Simulation of a Solid Oxide Fuel Cell: Heat-up, Start-up Process and the Thermal Stress Analysis

• Main Authors: Ming-HongChen, 陳銘宏
• Other Authors: Tsung-Leo Jiang
• Format: Others
• Language: en_US
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/90930793779363272961

博士 === 國立成功大學 === 航空太空工程學系碩博士班 === 98 === The SOFC (Solid Oxide Fuel Cell) meets the duo requirements of high energy-conversion efficiency as well as ultra-low pollution, and has been treated as one of the most important energy converting devices. The present study focuses on three critical issues for the development of the planar SOFC. The first one is to evaluate the performance of different heat-up modes and to identify the optimal one. The second one is to estimate the effect of the fuel type, the anode-recycling mechanism, and the fixed-temperature-difference mechanism on the start-up time and the temperature gradient. The last one is to investigate the performance of the bonded compliant seal (BCS) technique for the planar SOFC under the practical operating condition. For the first part, it was found that from the aspect of the heat-up time, the single-channel mode leads to about 2.7 folds longer than that of the dual channel mode, making it impractical to be employed in the SOFC heat-up process. The required time for the counter-flow configuration is about 25% less than that of the co-flow configuration. From the aspect of the maximum-temperature-gradient, the slowest single-channel mode leads to the smallest temperature gradient. For the counter-flow configuration, its temperature gradient is averagely about 17% larger than that of the co-flow configuration. For both the co-flow and the counter-flow configurations, the maximum-temperature-gradient occurs at the beginning of the heat-up process, while it occurs at the end of the heat-up process for the single-channel mode. From the aspect of the required energy, it is approximately constant for the counter-flow configuration as the burner power larger than 10kW, while it gradually increases for the co-flow configuration. The total energy required for the counter-flow configuration is about 20% less than that of the co-flow configuration. The optimal heat-up approach is identified as the counter-flow configuration. Under different burner powers, the optimal counter-flow configuration results in the smallest required energy and the shortest time for the heat-up process of the planar, anode-supported SOFC at the expense of a higher maximum temperature-gradient. For the second part, it was found that the effective maximum temperature-gradient generates at the beginning of the start-up process for both hydrogen and methane. The required time for the case using methane is 3.2 folds longer than that using hydrogen. For the APU system, a 40 minutes start-up process is too slow. For the effective maximum-temperature-gradient, the case using hydrogen is 2.2 folds larger than that utilizing methane. The statement about the endothermic internal reforming reaction generating large temperature gradient is only valid in the steady-state, while it is a positive effect on the accommodation of the temperature uniformity during the start-up process. There is no significant effect of the anode-recycling mechanism on the effective maximum temperature-gradient. However, the effect of the anode-recycling on the start-up time is significant. Comparing to the base case, i.e., without anode-recycling, the start-up time decreases by 48.58% under the condition of 70% anode-recycling. For the fixed-temperature-difference mechanism in the start-up process, a properly selected temperature difference leads to a smaller effective maximum temperature-gradient at the beginning of the start-up process and a shorter start-up time by accelerating the start-up pace at the later stage. Therefore, if the maximum sustainable effective maximum-temperature-gradient for a specific SOFC is known, the start-up process can be accelerated by choosing as high as possible fixed temperature difference, while keeping the effective maximum-temperature-gradient under the allowable threshold. For the last part, it was found that the predictions with the assumption of a uniform temperature might underestimate the thermal-stress of the present investigated SOFC with the BCS design by 28% for the cell and 37% for the metal frame in comparison with those in practical operating conditions where the temperature is non-uniform. The dominant factor for the thermal-stress is location dependent. The contribution of the temperature gradient to the thermal-stress of the cell for an SOFC using the BCS design is comparable with that using the bonded glass–ceramic seal. With a lower voltage, the thermal-stress of the cell is relatively lower, while the contribution of the temperature gradient to the thermal-stress is higher.