[hejiandong]

Could anyone attached a cfg file for optimization for unsteady flows?

[economon]

Originally Posted by hejiandong

Could anyone attached a cfg file for optimization for unsteady flows?

Hi,

Thanks for the note, and glad to hear that you're interested in unsteady adjoints. Adjoint-based aerodynamic design for unsteady flows on dynamic meshes is a current research topic for us in SU2, and we are still working on these features. That being said, I am attaching a config file for an optimization problem involving a pitching airfoil. The shape design problem is for time-averaged drag minimization with a lift and a geometric constraint on the area of the airfoil. It should work with the standard shape_optimization.py script with the most recent version of the code on GitHub. Please feel free to give it a try!

All the best,
Tom

[hejiandong]

Originally Posted by economon

Hi,

Thanks for the note, and glad to hear that you're interested in unsteady adjoints. Adjoint-based aerodynamic design for unsteady flows on dynamic meshes is a current research topic for us in SU2, and we are still working on these features. That being said, I am attaching a config file for an optimization problem involving a pitching airfoil. The shape design problem is for time-averaged drag minimization with a lift and a geometric constraint on the area of the airfoil. It should work with the standard shape_optimization.py script with the most recent version of the code on GitHub. Please feel free to give it a try!

All the best,
Tom

Many thanks for your help!
I have successfully done this test case.
But there is a problem, when i tried SU2_CFD NACA64A010......the result of lift cofficient is different from the result in your paper "Unsteady Aerodynamic Design on Unstructured Meshes with Sliding Interfaces"

[hejiandong]

Originally Posted by hejiandong

Many thanks for your help!
I have successfully done this test case.
But there is a problem, when i tried SU2_CFD NACA64A010......the result of lift cofficient is different from the result in your paper "Unsteady Aerodynamic Design on Unstructured Meshes with Sliding Interfaces"

Sorry! i find the reason!
i made a mistake when translate the coordinate from time step to angle of attack.