Journal of Power Sources, Vol.186, No.1, 10-21, 2009
Geometry optimization for proton-exchange membrane fuel cells with sequential quadratic programming method
Integration between COMSOL Multiphysics (TM) and MATLAB (TM) offers a useful option for the self-automated geometry optimization in proton-exchange membrane fuel cells (PEMFCS). It overcomes the difficulties of automatically re-generating high-quality computational meshes and subsequently running the simulations to evaluate the objective function values using commercial software in computational fuel cell dynamics-based designs. Geometry optimization studies of an air-breathing PEMFC searching for the optimum channel ratio at the anode and the optimum open ratio at the cathode, are undertaken. A sequential quadratic programming method is selected to deal with the constrained design problems, while the objective functions are evaluated by running the three-dimensional simulation script of COMSOL (TM) under the MATLAB (TM) environment. Simulation results show that for the air-breathing PEM fuel cell operated at 353 K and one standard atmosphere pressure, when the anode channel ratio is fixed at 10%. the optimum cathode open ratios are very similar for the cell operated at voltages of 0.7 and 0.4V, namely, 49.8% for 0.7 V and 49.5% for 0.4 V. When the cathode open ratio is set at 80% with a cell voltage of 0.7 V, the optimum anode channel ratio is found to be 34.7%. (c) 2008 Elsevier B.V. All rights reserved.
Keywords:Geometry optimization;Proton exchange membrane fuel cell;Sequential quadratic programming method