Numerical Simulation of Turbine Blade Heat Transfer Using Two-Equation Turbulence Models

• Main Author: Elfaghi, Abdul Hafid M.
• Format: Thesis
• Language: English; English
• Published: 2000
• Subjects: Turbines - Lubrication; Turbulence
• Online Access: http://psasir.upm.edu.my/id/eprint/10493/1/FK_2000_27.pdf; http://psasir.upm.edu.my/id/eprint/10493/
• Institution: Universiti Putra Malaysia

The development of high performance gas turbines requires high turbine inlet temperatures that can lead to severe thermal stresses in the turbine blades, particularly in the first stages of the turbine. Therefore, the major objective of gasturbine designers is to determine the thermal and aero-dynamical characteristics of the turbulent flow in the turbine cascade. This work is a numerical simulation of fluid flow and heat transfer in the turbine blade using different two-equation turbulence models. The turbulence models used here were based on the eddy viscosity concept, which determined the turbulent viscosity through time-averaged Navier-Stokes differential equations. The most widely accepted turbulence models are the two-equation models, which involves the solution of two transport equations for the turbulent kinetic energy, k, and its rate of dissipation, e or w. In the present simulation, four two-equation turbulence models were used, the standard k-e model, the modified Chen-Kim k-e model, RNG model and Wilcox standard k-w turbulence model. A comparison between the turbulence models and their predictions of the heat flux on the blade were carried out. The results were also compared with the available experimental results obtained from a research carried out by Arts et al.(1990) at the von Karman Institute of Fluid Dynamics (VKI). The simulation was perfonned using the general-purpose computational fluid dynamics code, PHOENICS, which solved the governing fluid flow and heat transfer equations. An H-type, body-fitted-co-ordinate (BFC) grid was used and upstream and downstream periodic conditions were specified. The grid system used was sufficiently fine and the results were grid independent. All models demonstrated good heat transfer predictions for the pressure side except close to the leading edge. On the suction side, standard model over-predicted the heat transfer, whereas Chen-Kim, RNG and k-w models captured the overall behaviour quite well. Unlike k-w model, all k-e models generated very high turbulence levels in the stagnation point regions, which gave rise to the heat transfer rates close to the leading edge.