Transient analysis of a solid oxide fuel cell unit with reforming and water-shift reaction and the building of neural network model for rapid prediction in electrical and thermal performance

Abstract

This study investigates the effect of reforming reaction, water-shift reaction, and operating parameters on the transient performance of a solid oxide fuel cell unit, because the transient analysis is necessary and helpful for the applications of a SOFC with cross-flow configuration. The primary results show that all properties approach the steady state at similar time except the cell temperature. The reforming and water-shift reaction obviously promote the average current density by 5%, and lower the maximum cell temperature by 20 K. The molar flow rate variation deduces about 15 K difference of maximum cell temperature. The effect of inlet temperature and operating voltage on the average current density and maximum cell temperature is more obvious than the molar flow rate effect. Moreover, this study builds a neural network model to predict the steady average current density and maximum cell temperature rapidly and correctly, which is helpful for the control of a SOFC.