Design optimization and thermal management of the PEMFC using artificial neural networks

Abstract

The subject of this study is to analyze the heat transfer inside the gas flow channel (GFC) of the proton exchange membrane fuel cell (PEMFC) numerically. Increasing the fluid-solid contact area, as well as, changing the cooling fluid velocity and temperature profiles can improve the heat transfer. Consequently, trapezoid porous ribs with three different geometrical parameters are placed in the GFC to meet this goal. These geometrical parameters are the base and the tip widths of the ribs in addition to their distance from each other. Numerical simulations arose the respective Nu numbers and friction factors of the system to evaluate the influence of these ribs. Afterward, these simulations data are used to train an artificial neural network (ANN) to model the system and produce data to perform the sensitivity analysis and sketch the corresponding three-dimensional diagrams. Finally, the optimum values of the mentioned geometrical parameters are calculated to maximize the heat transfer rate with minimum friction losses. Results indicate that although the lower distance between porous ribs has led to the higher Nu, this also causes the higher friction factor. Therefore, it is better to utilize higher distances to meet the higher performance evaluation criterion.