Assessment of stabilization strategies for high-order discontinuous Galerkin methods
Please login to view abstract download link
High-order simulations of high-speed turbulent flows require stabilization strategies to obtain physically relevant results or avoid blow-up of the computation. A large variety of approaches is available, which should be chosen in regard of the flow regime, type of simulation, accuracy and stability constraints, and cost limitations. Promising strategies for the discontinuous Galerkin method (DG) include entropy stable methods (ES) [1], physics based artificial viscosity (AV) [2], entropy stable subcell blending [3], a novel subcell solution limiting strategy in development in our group, and hp-adaptation. The present contribution aims to describe and compare stabilization strategies for scale-resolving simulations of high-speed turbulent flows, using the DG solver ArgoDG of Cenaero. Focusing on perfect gas simulations, the comparisons are performed on classical test cases including shock-turbulence interaction from [4], the supersonic Taylor-Green vortex (TGV), the 3D shock-boundary layer interaction [5], and the Daru-Tenaud viscous shock tube [6] test cases. The efficiency and effectiveness of each strategy are assessed, providing guidance for their extension to more complex physical models.
