抄録
Shape optimization of heat transfer augmentation device employed in turbine blade internal cooling passage has been performed numerically using single as well as multi-objective optimization procedures. Polynomial response surface approximation method and multi-objective genetic algorithm are used for single and multi-objective optimizations, respectively. Problem to enhance heat transfer rate considering staggered dimples on single surface of cooling passage has been formulated, and Reynolds-averaged Navier-Stokes equations are solved to analyze the flow field and the heat transfer. Three design variables defining channel and dimple dimensions, and two objective functions related to Nusselt number and friction drag are employed. Latin hypercube sampling is used to generate sampling points in design space, and the evaluated objectives are used to generate a set of optimal designs. Optimal shapes show higher heat transfer rates in the case of lower channel height, higher dimple depth and higher dimple spacing, but these geometries also produce higher pressure drops.
