|
[Sponsors] |
July 18, 2017, 12:20 |
DV_PARAM and DEFINITION_DV
|
#1 |
Member
Join Date: May 2017
Posts: 41
Rep Power: 9 |
Hi,
it may be a basic question but I'm confused about the usage of "DV_PARAM". We are already defining the design variables in "DEFINITION_DV", so why would we need to define "DV_PARAM" and what's it used for? For instance, when we have DV_PARAM= ( 1, 0.5 ) what are "1" and "0.5" representative of? Thanks. |
|
August 12, 2017, 10:56 |
|
#2 |
Super Moderator
Francisco Palacios
Join Date: Jan 2013
Location: Long Beach, CA
Posts: 404
Rep Power: 15 |
Hi Ryan, this is a very good question.
DEFINITION_DV is only used by the shape_optimization.py script. In other words SU2 uses DEFINITION_DV to specify all the design variables in your optimization problem. As you can see, DEFINITION_DV is a combination of the kind of design variable, the scaling factor, where are you going to apply the design variable, and a definition of the design variable (everything is there). On the other hand DV_PARAM is only used in case you want to manually deform a computational grid using SU2_DEF. if you manually define DV_KIND, DV_MARKER, DV_PARAM and DV_VALUE you would be able to apply that particular deformation to the surface. e.g. if you set DV_KIND=HICKS_HENNE DV_MARKER= ( airfoil ) DV_PARAM= ( 1, 0.5 ) DV_VALUE= 0.01 and execute SU2_DEF, then SU2 will apply the bump to the airfoil surface. Long story short, DEFINITION_DV is for shape design and DV_PARAM is for grid/surface independent deformations in case you are interested. Best, Francisco |
|
August 14, 2017, 23:28 |
|
#3 | |
Member
Join Date: May 2017
Posts: 41
Rep Power: 9 |
Quote:
|
||
Tags |
design varibles, optimizaion, su2, su2 adjoint, su2 optimization |
Thread Tools | Search this Thread |
Display Modes | |
|
|