Numerical flux functions for Reynolds-averaged Navier-Stokes andkωturbulence model computations with a line-preconditionedp-multigrid discontinuous Galerkin solver

2012 ◽  
Vol 71 (8) ◽  
pp. 1055-1072 ◽  
Author(s):  
Marcel Wallraff ◽  
Tobias Leicht ◽  
Markus Lange-Hegermann
Author(s):  
Fan Feng ◽  
Chunwei Gu ◽  
Xuesong Li

In this paper Discontinuous Galerkin Method (DGM) is applied to solve the Reynolds-averaged Navier-Stokes equations and S-A turbulence model equation in curvilinear coordinate system. Different schemes, including Lax-Friedrichs (LF) flux, Harten, Lax and van Leer (HLL) flux and Roe flux are adopted as numerical flux of inviscid terms at the element interface. The gradients of conservative variables in viscous terms are constructed by mixed formulation, which solves the gradients as auxiliary unknowns to the same order of accuracy as conservative variables. The methodology is validated by simulations of double Mach reflection problem and three-dimensional turbulent flowfield within compressor cascade NACA64. The numerical results agree well with the experimental data.


Sign in / Sign up

Export Citation Format

Share Document