Computational fluid dynamics-based multiobjective optimization for catalyst design

For an industrial secondary methane steam reformer with regular packing, catalyst design is accomplished by an integrated optimization approach, which includes the design of experiment, computational fluid dynamics (CFD) simulation, a response surface method, and a genetic algorithm, for multiobjective optimization. Both spherical and cylindrical catalysts are studied. The reactor performance considered for the catalyst design includes the pressure drop and hydrogen production, which constitute the binary objective functions for optimization. The optimal solutions reveal that a large pore diameter, near 1 μm, should be adopted for spherical catalysts. For cylindrical catalysts, the optimal design suggests the use of a 1-big-hole shape with a larger particle and pore size, 10-13 mm and near 1 μm, or a 4-hole shape with a smaller particle size of 6-8 mm.