ANALYSIS AND IMPLEMENTATION OF THE GAS-KINETIC BGK SCHEME FOR COMPUTATIONAL GAS DYNAMICS
✍ Scribed by Chongam Kim; Kun Xu; Luigi Martinelli; Antony Jameson
- Publisher
- John Wiley and Sons
- Year
- 1997
- Tongue
- English
- Weight
- 587 KB
- Volume
- 25
- Category
- Article
- ISSN
- 0271-2091
No coin nor oath required. For personal study only.
✦ Synopsis
Gas-kinetic schemes based on the BGK model are proposed as an alternative evolution model which can cure some of the limitations of current Riemann solvers. To analyse the schemes, simple advection equations are reconstructed and solved using the gas-kinetic BGK model. Results for gas-dynamic application are also presented. The ®nal ¯ux function derived in this model is a combination of a gas-kinetic Lax±Wendroff ¯ux of viscous advection equations and kinetic ¯ux vector splitting. These two basic schemes are coupled through a non-linear gas evolution process and it is found that this process always satis®es the entropy condition. Within the framework of the LED (local extremum diminishing) principle that local maxima should not increase and local minima should not decrease in interpolating physical quantities, several standard limiters are adopted to obtain initial interpolations so as to get higher-order BGK schemes. Comparisons for well-known test cases indicate that the gas-kinetic BGK scheme is a promising approach in the design of numerical schemes for hyperbolic conservation laws.
📜 SIMILAR VOLUMES
The governing equations of relativistic computational fluid dynamics (CFD) are integrated numerically. The equation of state (EOS) for a gas at relativistic temperature (the thermal energy of a gas particle is on the order of its rest mass energy) is obtained as a polynomial approximation for a gas
Recently, Chae, Kim, and Rho proposed a new gas-kinetic BGK scheme [1]. In their approach, they modified the EFM or KFVS flux component in a gas-kinetic scheme through techniques based on Mach number splitting and Osher's linear subpath solution; see Eqs. ( 30) and ( 31) in [1]. In order to demonstr