OpenFOAM and Adjoint solution

aerospain · March 6, 2014, 12:18

Hi everyone,

I have read about the capability in OpenFOAM to run the Adjoint "version" of the Navier-Stokes equations for shape optimization.

Is there a way of just calculating the Adjoint solution so I can locate the more sensible regions in an afterbody (cartesian) wake in order to design wake stability measures?

Thanks!

sylvester · March 7, 2014, 06:43

Hi,

Probably best to start with the 'adjointShapeOptimizationFoam' solver (a simpleFoam derivative), while maintaining alpha = 0.

Beware though, adjoint simulations are notoriously unstable.

Good luck!

Sylvester

aerospain · March 7, 2014, 06:45

Thank you Sylvester,

I'll give it a go during the weekend.

Where could I find info about keeping stability under control?

Thanks!

sylvester · March 7, 2014, 07:31

Quote:

Originally Posted by aerospain

Where could I find info about keeping stability under control?

If I knew that, my (CFD) life would be so much better!

You could look up my old posts on adjoint, they might help you a bit. Since then I haven't worked the subject anymore, so I haven't made much additional progress.

regards,
Sylvester

March 6, 2014, 12:18	OpenFOAM and Adjoint solution	#1
aerospain Senior Member aerospain Join Date: Sep 2009 Location: Madrid, Spain Posts: 149 Rep Power: 17	Hi everyone, I have read about the capability in OpenFOAM to run the Adjoint "version" of the Navier-Stokes equations for shape optimization. Is there a way of just calculating the Adjoint solution so I can locate the more sensible regions in an afterbody (cartesian) wake in order to design wake stability measures? Thanks!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
adjoint Shape Optimization	mechy	OpenFOAM Running, Solving & CFD	11	June 19, 2017 10:05
Adjoint solver for flow problems	jelmer	OpenFOAM Running, Solving & CFD	11	March 12, 2017 07:52
adjoint method- uniformity	Envolly	OpenFOAM Programming & Development	8	July 9, 2013 02:29
Adjoint Methods in OpenFOAM	treima	OpenFOAM Programming & Development	4	May 24, 2012 06:57
OpenFOAM 1.7.1: Oscillating CFL number	LarsPT	OpenFOAM Running, Solving & CFD	4	September 18, 2010 04:55

March 7, 2014, 06:43		#2
sylvester Member Roland Join Date: Mar 2009 Location: Netherlands Posts: 92 Rep Power: 17	Hi, Probably best to start with the 'adjointShapeOptimizationFoam' solver (a simpleFoam derivative), while maintaining alpha = 0. Beware though, adjoint simulations are notoriously unstable. Good luck! Sylvester

March 7, 2014, 06:45		#3
aerospain Senior Member aerospain Join Date: Sep 2009 Location: Madrid, Spain Posts: 149 Rep Power: 17	Thank you Sylvester, I'll give it a go during the weekend. Where could I find info about keeping stability under control? Thanks!