[Amr Emad Ezzat]

Hello,

Is there any resource or algorithm for how to apply the flux splitting upwind method for compressible flow using the finite difference method so I can follow?

I want to simulate shockwave reflection on a surface and the central scheme did not do well in this situation.

[FMDenaro]

Originally Posted by Amr Emad Ezzat

Hello,

Is there any resource or algorithm for how to apply the flux splitting upwind method for compressible flow using the finite difference method so I can follow?

I want to simulate shockwave reflection on a surface and the central scheme did not do well in this situation.

Have a look here http://www.clawpack.org/gallery/index.html


of course in the framework of "flux" splitting you have to think about the FV formulation.

[sbaffini]

I am afraid that there is no such concept as flux splitting in FD, as it specifically involves splitting a flux on a given face in relation to the underlying Riemann problem there.

You can certainly arrange a FV approach on the same FD grid you are using and convert it back to the FD formulation, but that would require some arbitraryness in positioning FV faces. Then, why not just using FV?

FV schemes are typically built as interpolation to faces + limiting + riemann problem. Flux splitting only solves one of the three parts of the scheme. FD schemes, in contrast, are typically built as a single piece for a convective derivative.