.png){kind=icon}
.png){kind=icon}
Return
Fuel ejector design and optimization for solid oxide fuel cells using response surface methodology and multi-objective genetic algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.applthermaleng.2023.121067.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Ejectors are the main devices used to circulate high-temperature anode exhaust gases for solid oxide fuel cells (SOFCs) for recovering and utilizing water vapor and heat. To optimize the performance of a high-temperature fuel ejector for SOFCs, we established a numerical model of the ejector and then validated it using the experimentally obtained primary flow and secondary flow mass flow rates. The influence of the primary nozzle exit position (NXP), length (Lm) and radius (Rm) of the mixing chamber, and other key structures, on ejector performance was examined via Response Surface Methodology (RSM). The performance of the ejector gradually decreased with increasing NXP and Lm and first increased and then decreased with increases in Rm. The sensitivity of the ejector performance to Rm, Lm, and NXP decreased sequentially, and its structure was optimized using a multi-objective genetic algorithm. The entrainment ratio of the optimized ejector reached 6.66, representing an increase of 23.37 % compared with the original ejector. This work proposes a design and optimization method for optimizing the ejector to meet the high entrainment ratio required by SOFCs. The response surface methodology was used to realize the research on the influence of multi-structural parameter coupling on the performance of the ejector, which significantly improved the optimization efficiency and accuracy. This study also introduced multi-objective genetic algorithm on the basis of response surface methodology to optimize the performance of the ejector. Through continuous gene mutation and iteration, the ejector has the best performance. This provides a new direction for future SOFC fuel ejector performance optimization, that is, introducing various optimization algorithms on the basis of multi-parameter coupling effects, and finally solving an optimal performance through continuous permutation, combination and iterative calculation.
Keywords:
Ejector
Solid oxide fuel cell
Anodic recirculation
Response surface methodology
Genetic algorithm
Journal
IF:
6.9
Papers:
2.7W
Citations:
10.6W