Mass Flux Schemes and Connection to Shock Instability

We analyze numerical mass fluxes with an emphasis on their capability for accurately capturing shock and contact discontinuities. The study of mass flux is useful because it is the term common to all conservation equations and the numerical diffusivity introduced in it bears a direct consequence to the prediction of contact (stationary and moving) discontinuities, which are considered to be the limiting case of the boundary layer. We examine several prominent numerical flux schemes and analyze the structure of numerical diffusivity. This leads to a detailed investigation into the cause of certain catastrophic breakdowns by some numerical flux schemes. In particular, we identify the dissipative terms that are responsible for shock instabilities, such as the odd?even decoupling and the so-called “carbuncle phenomenon”. As a result, we propose a conjecture stating the connection of the pressure difference term to these multidimensional shock instabilities and hence a cure to those difficulties. The validity of this conjecture has been confirmed by examining a wide class of upwind schemes. The conjecture is useful to the flux function development, for it indicates whether the flux scheme under consideration will be afflicted with these kinds of failings. Thus, a class of shock-stable schemes can be identified. Interestingly, a shock-stable scheme's self-correcting capability is demonstrated with respect to carbuncle-contaminated profiles for flows at both low supersonic and high Mach numbers.

[1]  B. Leer,et al.  Flux-vector splitting for the Euler equations , 1997 .

[2]  P. Roe.  Approximate Riemann Solvers, Parameter Vectors, and Difference Schemes , 1997 .

[3]  Richard Sanders,et al.  Regular ArticleMultidimensional Dissipation for Upwind Schemes: Stability and Applications to Gas Dynamics☆ , 1998 .

[4]  Meng-Sing Liou,et al.  Hybrid Upwind Splitting (HUS) by a Field-by-Field Decomposition , 1995 .

[5]  Richard Sanders,et al.  Multidimensional Dissipation for Upwind Schemes , 1998 .

[6]  A. Jameson.  ANALYSIS AND DESIGN OF NUMERICAL SCHEMES FOR GAS DYNAMICS, 2: ARTIFICIAL DIFFUSION AND DISCRETE SHOCK STRUCTURE , 1994 .

[7]  Probing numerical fluxes - Mass flux, positivity, and entropy-satisfying property , 1997 .

[8]  J. Quirk.  A Contribution to the Great Riemann Solver Debate , 1994 .

[9]  S. Osher,et al.  Upwind difference schemes for hyperbolic systems of conservation laws , 1982 .

[10]  Meng-Sing Liou,et al.  An Accurate and Robust Flux Splitting Scheme for Shock and Contact Discontinuities , 1997, SIAM J. Sci. Comput..

[11]  P. Roe,et al.  On Godunov-type methods near low densities , 1991 .

[12]  Domenic D'Ambrosio,et al.  Upwind methods and carbuncle phenomenon , 1998 .

[13]  J. Edwards.  A low-diffusion flux-splitting scheme for Navier-Stokes calculations , 1997 .

[14]  R. Schwane,et al.  ON THE ACCURACY OF UPWIND SCHEMES FOR THE SOLUTION OF THE NAVIER-STOKES EQUATIONS , 1987 .

[15]  M. Liou.  A Sequel to AUSM , 1996 .

[16]  R. C. Swanson.  On Some Numerical Dissipation Schemes , 1998 .

[17]  Shigeru Obayashi,et al.  Practical formulation of a positively conservative scheme , 1994 .

[18]  S. Imlay,et al.  Blunt-body flow simulations , 1988 .

[19]  M. Liou,et al.  A New Flux Splitting Scheme , 1993 .

[20]  Bernd Einfeld.  On Godunov-type methods for gas dynamics , 1988 .

[21]  Hong-Chia Lin,et al.  Dissipation additions to flux-difference splitting , 1991 .