In order to reach higher power output and increase the thermodynamic efficiency of gas turbines, the turbine entry temperatures have to be further increased. These higher entry temperatures cause severe demands on the materials. For example, the lifetime of the complex and expensive turbine blades and vanes is highly dependent on the gas temperature. In modern gas turbines, a lot of effort is made in order to cool the turbine.
A full understanding of the various physical phenomena is vital to improve the blade cooling methods. The cooling of the blades can be divided into two main fields: the inner and the outer cooling, both governed by different flow regiemes.
At the inner blade surface a massively turbulent flow with numerous flow separations is generated to improve the heat transfer from the cooling air into the solid material. In order to understand thesethermal boundary layers anisotropic turbulence models are currently under development. However, at the outer blade surface, a very low heat transfer from the hot gas to the solid blade is of interest. Here dominant aspects are shock/boundary layer and cooling-air/flow-field interactions in flow fields with strong pressure gradients. Additional to the fluid flow modeling, the heat conduction in the solid materials must be modeled. This is because the temperature distribution of the blade surface influences the fluid boundary layer and vice versa. A coupling of the FEM-solver for heat conduction and the URANS-solver for the flow is a possible approach.
To simulate the flow the URANS-Solver TRACE is used. In TRACE the basic assumption for modeling the heat transfer is the Reynolds Analogy, which relates the analogue development of the heat and the momentum boundary layer. This must not necessarily be valid in all types of turbomachinery flows. Therefore more sophisticated models describing the thermal boundary layer are another field of research.