Hello Everyone ,
As we know that at higher angle of attacks and low Reynolds's number regimes delta wings tend to undergo oscillation about longitudinal axis , that is Rolling Oscillation . The commonly know as Wing Rock Phenomenon .
How can I go about modelling these kind of oscillations in a commercial CFD package like Ansys Fluent.
Kindly enlighten me with your kind suggestions .
Regards
Hi Dear Aamir
Try spallart-allmaras turbulence model.
Best regards
Hello Aamir,
You should be able to model this kind of cases by some tools available in Fluent. If you are after a 2-way coupling, so, the "MpCCI from Fraunhofer SCAI" external software tool to be used and would be the best option. However, if you expect a small structural deformation, I would suggest the "FSI mapping" toolkit which does not need any external tool as in 2-way coupling.
I hope this helps
Best regards
Hayder
Thank you so much Hayder Al-Jelawy for you kind response .
Actually I am using a rigid wing with no structural deformation to take into account , I just want to see the spatial oscillations of the rigid delta wing about its longitudinal axis as it enters the high alpha regimes at low Reynolds's number . Hope you are getting my point .
Thanks Again
Reza Khazaee : Excuse me, but what has a 1-equation turbulence model to do with the question asked by Aamir Sultan ?
Hayder Al-Jelawy : Doing system coupling with MpCCI is an outdated method of doing coupling between a CFD code and a mechanical code. And it is not at all required, if you couple CFD and structural mechanics code of the same software vendor like ANSYS.
Aamir Sultan : If you keep the wing rigid, than you cannot predict the oscillations of the wing, since it is a response of a flexible wing to pressure oscillations in the flow field. So if that is a response to pressure fluctuations, than we of course 1st need to resolve these pressure fluctuations (which occur as a result of flow separation and forming separating vortices) accurately by the CFD solver and the used turbulence model. It is a well known fact, that 1- and 2-equation turbulence models cannot predict the separation of turbulent vortices with all there time and spatial scales very well. They give us a more or less accurate representation of the MEAN velocity filed information, but not an accurate result for the transient behavior. But oscillations are a transient phenomenon.
Consequence: we need at least a good hybrid RANS-LES turbulence model like SAS-SST or even better SBES. If time and resources do not count, than a high-fidelity ELES, DDES or WMLES can be applied. By that we get the timely resolved pressure fluctuations at the wing surface.
With that we can now run what ANSYS now calls the System Coupling (1- or 2-way FSI) between ANSYS Fluent and ANSYS Structural Mechanics solvers. This is a well established coupling method in the ANSYS software universe, also it is not totally trivial to run it in transient, on a larger Linux Cluster and for a longer coupling period in a stable manner in order to get the structural response of the wing to the pressure oscillations from the vortex flow around the wing.
To save time and resources, the fully developed flow field should be computed with only Fluent and using a rigid wing. Using monitors in the flow field close to interesting locations on the wing or close to the wing and analyzing the monitor data for temporal oscillations, potentially already a first conclusion about wing oscillations can be drawn even before actually running a coupled FSI.
Best regards,
Dr. Th. Frank.
- Badges
- Science topic