# Initializing transient analysis using static analysis in two-way FSI simulation

 Register Blogs Members List Search Today's Posts Mark Forums Read

 April 30, 2014, 13:35 Initializing transient analysis using static analysis in two-way FSI simulation #1 Senior Member   Daniel Join Date: Mar 2013 Location: Noshahr, Iran Posts: 348 Rep Power: 21 Hello to all I am planning to do fluid-structure interaction using Fluent an structural in workbench. I know how to setup transient and static two-way FSI separately. However I don't know how to use static analysis results as an initial condition for transient analysis to make it stable. Is it possible to do it in workbench interface? any help would be appreciated Best wishes, Danial Khazaei

 May 3, 2014, 12:14 #2 Senior Member   Daniel Join Date: Mar 2013 Location: Noshahr, Iran Posts: 348 Rep Power: 21 any help? Best wishes

 May 5, 2014, 02:40 #3 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Dear Daniel, Unfortunately i do not know how to solve your problem, but i think that you can help me. I see that you solved static 2 way FSI. Please, give me some advise how to setup this problem. My problem based on calculating lift force on flexible wing. Thank you in advance, sf

May 11, 2014, 12:37
#4
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Dear Daniel, Unfortunately i do not know how to solve your problem, but i think that you can help me. I see that you solved static 2 way FSI. Please, give me some advise how to setup this problem. My problem based on calculating lift force on flexible wing. Thank you in advance, sf
I have gone through system coupling tutorials in order to learn how to setup a case for two-way FSI problem.

Best wishes

 May 12, 2014, 11:14 #5 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Daniel, Is it necessary to write UDF for defining "Deforming" option within "Geometry Definition"? This zone (geometry) surrounding wing geometry (interior field). Thanks, sf

May 12, 2014, 17:24
#6
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Daniel, Is it necessary to write UDF for defining "Deforming" option within "Geometry Definition"? This zone (geometry) surrounding wing geometry (interior field). Thanks, sf
Do you need a prescribed motion for the wing?

 May 13, 2014, 06:01 #7 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 I want to calculate the resulting wing deformation caused by aerodynamic simulation and than lift over deformed wing (fluent-->mechanical-->fluent, with system coupling). Calculated lift force over deformed wing is about 20 % lesser then over rigid configuration and this is pretty realistic situation. It is static 2-way fsi, isn't it? Thank you Daniel for conversation. All the best.

May 13, 2014, 09:41
#8
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid I want to calculate the resulting wing deformation caused by aerodynamic simulation and than lift over deformed wing (fluent-->mechanical-->fluent, with system coupling). Calculated lift force over deformed wing is about 20 % lesser then over rigid configuration and this is pretty realistic situation. It is static 2-way fsi, isn't it? Thank you Daniel for conversation. All the best.
Yes it's a two-way fsi simulation.
Regarding the deformation:

1) prescribed motion of wing like plunging and pitching: you need to write a UDF

2) deformation of wing due to aerodynamic forces only: no need for UDF

 May 14, 2014, 12:19 #9 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Ok, thank you very much. What is about flutter - need or no need for UDF? Did you overcome your problem (how to use static analysis results as an initial condition for transient analysis)? I have read some proposal that solves the problem. It is all about Time Integration ON/OFF within Analysis Settings - try with OFF in 1st second to reach static condition and go forward with transient starting from 1st second.

May 29, 2014, 06:49
#10
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Ok, thank you very much. What is about flutter - need or no need for UDF? Did you overcome your problem (how to use static analysis results as an initial condition for transient analysis)? I have read some proposal that solves the problem. It is all about Time Integration ON/OFF within Analysis Settings - try with OFF in 1st second to reach static condition and go forward with transient starting from 1st second.
Sorry for the late response. I was too busy with rest of my works.

Do you need the flutter to move based on the prescribed motion?

---------

Regarding my question, what I understand is:

1) setting up Fluent and transient structural for two coupling.

2) in order to obtain steady-state solution:

a) Turning off Time Integration in transient structural.

3) after obtaining the solution, start the real transient analysis.

Am I correct?

Best wishes

 June 2, 2014, 18:58 #11 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Correct, but, sorry, I was wrong Daniel. That proposal is for CFX, so, Mechanical don't allow to conduct transient simulation by setting up prestress on that way (within system coupling environment). Try this way: 1) connect (export) steady (converged and saved) fluent with transient fluent 2) connect transient fluent (without initializing) with system coupling engine 3) connect transient structural (with Time Integration ON) with sys coupl eng 4) within sys coupl (Co-Sim. Sequence) set trans struct to do first sequence and fluent 2nd I'm not sure but I think that you want to do this. I did it. Inform me about your progress. Good luck. Daniel_Khazaei likes this.

