Non-conformal periodic boundary conditions in su2

Black Margician · June 5, 2024, 03:45

Hi friends,

I'm trying to simulate a single-passage turbine vane using su2. However, because the mesh on both sides of the computational domain is non-conformal (i.e. not 1:1), the solver gives me the following message:

Bad match for point 298530. Nearest donor distance: 2.9609723857e-05.

Finally the solver diverges in less than 10 iterations. How can I set the periodic boundary conditions to solve this problem? Should I set some interpolation boundary conditions for the periodic faces? If anyone has had this type of problem and can help, I would appreciate it.

Thanks,
JZ

Tongtong1919 · June 7, 2024, 22:39

I have encountered this issue before and was fortunate to resolve it at that time. Now I am sharing my experience with you in the hope that it will be helpful.

The problem with point mismatch is usually due to improperly set periodic boundaries. You can troubleshoot the error from the following aspects:

Check the direction of node matching, whether it is counterclockwise or clockwise.
Check the units of the mesh nodes. I am not sure how you generated the mesh. For turbomachinery, my habit is to use AutoGrid, where the default unit is either mm or m (I can't quite remember). Usually, importing meshes generated by commercial software into other commercial software shouldn't be an issue (even if they are from different companies), but it might not always work when importing into SU2. Ensure that the units of the node coordinates in your mesh file are consistent with those in the tutorial mesh file; this should have an effect.
Good luck!

Black Margician · July 16, 2024, 09:02

Thanks Tongtong,

Actually, I have checked the direction and the units of the mesh. However, since the periodic boundary mesh in my case is naturally non-conformal (the mesh nodes are naturally not matching!), maybe the general 1:1 periodic boundary condition cannot be applied to my case. So I just wonder if su2 can deal with this kind of problem?

Regards,
JZ

joshkellyjak · July 18, 2024, 11:34

To my knowledge SU2 is not capable of non-conformal periodicity in the current release.

June 5, 2024, 03:45	Non-conformal periodic boundary conditions in su2	#1
Black Margician New Member Jz Fan Join Date: Jun 2024 Posts: 2 Rep Power: 0	Hi friends, I'm trying to simulate a single-passage turbine vane using su2. However, because the mesh on both sides of the computational domain is non-conformal (i.e. not 1:1), the solver gives me the following message: Bad match for point 298530. Nearest donor distance: 2.9609723857e-05. Finally the solver diverges in less than 10 iterations. How can I set the periodic boundary conditions to solve this problem? Should I set some interpolation boundary conditions for the periodic faces? If anyone has had this type of problem and can help, I would appreciate it. Thanks, JZ

July 18, 2024, 11:34		#4
joshkellyjak New Member Josh Kelly Join Date: Dec 2018 Posts: 13 Rep Power: 7	To my knowledge SU2 is not capable of non-conformal periodicity in the current release. Tongtong1919 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
3D Windturbine simulation in SU2	k.vimalakanthan	SU2	15	October 12, 2023 05:53
Periodic boundary conditions using Fluent Meshing	Alice_1994	FLUENT	0	November 21, 2022 09:24
Turbomachinery Mass imbalance	sheaker	CFX	12	September 5, 2019 08:09
CGNS Boundary conditions using SU2	denzell	SU2	3	July 9, 2018 05:58
Error finding variable "THERMX"	sunilpatil	CFX	8	April 26, 2013 07:00

June 7, 2024, 22:39		#2
Tongtong1919 New Member Tongtong Join Date: Jan 2024 Posts: 24 Rep Power: 2	I have encountered this issue before and was fortunate to resolve it at that time. Now I am sharing my experience with you in the hope that it will be helpful. The problem with point mismatch is usually due to improperly set periodic boundaries. You can troubleshoot the error from the following aspects: Check the direction of node matching, whether it is counterclockwise or clockwise. Check the units of the mesh nodes. I am not sure how you generated the mesh. For turbomachinery, my habit is to use AutoGrid, where the default unit is either mm or m (I can't quite remember). Usually, importing meshes generated by commercial software into other commercial software shouldn't be an issue (even if they are from different companies), but it might not always work when importing into SU2. Ensure that the units of the node coordinates in your mesh file are consistent with those in the tutorial mesh file; this should have an effect. Good luck!

July 16, 2024, 09:02		#3
Black Margician New Member Jz Fan Join Date: Jun 2024 Posts: 2 Rep Power: 0	Thanks Tongtong, Actually, I have checked the direction and the units of the mesh. However, since the periodic boundary mesh in my case is naturally non-conformal (the mesh nodes are naturally not matching!), maybe the general 1:1 periodic boundary condition cannot be applied to my case. So I just wonder if su2 can deal with this kind of problem? Regards, JZ