Numerical investigation of a single jet impingement on a flat surface using a cubic k-ε non-linear eddy viscosity model, to predict the effect of cooling on gas turbine blades

Abstract

The ability to accurately predict the effect of cooling on gas turbine blades is essential in designing the blades that will operate at extremely high temperature. The standard k-ε linear eddy viscosity model is known to be inaccurate in predicting highly complex flows. Thus, a relatively new cubic k-ε non-linear eddy viscosity model was tested to ascertain whether it has improved the performance of eddy viscosity models. A single jet impingement on a flat plate with surface-to-nozzle distance of H/D = 6 was investigated numerically using a cubic k-ε non-linear eddy viscosity model of Craft et. al. [1] and high-Re k-ε linear eddy viscosity model of Jones & Launder [2]. Both use standard wall-function to model the near-wall flow. Dynamic field profiles taken at certain distances away from the impingement point were compared with experimental results of Cooper et al. [3]. The heat transfer field results were compared with the experimental data of Baughn et al. [4]. The dynamic field results show that the cubic non-linear model gives a much better prediction than the linear model. The heat transfer results showed that the linear model over-predicted the heat transfer rate at the stagnation point whilst the non-linear model gave under-prediction due to a lower prediction of the turbulent kinetic energy at that region.