How is the normal unit vector calculated for 2D airfoil surface gradients?

Ry10 · April 28, 2016, 17:23

I am finding the surface gradient of a 2D airfoil. I have read that the surface gradient that surface_adjoint.csv gives you is the gradient in the normal direction at each point. I want to know how you calculate that normal direction. Is the normal direction at each point the summation of the normal vectors from the lines to the two adjacent points? Or is the normal vector calculated in some other way? I am using my own shape design variable (CST) and I need to know which direction the surface gradient is pointing at all the points.

Thanks!

kmandar · May 2, 2016, 13:06

Take a lot at this thread:
http://www.cfd-online.com/Forums/su2...y-normals.html

JPBLourenco · February 21, 2018, 05:41

Hi everyone.

Does anyone know how to use CST method (class shape transformation) in SU2? In the configure file it's not clear how to do it. I try this configuration
DEFINITION_DV= (20, 1.0 |airfoil| 0, 1, 8) for shape optimization airfoil but nothing change in the airfoil.

I'm using the SU2 version5.0

Thanks everyone

PawHewitt · February 22, 2018, 09:14

I'm not entirely sure if the CST has been fully implemented for deformation yet. It seems the subroutine exits in the grid_movement_structure file but I don't see a call to it in the SetSurface_Deformation function which works out the displacements.
You might have add the call yourself, following the same pattern as the Hicks Hennes

fpalacios · February 24, 2018, 09:57

Yep, it is an interesting hidden (?) capability implemented by Ryan Barrett. Looking at the code it seems almost complete:
void CSurfaceMovement::SetCST(CGeometry *boundary, CConfig *config, unsigned short iDV, bool ResetDef) is there
it is also in the python scripts (DV number 51) in tools.py
and I see the reference in SU2_DOT and SU2_GEO...
however it is not in the config_template config dile.

If I were you I'll try again but using number 51 instead of 20 and checking what are the input parameters. If you make it work then it would be great if you could create a pull request in GitHub and add the info to the config template.

I'm looking at SU2 v6.0. Another excellent possibility is to contact Ryan Barrett (look for his pull request in GitHub).

By the way, we are organizing the SU2 community of developers in the SU2 IDS. If you have a while please take a look at www.su2devsociety.org

Best,
Francisco

JPBLourenco · February 27, 2018, 06:28

Thank you for your quick reply Philip Hewitt and Francisco Palacios.

From what I realized, the code doesn't have the CST fully implemented. The subroutine exists in the grid_movement_structure file and also some python files. When I try to run a shape optimization simulations the objective function still with the same value in all functions evaluated because the CST method isn't called. In fact, CST is SU2_DOT and SU2_GEO but something is missing. I will look to Hicks Hennes and try to follow the same procedure.

In SU2 v5.0, when I look to the tools.py file, the DV number is set to 20 and not to 51 (like v6.0). Case I make it works, I will create a pull request in Github and add the info.

Thanks to share the information. I will contact if I have some doubts and check the SU2 community of developers.

Best,
João Lourenço

April 28, 2016, 17:23	How is the normal unit vector calculated for 2D airfoil surface gradients?	#1
Ry10 New Member Ryan Barrett Join Date: May 2015 Posts: 15 Rep Power: 10	I am finding the surface gradient of a 2D airfoil. I have read that the surface gradient that surface_adjoint.csv gives you is the gradient in the normal direction at each point. I want to know how you calculate that normal direction. Is the normal direction at each point the summation of the normal vectors from the lines to the two adjacent points? Or is the normal vector calculated in some other way? I am using my own shape design variable (CST) and I need to know which direction the surface gradient is pointing at all the points. Thanks!

February 21, 2018, 05:41	CST configure file	#3
JPBLourenco Member João Lourenço Join Date: Jan 2017 Posts: 32 Rep Power: 9	Hi everyone. Does anyone know how to use CST method (class shape transformation) in SU2? In the configure file it's not clear how to do it. I try this configuration DEFINITION_DV= (20, 1.0 \|airfoil\| 0, 1, 8) for shape optimization airfoil but nothing change in the airfoil. I'm using the SU2 version5.0 Thanks everyone

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to get the surface normal vector	ohrmond	CFX	3	February 1, 2018 05:45
[snappyHexMesh] determining displacement for added points	CFDnewbie147	OpenFOAM Meshing & Mesh Conversion	1	October 22, 2013 09:53
Problem with restart solution in shape_optimization.py	robyTKD	SU2 Shape Design	21	May 29, 2013 09:26
Normal - Helical Surface	m. malik	Main CFD Forum	3	February 3, 2006 12:56
CFX4.3 -build analysis form	Chie Min	CFX	5	July 12, 2001 23:19

May 2, 2016, 13:06		#2
kmandar Member Mandar Kulkarni Join Date: Nov 2013 Location: Virginia Tech, Blacksburg, VA Posts: 52 Rep Power: 12	Take a lot at this thread: http://www.cfd-online.com/Forums/su2...y-normals.html

February 22, 2018, 09:14		#4
PawHewitt New Member Philip Hewitt Join Date: Oct 2013 Posts: 4 Rep Power: 12	I'm not entirely sure if the CST has been fully implemented for deformation yet. It seems the subroutine exits in the grid_movement_structure file but I don't see a call to it in the SetSurface_Deformation function which works out the displacements. You might have add the call yourself, following the same pattern as the Hicks Hennes

February 24, 2018, 09:57		#5
fpalacios Super Moderator Francisco Palacios Join Date: Jan 2013 Location: Long Beach, CA Posts: 404 Rep Power: 15	Yep, it is an interesting hidden (?) capability implemented by Ryan Barrett. Looking at the code it seems almost complete: void CSurfaceMovement::SetCST(CGeometry boundary, CConfig config, unsigned short iDV, bool ResetDef) is there it is also in the python scripts (DV number 51) in tools.py and I see the reference in SU2_DOT and SU2_GEO... however it is not in the config_template config dile. If I were you I'll try again but using number 51 instead of 20 and checking what are the input parameters. If you make it work then it would be great if you could create a pull request in GitHub and add the info to the config template. I'm looking at SU2 v6.0. Another excellent possibility is to contact Ryan Barrett (look for his pull request in GitHub). By the way, we are organizing the SU2 community of developers in the SU2 IDS. If you have a while please take a look at www.su2devsociety.org Best, Francisco

February 27, 2018, 06:28		#6
JPBLourenco Member João Lourenço Join Date: Jan 2017 Posts: 32 Rep Power: 9	Thank you for your quick reply Philip Hewitt and Francisco Palacios. From what I realized, the code doesn't have the CST fully implemented. The subroutine exists in the grid_movement_structure file and also some python files. When I try to run a shape optimization simulations the objective function still with the same value in all functions evaluated because the CST method isn't called. In fact, CST is SU2_DOT and SU2_GEO but something is missing. I will look to Hicks Hennes and try to follow the same procedure. In SU2 v5.0, when I look to the tools.py file, the DV number is set to 20 and not to 51 (like v6.0). Case I make it works, I will create a pull request in Github and add the info. Thanks to share the information. I will contact if I have some doubts and check the SU2 community of developers. Best, João Lourenço