Applications of Artificial Intelligence for Analysis of Two-Phase Flow in PEM Fuel Cell Flow Fields

This work probes the use of machine learning to uncover complex relationships between liquid phase behavior and pressure drop in polymer-electrolyte fuel cell reactant channels. Using a supervised decision tree (DT) regression algorithm, the liquid distributions in an ex-situ experimental reactant channel are correlated to the two-phase pressure drop for a range of air flow rates. Liquid phase distribution data collection is accomplished by capturing 2D images via a CCD camera of a transparent flow channel simulating polymer-electrolyte fuel cell conditions. Pressure transducers measure pressure drop along the length of the channel and are synchronized with the image collection. The images are processed for noise reduction and feature extraction and then used with pressure data for training. Overall, the DTs predict pressure drop using liquid distributions as inputs at rates exceeding 88.9% accuracy and were able to capture complex behavior such as pressure changes due to breakup of liquid slugs.