Aerodynamic and heat transfer design optimization of internally cooling turbine blade based different surrogate models

摘要

This study presents a numerical procedure to optimize the cooling passage structure of turbine blade to enhance aerodynamic and heat transfer. Surrogate model based optimization technique is used with Navier-Stokes analysis of fluid flow and heat transfer with RNG k-epsilon transport turbulence model. The objective function is defined as a nonlinear combination of heat transfer and pressure loss with K-S function. Optimal Latin Hypercube Sampling is used to determine the training points as a mean of design of experiment. Two Loops Dynamic Optimization System (TLDOS) is performed to implement the cooling blade optimization. Blade performance improves obviously, especially the kriging model based system. Result shows a significant impact of rib positions for blade heat transfer but slightly for total pressure loss. Numerical simulation proves the feasibility and validity of the TLDOS methods.