June 3, 2014, 02:01
#12
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Correct, but, sorry, I was wrong Daniel. That proposal is for CFX, so, Mechanical don't allow to conduct transient simulation by setting up prestress on that way (within system coupling environment). Try this way: 1) connect (export) steady (converged and saved) fluent with transient fluent 2) connect transient fluent (without initializing) with system coupling engine 3) connect transient structural (with Time Integration ON) with sys coupl eng 4) within sys coupl (Co-Sim. Sequence) set trans struct to do first sequence and fluent 2nd I'm not sure but I think that you want to do this. I did it. Inform me about your progress. Good luck.
Thank you.

Yes I have tried that method and system coupling does not allow transient mechanical with stead-state Fluent.

----

Your new suggestion seems reasonable, I will try it soon.

Best wishes

 June 4, 2014, 05:47 #13 Senior Member   Daniel Join Date: Mar 2013 Location: Noshahr, Iran Posts: 348 Rep Power: 21 I have another problem regarding mapping summery in system coupling log file. It reports 0 percent mapped on both sides of FSI interface. I have generated the fluid mesh in ICEM, then I have deleted the volume mesh. After that, surface mesh of FSI interface on fluid side is extruded in order to get solid wall volume mesh. I have tested both mesh with OpenFOAM FSI solver and it reports an almost zero interpolation error at the interface (1.2e-17)

 June 4, 2014, 10:58 #14 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Share some images of your problem, please. I'm not sure I understand your problem well. What does it mean "...I have deleted the volume mesh"? Do you want to simulate some hemodynamics problem? Please, depict your problem.

June 4, 2014, 11:55
#15
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Share some images of your problem, please. I'm not sure I understand your problem well. What does it mean "...I have deleted the volume mesh"? Do you want to simulate some hemodynamics problem? Please, depict your problem.
Yes, I am planning to perform a fluid-structure interaction in a vessel.

1) fluid mesh is generated in ICEM (att, fluid)

2) surface mesh which is the same on both sides (fluid & solid). (att, surfaceMesh)

3) solid mesh is the generated by extruding the surface mesh by an specific thickness.

However, it seems that system coupling is unable to read that and mapping summery reports 0 percent mapped.
Attached Images
 fluid.jpg (96.6 KB, 43 views) surfaceMesh.jpg (82.0 KB, 37 views) solid.jpg (96.2 KB, 39 views)

 June 5, 2014, 11:01 #16 New Member   NeNaD Join Date: Mar 2010 Posts: 16 Rep Power: 16 Check which side of the wall you selected to be Fluid Solid Interface in Mechanical (inner or outer wall of the vessel). I saw hemodynamics simulation with system coupling. It works.

June 14, 2014, 06:18
#17
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by skinnyfluid Check which side of the wall you selected to be Fluid Solid Interface in Mechanical (inner or outer wall of the vessel). I saw hemodynamics simulation with system coupling. It works.
I think I have found the problem, however I don't have any alternatives. You can see the project schematic in the att. I have tried with much simpler problem which I have tested with the same method of system coupling tutorial and it worked. However, after using my current project schematic on it, I get 0 percent area mapped here too.

 June 18, 2014, 02:46 #18 Senior Member   Matthias Voß Join Date: Mar 2009 Location: Berlin, Germany Posts: 449 Rep Power: 20 Did you try to break and resetup the very last connection? Maybe also delete the *.ds file from ansys (which is basically the same as the first). Daniel_Khazaei likes this.

June 18, 2014, 06:45
#19
Senior Member

Daniel
Join Date: Mar 2013
Location: Noshahr, Iran
Posts: 348
Rep Power: 21
Quote:
 Originally Posted by mvoss Did you try to break and resetup the very last connection? Maybe also delete the *.ds file from ansys (which is basically the same as the first).
Thank you. I will try to test your suggestion.
However I did not understand the second part of your suggestion regarding deleting the *.ds file from ansys.

--------

edit:

I have tried the first part of your suggestion with no luck. I still get 0 percent area mapped.

Last edited by Daniel_Khazaei; June 18, 2014 at 07:46.

 June 30, 2014, 10:02 #20 Senior Member   Daniel Join Date: Mar 2013 Location: Noshahr, Iran Posts: 348 Rep Power: 21 any help? Best wishes