ABSTRAKT

A three-dimensional model of a serpentine flow channel design of the bipolar plate of a proton exchange membrane fuel cell(PEMFC) with different landing to channel configurations (1:1, 2:2, 2:1, 1:2) has been developed using commercially available computational fluid dynamics (CFD) software to analyse the electrochemistry and transport phenomena of the reactants and products inside the fuel cell. The active areas considered for the numerical analysis were (10, 15, 20, 25, 30, 50 and 70 cm²). An artificial neural network (ANN) is used for the prediction of power density of different active areas and landing to channel ratios. An optimized configuration with a particular landing to channel ratio and active area can be obtained using artificial neural network from a limited number of numerical results, which will greatly reduce the effort and the computational time.