selective frequency damping in SU2

Sigi · November 6, 2024, 15:16

Hi,

Is there any way to apply the selective frequency damping (SFD) technique in SU2?

I am wondering if someone has already implemented this.

Thanks.

bigfootedrockmidget · November 7, 2024, 04:55

I do not think this was ever implemented in SU2. But it might be very beneficial to have it, especially in combination with the multigrid method for convergence acceleration.

What I understand is that after solving for the flow variables q in (pseudo-)time using implicit Euler, you have to solve a second system that solves the filtered flow variables (q, q_f). This does not sound too intrusive. If you would like to implement something like this, please go ahead!

Link to some research:
https://research.tudelft.nl/files/46..._etal_2018.pdf

November 6, 2024, 15:16	selective frequency damping in SU2	#1
Sigi New Member Join Date: Jan 2023 Posts: 12 Rep Power: 3	Hi, Is there any way to apply the selective frequency damping (SFD) technique in SU2? I am wondering if someone has already implemented this. Thanks.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Dynamic FSI with SU2 Compressible Solver and SU2 FEM Solver Diverges when restarted	sangeet	SU2	7	August 3, 2022 02:54
Introducing SU2 International Developers Society (IDS)	fpalacios	SU2 News & Announcements	1	June 17, 2019 23:38
Setting alpha and beta damping on two different materials (Ansys Workbench)	walteppich	ANSYS	1	August 29, 2017 03:56
wrong SU2 calculation for lift and drag coefficient for NAC4421	mechy	SU2	7	January 9, 2017 06:18
Input for Mass, Frequency and Damping	Sham	FLUENT	0	March 29, 2007 23:37

November 7, 2024, 04:55		#2
bigfootedrockmidget Senior Member bigfoot Join Date: Dec 2011 Location: Netherlands Posts: 676 Rep Power: 21	I do not think this was ever implemented in SU2. But it might be very beneficial to have it, especially in combination with the multigrid method for convergence acceleration. What I understand is that after solving for the flow variables q in (pseudo-)time using implicit Euler, you have to solve a second system that solves the filtered flow variables (q, q_f). This does not sound too intrusive. If you would like to implement something like this, please go ahead! Link to some research: https://research.tudelft.nl/files/46..._etal_2018.pdf