Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
Jing Sun, Pengzhu Lin, Lin Zeng, Zixiao Guo, Yuting Jiang, Cailin Xiao, Qinping Jian, Jiayou Ren, Lyuming Pan, Xiaosa Xu, Zheng Li, Lei Wei, Tianshou Zhao
Abstract
Rational design of gas diffusion layers (GDL) is an example of a long-standing pursuit to increase the power density and reduce the cost of proton exchange membrane fuel cells (PEMFC). However, current state-of-the-art GDLs are designed by trial-and-error, which is a time-consuming endeavor. Here, we propose a closed-loop workflow of Bayesian machine learning approach to guide the design of GDL structures. With artificial neural network accelerating the calculation of anisotropic transport properties of reconstructed GDLs, Bayesian optimization algorithm identifies optimal structures in only 40 steps, maximizing the PEMFC’s limiting current density. Results suggest that the optimal porous GDL structure consists of highly orientated fibers with moderate diameters, which is successfully fabricated with a controlled electrospinning technique. The PEMFC demonstrates a high power density of 2.17 W cm-2 and a limiting current density of ~7200 mA cm-2, far exceeding that with commercial GDL (1.33 W cm-2 and ~2700 mA cm-2). Rational design of gas diffusion layers (GDL) is critical for enhancing the performance of proton exchange membrane fuel cells. Here, the authors employ Bayesian machine learning to design aligned GDLs, achieving a limiting current density of 7200 mA cm-2.