Method of predicting liquid regions and vapor regions in bipolar plates of a fuel cell

A method of designing a fuel cell, the method comprising:

• by one or more computing devices having one or more processors:
• predicting a location of one or more liquid regions and one or more vapor regions in microchannels at an air layer of a plate of the fuel cell, wherein the prediction comprises executing an optimization method to identify, in the microchannels of the air layer, one or more regions where oxygen concentration is less than a predetermined threshold concentration;
• wherein predicting further comprises applying, iteratively, material properties of a liquid phase and a vapor phase to the identified one or more regions until convergence is achieved;
• passing the material properties of the liquid phase and the vapor phase to the optimization method after the convergence is achieved between the liquid phase and the vapor phase;
• optimizing, based on the prediction, the material properties of the liquid phase and the vapor phase, and via a homogenized flow optimization, fluid flow networks for the air layer, a hydrogen layer, and a coolant layer of the fuel cell;
• generating, based on the homogenized flow optimization, one or more multi-scale Turing-patterned microstructures for the air layer and the hydrogen layer, the generating the one or more multi-scale Turing-patterned microstructures comprising propagating anisotropic diffusion coefficient tensors for reaction-diffusion equations through time;
• and
• creating a fuel cell to convert chemical potential energy into electrical energy by, at least in part, stacking the air layer and the hydrogen layer in parallel to each define half channels for the coolant layer.

A method of designing a fuel cell, the method comprising:

• by one or more computing devices having one or more processors:
• predicting a location of one or more liquid regions and one or more vapor regions in microchannels of an air layer of a plate of the fuel cell, and optimizing, via homogenized flow optimization, fluid flow networks for the air layer based on the prediction, wherein the prediction comprises executing an optimization method to identify, in the microchannels of the air layer, one or more regions where oxygen concentration is less than a predetermined threshold concentration;
• wherein predicting further comprises applying, iteratively, material properties of a liquid phase and a vapor phase to the identified one or more regions until convergence is achieved;
• passing the material properties of the liquid phase and the vapor phase to the optimization method after the convergence is achieved between the liquid phase and the vapor phase;
• optimizing, based on the prediction, the material properties of the liquid phase and the vapor phase, and via a homogenized flow optimization, fluid flow networks for the air layer, a hydrogen layer, and a coolant layer of the fuel cell;
• generating, in response to the optimizing, one or more multi-scale Turing-patterned microstructures for the air layer, the generating the one or more multi-scale Turing-patterned microstructures comprising propagating anisotropic diffusion coefficient tensors for reaction-diffusion equations through time, wherein the one or more multi-scale Turing-patterned microstructures include multi-scale channels such that a larger flow structure interfaces with a smaller flow structure;
• and
• creating a fuel cell to convert chemical potential energy into electrical energy by, at least in part, stacking the air layer and a hydrogen layer in parallel to define a coolant layer.