|
[Sponsors] |
How to perform larger deformations while mantaining good mesh with adjoint solver |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
October 13, 2023, 12:27 |
How to perform larger deformations while mantaining good mesh with adjoint solver
|
#1 |
New Member
Join Date: Oct 2023
Posts: 1
Rep Power: 0 |
Hello everyone. I am starting with the adjoint solver for my master thesis, in particular, I am doing an adjoint shape optimisation for a dual-stage wing (tutorial of adjoint shape optimisation). Following the tutorial, I used the adjoint mesh sensitivities to deform the volumetric mesh, as specified in the tutorial. However, after some loops executing the adjoint, deforming the mesh and running the primal flow again, I get an error of having negative volume cells. After doing some research, I found that the mesh quality of the volumetric mesh is very bad after all the morphing made.
Hence, I am looking to still use control points but to be able to perform larger deformations. I tried to do the following operations: 1)Solve adjoint 2)Deform mesh 3)Mesh 3.1)Surface mesh 3.2)Volume mesh 4)Run primal flow again 5)Do steps 1 to 4 again After, performing step 2, the mesh was deformed, but when doing step 3.a) Surface mesh the output mesh is the same original mesh that I had without any deformation. So how can I make a new surface mesh from the deformed geometry (volume mesh)? How can I fix this? Secondly, I tried using the morph surface option, as I read it can be used to perform larger displacements. However, I can not use it and still use the control points. The morph surface mesh only works with surface sensitivities and not control points, right? The steps using the surface mesh are: 1)Solve adjoint with surface sensitivity 2)Mesh 2.1)Morph surface mesh Field Function: related to the surface sensitivy 2.2)Surface mesh 2.3)Volume mesh 3) Run primal flow again 4) Do steps 1 to 3 again In any case, I am looking to find a way to have larger deformations while maintaining good mesh quality, as I am sure there has to be a way Thank you for your help |
|
Tags |
adjointoptimisation, mesh, star ccm+ |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[snappyHexMesh] Is it a good mesh for a square cylinder ? | belier1988 | OpenFOAM Meshing & Mesh Conversion | 2 | September 10, 2020 12:59 |
Gambit problems | Althea | FLUENT | 22 | January 4, 2017 04:19 |
[snappyHexMesh] No layers in a small gap | bobburnquist | OpenFOAM Meshing & Mesh Conversion | 6 | August 26, 2015 10:38 |
Moving mesh | Niklas Wikstrom (Wikstrom) | OpenFOAM Running, Solving & CFD | 122 | June 15, 2014 07:20 |
Icemcfd 11: Loss of mesh from surface mesh option? | Joe | CFX | 2 | March 26, 2007 19:10 |