Optimizing current loading strategies for PEMFC stacks and predictions of needed powers for successful cold starts

Enhancing the cold-start capability of proton exchange membrane fuel cell (PEMFC) stacks is essential for their reliable deployment in low-temperature environments. This study develops a one-dimensional multiphysics model to evaluate the cold-start behavior of PEMFC stacks and proposes a two-stage current ramp loading strategy . A data-driven model combined with the genetic algorithm is then used to determine the optimal parameters for the two-stage current ramp loading strategy. Results show that the proposed strategy achieves a 64 % reduction in start-up duration compared to conventional approaches, with the highest effective heat generation efficiency. The existence of endplates was found to significantly suppress the internal temperature rise by approximately 80 % in a 10-cell stack, contributing to cold-start failure and necessitating additional heating power. A fast prediction model , based on a neural network, was developed to estimate the minimum auxiliary heating power required for a successful cold start, with deviations within ±10 % of numerical results. The insights provided in this work offer practical guidance for optimizing PEMFC cold-start strategies, ensuring both improved performance and minimized auxiliary energy consumption.