[kayla1994]

hello,

I'm running the shape optimization to minimize the drag of a NACA airfoil. I used the configuration present in the test cases and i substituted the native mesh with my own mesh, but i don't know why the process stops after the first iteration.
Any idea about the source of the problem?
Also i don't understand what is the scale factor of the objective function.


Thanks for help

[kayla1994]

i read the the force_breakdown.dat in the DIRECT folder prduced by the first step of the optimization and i noticed all the values are zero (CL, CD etc.), so this is the problem that stops the optimization at the first step. I don't think is a mesh problem because i run the same mesh and run correctly in the direct simulation, so i think is the .cfg

[kayla1994]

Maybe i found the problem

here a screen of my terminal

http://i.imgur.com/mcJPray.jpg

It seems i did a mistake in the "initial guess of indipendent variables". Honestly i have no idea whis is the right manner to set di parameters

[kayla1994]

Now i think i set correctly all the parameters, and these are the errors

http://imgur.com/a/7kDpx

[talbring]

Hi kayla1994,

you can have a look into DESIGNS/DSN_002/log_Direct.out to see why it fails.

[kayla1994]

Originally Posted by talbring

Hi kayla1994,

you can have a look into DESIGNS/DSN_002/log_Direct.out to see why it fails.

Thanks for your answer talbring

this is the message in log_DIRECT.out

\
!!! Error !!!
NACA5512_2_deform.cgns is not a CGNS file.
Now exiting...
\

NACA5512_2 is the name of my mesh in cgns format

You can find the part of the .cfg file that contains the INPUT/PUTPUT information at this link http://imgur.com/8G2Ls3p

Maybe it can help


Thanks for help

[fpalacios]

Hi Kayla,

Please keep in mind that the shape design of SU2 only works with .su2 format (not cgns). This is because the grid deformation tool are not able to write the deformed grid in cgns format.

Best,
Francisco Palacios