Intelligent optimization: Novel application of PCC, MCDM, and ANN + NSGA-III in integrated optimization of the flow field and porous layer structures for unitized regenerative fuel cell
Ke Chen, Wenshang Chen, Guofu Zou, Ben Chen
Abstract
Unitized Regenerative Fuel Cells (URFCs) are a promising technology that utilizes renewable energy sources to efficiently convert them into electricity while offering potential for renewable energy storage . These cells facilitate the conversion between electrical and chemical energy , enabling processes such as electrolysis and hydrogen gas synthesis from water. This study aims to propose a more efficient and stable mass transfer solution for URFCs through the integrated optimization of flow field structure and porous transport layer configuration . Leveraging Taguchi orthogonal design, Pearson correlation coefficient , contour analysis, multi-criteria decision making, and the integration of artificial neural networks with non-dominated sorting genetic algorithm-III, an optimized selection is performed. Results indicate that the optimized structure exhibits improved performance in both electrolytic cell (EC) and fuel cell (FC) modes. At a current density of 1.5 A/cm 2 , compared to traditional structures, the voltage decreases by 8.2 mV in the EC mode. In the FC mode at a current density of 1.0 A/cm 2 , performance improves by 5.748%, and URFC round-trip efficiency increases by 6.183%. The assessment of mass transfer capability reveals that the optimized structure promotes gas transfer processes in different modes, leading to significant overall performance enhancement of URFC. These findings provide valuable guidance for the enhanced performance of URFC.