Advanced design of cathode array protrusion structure of solid oxide fuel cell based on NSGA-II multi-objective optimization

• Published in: International journal of heat and mass transfer 2024-07, Vol.226, p.125457, Article 125457
• Main Authors: Cui, Yi, Wang, Zhen, Yang, Laishun, Jia, Huiming, Ren, Yunxiu, Song, Lei, Yue, Guangxi
• Format: Article
• Language: English
• Subjects: Multi-Objective Optimization Design; Response Surface Methodology; Solid Oxide Fuel Cells; Structured Array Design
• ISSN: 0017-9310; 1879-2189
• DOI: 10.1016/j.ijheatmasstransfer.2024.125457

•Biomimetic protrusions were utilized to enhance SOFC mass transfer.•Vortices generated by the protrusions facilitated reactant diffusion.•Power density of Pareto solutions saw an increase of 16.5% to 23.6%.•Pressure drop of Pareto solutions escalated marginally by 1.5% to 8.9%. As fuel cells gain widespread utilization, augmenting the power density and combination property of Solid Oxide Fuel Cells (SOFCs) has emerged as a pivotal area of research. This study introduces an innovative arrayed protrusion structure within the cathode flow channels of SOFCs to induce secondary flow. The influence of this arrayed structure on fuel cell combination property is meticulously analyzed through comprehensive numerical simulations. Furthermore, this paper employs the Response Surface Methodology to evaluate the synergistic effects of cathode operational conditions, particularly temperature, and arrayed structure parameters (including base width, height, and quantity) on SOFC performance. Correlations among these variables have been established to show synergistic associations between variables and performance index. The findings reveal that the inclusion of an arrayed structure in the cathode flow channels generates multi-directional vortices, thereby intensifying gas disturbance, facilitating reactant diffusion, and optimizing oxygen concentration distribution. This results in a marked enhancement of the fuel cell's overall performance. Among the operational parameters analyzed, temperature exerts the most substantial influence on power density, followed by protrusion height, base width, and number. Ultimately, using the Non-dominated Sorting Genetic Algorithm-II (NSGA-II), Pareto optimal solutions were derived, yielding a superior balance of higher power density and reduced pressure drop. Compared to the baseline, the power density in these Pareto optimal solutions saw an increase of 16.5% to 23.6%, while pressure drop escalated marginally by 1.5% to 8.9%.