CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Flux vector splitting of Inflow/outflow BC (http://www.cfd-online.com/Forums/main/4569-flux-vector-splitting-inflow-outflow-bc.html)

Quain Tchew March 24, 2002 22:30

Flux vector splitting of Inflow/outflow BC
 
Hi

I'm reading a code. In the code, inflow/outflow bc is handled with Flux vector splitting:

normal flux on the boundary face of the cell = integrate_on_the_face ( ((A+)*(Win)+(A-)*(Wex))*n )

A+ is the positive Jacobian Matrix A- is the negative Jacobian Matrix Win is the primal var on the inner side of the face Wex is the primal var on the outer side of the face

What does that mean? Where can I fined a detail introduction about Flux vector splitting of Inflow/outflow BC on web?

gita March 25, 2002 10:43

Re: Flux vector splitting of Inflow/outflow BC
 
Hi,

You can refer to this book: C. Hirsch, Numerical Computation of Internal and External Flows, Vol. 1 and 2, 1988 Wiley.

There are different types of flux vector splitting schemes like van Leer(based on Mach No),Stegar-Warming(eigenvalue based),hybrid FVS(flux vector splitting) like AUSM(Advection Upstream Splitting Method-pressure based splitting).

If you're new to these schemes(flux-difference and flux-vector),then I suggest you should read Hirsh so that you gain some insight into those numerical methods. Basically these compressible schemes split the fluxes into positive and negative fluxes(inviscid) across each cell of the volume so that the correct flow information is incorporated(direction of the propagation of wave speeds). When you deal with compressible NS flows,you apply these flux-difference/flux-vector/hybrid splitting schemes only for the inviscid part. Do you want to know the physical meaning behind these different splittings? I can explain... Regards

Quain Tchew March 25, 2002 10:54

Re: Flux vector splitting of Inflow/outflow BC
 
Hi! gita, Thank you.

I have read some books about that. But I still don't know how to apply flux vector splitting in inflow and out flow boundary.

I met this problem when I was reading the NSC2KE code(http://pauillac.inria.fr/cdrom/prog/unix/nsc2ke/eng.htm). In the page 10 of the guide of NSC2KE, there is a paragraph about "Inflow and outflow weak boundary". But I can not understand.

Regards Quain Tchew

Praveen Chandrashekar March 25, 2002 12:25

Re: Flux vector splitting of Inflow/outflow BC
 
Refer to the paper by Steger and Warming as suggested in the manual.

Quain Tchew March 25, 2002 20:43

Re: Flux vector splitting of Inflow/outflow BC
 
oh, I can't find it.

But I think I have catched on it.


All times are GMT -4. The time now is 15:59.