Turbulence Modelling of Flows Related to Wall-Cooling Applications
Engineering flows are generally characterised by complex strain field and turbulence structure arising from a combination of shear, curvature, separation and recirculation etc. Under such flow conditions, transport by turbulent velocity fluctuations can make a significant contribution to the transport of momentum, heat and mass. This thesis is focussed on turbulence modelling based on the scalar eddy-viscosity concept, as well as on more elaborate second-order-moment closures, in which the modelled transport equations for the individual Reynolds stresses are solved for. The aim is to numerically simulate the behaviour of turbulent flows relevant to industrial wall-cooling applications, such as cooling of the blades in a gas turbine or of the walls in a combustion chamber. In order to determine the temperature distribution at the wall, it is necessary to accurately predict the flow in the near-wall region. In the vicinity of a solid wall, the viscous effects on the turbulence is of major importance and are usually taken into account in the simulation procedure by adopting a near-wall turbulence model. Increased computing power has made it feasible to adopt low-Reynolds number turbulence models in complex flows and these provide greater scope for representing the turbulent transport processes in the sublayer than standard wall functions. The present contribution considers different approaches to handle near-wall turbulence and presents examples of their relative performance in predicting convective heat transport in several flows of various complexity, relevant to wall-cooling applications.
low-Reynolds number models
turbulent heat transfer