Hi, cfd gurus!

Can anybody tell me if there exists any technique for imposing non-reflective BC to 1D and 2D problems governed by the Euler equations?

I mean the technique which can help to essentially avoid reflections from the far field, to assure full absorption of the notable moving finite-amplitude waves?

Any advice will be highly appreciated.

Thanks in advance,

Andrei

In order to impose non reflective boundary conditions at the boundary one has to impose the conditions on the characteristics of the flow rather than on the primitive variables. The best I can do is to give you some references on the subject, and if you have some more particular questions, then just post it here and I'll try to answer them.

Review Paper: Givoli, 1991, Journal of Computational Physics, volume 94, page 1.

See the open boundary treatment and discussions in the following papers:

Abarbanel et al. 1991, J. Fluid Mechanics, 225,557

Gottlieb, Gunzburger and Turkel, 1982, SIAM J. Numer. Anal. 19, 671.

Wasberg and Andreassen, Comput. Meth. Appl. Mech. Engrg, 80, 459 (1990).

Godon and Shaviv, Comput. Meth. Appl. Mech. Engrg, 1992, 110, 171.

Cheers,

Patrick

Take a look at the work by Poinsot and Lele (it is a JCP paper from 1992, I forget the exact vol. no and pp). It works excellently for pure euler flows, even works very well if carefully done for vortex convection through boundaries.

Good Luck Srinivasan