Multiobjective optimization of staggered elliptical pin-fin arrays

This work presents a numerical procedure for multiple objectives to optimize staggered elliptic-shaped short pin-fin arrays. The multiobjective problem is to achieve an acceptable compromise between augmentation of turbulent heat transfer and reduction in friction loss. Four nondimensional variables, pin-fin height-to-channel height ratio, major axis length-to-channel height ratio, minor-axis length-to-channel height ratio, and pin-fin pitch-to-channel height ratio are chosen as design variables. The D-optimal method is used to determine the training points. The response surface method is used to approximate the Pareto optimal front with Reynolds-averaged Navier-Stokes analysis of the flow and heat transfer using the shear stress transport (SST) turbulence model. The Pareto-optimal solutions are obtained using a combination of an evolutionary algorithm and a local search.