. This study provides useful information to understand the dynamic behaviors of HT-PEMECs driven by excess renewable power.