Coupling Boundary Layer Methods with Inviscid Flow Solvers
Tuesday, April 24, 2012
3:30PM – 5PM
In the 1980’s, a great deal of work focused on interactive boundary layer methods coupled with inviscid flow solvers. One of the best of these methods was developed by Prof. Mark Drela of MIT. At Boeing, we developed the TRANAIR code by coupling a solution adaptive full potential solver to Drela’s strip integral boundary layer with various 3 dimensional corrections. Research in the field of CFD has focused almost exclusively on Reynolds Averaged Navier Stokes starting around 1992.
In this seminar, we will discuss extension of TRANAIR to separated flows and show why these methods deserve a second look by comparing the results to Navier Stokes codes and to test data. We will also discuss some of the weaknesses of steady state Navier Stokes simulations, including the existence of multiple solutions, the role of time dependent flow phenomena such as vortex shedding, and the long continuing struggle to successfully implement solution adaptive grid refinement in these methods.
Additionally, in multi-disciplinary simulations, cycle time is critical because of the large number of flight conditions that must be considered. This leads us to the development of an Euler inviscid solver coupled with integral boundary layer methods.
Hosted by Leszek Demkowicz