CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS (https://www.cfd-online.com/Forums/ansys/)
-   -   Airfoil - Inverse Design/ Shape Optimization (https://www.cfd-online.com/Forums/ansys/236741-airfoil-inverse-design-shape-optimization.html)

Opal June 14, 2021 04:04

Airfoil - Inverse Design/ Shape Optimization
 
Good day,

I have a question about the following task:
Based on a given NACA profile and the pressure distribution evident from an initial simulation, I would like to generate a new profile with the same/ a very similar pressure distribution after adjusting the flow conditions.

Now I need a hint how to realize the project in ANSYS.

I have already parameterized the pressure side of the single profile in the DesignXplorer (x,y-coordinates). However, I currently do not know how I can integrate the static pressures along the PS as target values. Is for the above mentioned problem another methodology possibly more advantageous (e.g. Adjoint Solver, BladeModeler)?
Especially with respect to the later following, intended geomteric study and airfoil generation in a cascade flow.


I would be pleased about your assistance.


With kind regards
Opal

hoyos98 March 7, 2022 16:20

You can find the optimal airfoil shape (minimizing the drag) but fulfilling a target lift coefficient on this website: https://aeroptimal.com/airfoil . The methodology is under review in journals. The CFD method employs OpenFOAM.

Indra March 8, 2022 04:21

Quote:

Originally Posted by Opal (Post 805980)
Good day,

I have a question about the following task:
Based on a given NACA profile and the pressure distribution evident from an initial simulation, I would like to generate a new profile with the same/ a very similar pressure distribution after adjusting the flow conditions.

Now I need a hint how to realize the project in ANSYS.

I have already parameterized the pressure side of the single profile in the DesignXplorer (x,y-coordinates). However, I currently do not know how I can integrate the static pressures along the PS as target values. Is for the above mentioned problem another methodology possibly more advantageous (e.g. Adjoint Solver, BladeModeler)?
Especially with respect to the later following, intended geomteric study and airfoil generation in a cascade flow.


I would be pleased about your assistance.


With kind regards
Opal

Adjoint solver is different from parameter-based optimisation in workbench. Adjoint will help you achieve your observable values(for eg: L/D ratio) by modifying the shape of airfoil, so you won't be running many design points. It is a gradient based solver that calculates different equations within fluent.

If you run a prameter study in workbench along with designexplorer, it'll give you the performance of foil at each input parameter and you can get a response surface that'll tell you how performance changes with you input parameter.

Both ways can be used, adjoint although gives you some unique shapes.


All times are GMT -4. The time now is 19:37.