Large Eddy Simulation Based Optimization of a Fan-Shaped Cooling Hole Geometry to Enhance Cooling Performance

Abstract

In this study, a shaped hole optimization approach based on Large Eddy Simulations and Efficient Global Optimization (EGO) is presented. The shape of a fan-shaped hole used for turbine film cooling is then optimized to maximize the film cooling performance and the numerical problem is modeled using flow configurations close to those in real gas turbine conditions. Four of the most important geometrical parameters defining a cooling hole shape, namely the depth of the expanded section, the hole inclination angle and the forward and the lateral expansion angles are selected as the design variables to obtain the optimal hole shape. Forty design cases at start are selected via an Optimal Latin Hypercube Sampling method (OLHS) and further more are added during the successive iteration steps of the optimization algorithm. The handling of these design cases including the CAD creation of the geometries, computational domains, meshes and finally the numerical setup is handled by the LES based autonomous tool which has been previously validated [1]. Finally, the Bayesian based EGO [2] method along with the Expected Improvement (EI) as the acquisition function is used to maximize the surface averaged film cooling effectiveness as the objective function. After several database enrichment steps to reduce the overall modal error of the response surface an optimal shape of the cooling hole with the highest cooling performance is obtained. The optimal geometry thus obtained has a significantly higher cooling performance than the reference hole shape which is also confirmed via the study of the fluid flow distribution in both the cases. Overall, this study shows that, Large Eddy Simulations can be successfully coupled along with an EGO based optimization approach to obtain the optimal shaped cooling hole in a computer-aided optimization setting.