Parameter study of high-temperature proton exchange membrane fuel cell using data-driven models

Abstract In this paper, a numerical model of high-temperature proton exchange membrane fuel cell (HT-PEMFC) was developed, in which the thermal and electrical properties were treated as temperature dependent. Based on the numerical simulation, the needed training data was acquired and used for the development of data-driven model via the artificial neural network (ANN) algorithm. The developed data-driven model was then used to predict the performance of HT-PEMFC. The simulation results indicated that the deviation of ANN prediction was less than 2.48% compared with numerical simulation. The effects of various influential factors, such as the geometry size of the gas flow channel, the thickness of the membrane and the operating temperature, could be predicted easily by using the ANN model. The ANN model prediction results showed that the more compact fuel cell and the higher operating temperature improved the performance of HT-PEMFC. The proposed ANN model and the parameters study will contribute to the further design and operation of HT-PEMFC.