Turbulence Modelling and Parallel Solver Development Relevant for Investigation of Gas Turbine Cooling Processes

Abstract: To enhance the performance of gas turbine engines, high top-cycle temperatures are desired. However, the temperature is limited by current materials. To achieve the higher performance, and meanwhile maintain the metal temperatures of the combustors, vanes and blades below the allowable limits, sophisticated cooling techniques, such as impingement cooling, film cooling and convective internal cooling, are essential. This thesis focuses on developing numerical tools for improved predictions of cooling processes of gas turbine blades and combustor walls. The cooling methods include impinging jets, rib roughness and film jets. An in-house Navier-Stokes solver was extended to multi-block and parallel computations, with implementation of a range of turbulence models. The evaluated turbulence models for the Reynolds stresses are: 1. Two-equation models: Shear Stress Transport model (SST model), Explicit Algebraic Stress Model (EASM model), and the Linear Eddy Viscosity Model (LEVM model); 2. Four-equation models: V2F model in three versions; 3. Seven-equation models (Reynolds Stress Transport Models (RSTM)): Standard RSTM, Stress-w model, and the Speziale-Sarkar-Gatski model (SSG model). In addition, a low-Re RSTM (combined SSG and SST), is developed, which solves the problem at reattachment points. All the models mentioned above are based on the Reynolds averaged Navier-Stokes (RANS) equations. RANS methods, however, filter out many physical details of the underlying flow field. Consequently, the validity of the results is often questionable, especially for complex geometries, where the turbulence is very an-isotropic. Therefore, the large eddy simulation (LES) method was also considered, where large eddies are resolved and small scales are modelled. A number of case studies are presented basically for the following three purposes: Validation of the parallel multi-block code; Validation and evaluation of turbulence models; Providing results for better understanding and enhancement of the cooling process relevant for gas turbine systems.