I am analyzing adjoint solution around NACA0012. In adjoint.vtk file I see 3 different fields: PsiRho, PsiE and Phi. Can anybody tell me if they have a physical interpretation? In particular, I am not sure how to interpret them in the flow region.
Another doubt is about sensitivity in the surface_adjoint.csv: is it Cd = Cd_old + S * \deltan
where S is the sensitivity and \deltan is the displacement[meter] in the normal to airfoil direction?
Last question: is it possible to perform a fully turbulent(no FROZEN viscosity) adjoint simulation with SA turbulent model?
Thank you very much,
I think this article http://www.stanford.edu/~fasispg/art..._March2012.pdf could be useful for you. In fact there is a non-frozen viscosity implementation that we are validating (our team have some articles this summer where the viscous adjoint is a key element).
|All times are GMT -4. The time now is 20:47.|