## Abstract A method for modelling of flow discontinuities based on deconvolution with a relaxation regularization (DDM) is compared with an essentially nonβoscillatory scheme (ENO) without further modelling and with a dynamic Smagorinsky subgridβscale model (DSM). For the DDM approach, a sufficien
A Subgrid-Scale Deconvolution Approach for Shock Capturing
β Scribed by N.A. Adams; S. Stolz
- Publisher
- Elsevier Science
- Year
- 2002
- Tongue
- English
- Weight
- 319 KB
- Volume
- 178
- Category
- Article
- ISSN
- 0021-9991
No coin nor oath required. For personal study only.
β¦ Synopsis
We develop a method for the modeling of flow discontinuities which can arise as weak solutions of inviscid conservation laws. Due to its similarity with recently proposed approximate deconvolution models for large-eddy simulation, the method potentially allows for a unified treatment of flow discontinuities and turbulent subgrid scales. A filtering approach is employed since for the filtered evolution equations the solution is smooth and can be solved for by standard central finite-difference schemes without special consideration of discontinuities. A sufficiently accurate representation of the filtered nonlinear combination of discontinuous solution components which arise from the convection term can be obtained by a regularized deconvolution applied to the filtered solution. For stable integration the evolution equations are supplemented by a relaxation regularization based on a secondary filter operation and a relaxation parameter. An estimate for the relaxation parameter is provided. The method is related to the spectral vanishing-viscosity method and the regularized Chapman-Enskog expansion method for conservation laws. We detail the approach and demonstrate its efficiency with the inviscid and viscous Burgers equations, the isothermal shock problem, and the one-dimensional Euler equations.
π SIMILAR VOLUMES
Several issues related to applications of the dynamic subgrid-scale (SGS) model in large-eddy simulation (LES) at low Reynolds number are investigated. A modified formulation of the dynamic model is constructed and its perfoxmance in low-Reynolds-number LES of channel flow is assessed through a comp
A finite difference scheme is presented for the solution of the two-dimensional shallow water equations in steady, supercritlcal flow. The scheme incorporates numerical characteristic decomposition, is shock capturing by design and incorporates space-marching as a result of the assumption that the f