Comparative Study Of Convergence Of CfdCommercial Codes When Simulating Dense Underflows
Abstract
In this paper, we present numerical simulations of saline, discontinuous density currents,
in two and three dimensions. The simulation of these flows poses a severe challenge for all codes. In
fact, the currents present characteristic flow instabilities at the interface which are constituted by
small spatial scales. A very fine resolution of these scales is needed to adequately capture the
instabilities.
The two-dimensional simulations reported herein were performed with two CFD codes. The first one
is a comprehensive finite-element platform, whereas the other one is a commercial code. The runs
were undertaken under quite similar conditions. Simulations show that only when the mesh employed
in the commercial code is strongly refined a convergent solution is attained, which is similar to the
solution obtained with the finite-element CFD code. This result would warn about the indiscriminate
use of commercial codes with supposedly “fine” meshes when simulating complex underflows. The
solution with the finite-element code, in turn, shows the shedding of large vortices containing salt.
These vortices do not have a physical basis, but they correspond to the true solution of the twodimensional
Navier-Stokes Equations.
in two and three dimensions. The simulation of these flows poses a severe challenge for all codes. In
fact, the currents present characteristic flow instabilities at the interface which are constituted by
small spatial scales. A very fine resolution of these scales is needed to adequately capture the
instabilities.
The two-dimensional simulations reported herein were performed with two CFD codes. The first one
is a comprehensive finite-element platform, whereas the other one is a commercial code. The runs
were undertaken under quite similar conditions. Simulations show that only when the mesh employed
in the commercial code is strongly refined a convergent solution is attained, which is similar to the
solution obtained with the finite-element CFD code. This result would warn about the indiscriminate
use of commercial codes with supposedly “fine” meshes when simulating complex underflows. The
solution with the finite-element code, in turn, shows the shedding of large vortices containing salt.
These vortices do not have a physical basis, but they correspond to the true solution of the twodimensional
Navier-Stokes Equations.
Full Text:
PDFAsociación Argentina de Mecánica Computacional
Güemes 3450
S3000GLN Santa Fe, Argentina
Phone: 54-342-4511594 / 4511595 Int. 1006
Fax: 54-342-4511169
E-mail: amca(at)santafe-conicet.gov.ar
ISSN 2591-3522