Applying an External Displacement in FSI Modelling

ashtonJ · August 23, 2012, 20:48

Dear all,

I am trying to model the motion of the coronary arteries while the heart is beating. During 1 second, I measured the average displacement of the artery, for example, at

t = 0: displacement=2mm
t= 0.1: displacement=2.4 mm and so on.

Now, I am trying to apply the above-mentioned time dependant displacement in the FSI modelling.

I would be too pleased if anybody can help me.

Kind regards,
AshtonJ

ghorrocks · August 23, 2012, 22:53

If you already know the displacement then it is not FSI. You can model it simply with moving mesh and this can be done entirely in CFX.

ashtonJ · August 23, 2012, 23:20

Thanks for your help.

I am trying to combine the measured displacement with displacement resulted from FSI modelling. As you know, the first displacement is due to the heart beating and the second one is due to the flow pulsatility, so, I need to see the effect of both motions simultaneously.

Kind regards,
AshtonJ

ghorrocks · August 24, 2012, 07:37

How do you propose to apply a measured displacement, but let it move? Then you aren't applying a measured displacement then, are you?

ashtonJ · August 27, 2012, 04:40

I am trying to combine to different motions in coronary arteries, one due to the pressure waveform and the other due to the heart contraction. I already modelled the motion due to the pressure waveform using FSI modelling. Using some experiments, I also measured the displacement of the artery due to the heart contraction. Now, I am trying to include the experimentally measured displacement in the FSI model that I already set up for the motion due to the pressure waveform. I am not sure whether it is feasible or not.

ghorrocks · August 27, 2012, 06:22

If you apply a displacement to the boundary to model the pressure waveform deflection then you have specified the position and it will not move further for other loading. Hence my question in post #4. There is no straight forward way of doing this.

You are going to have to think up how the physics of what you are tryign to do will work and then try to get the simulation to model it. Until you know the physics you cannot model it.

singer1812 · August 27, 2012, 09:59

You need to apply the fixed displacement on the ANSYS side not the CFX side. Once you set the displacement, you can remove it and the FSI simulation will commence with calcualted displacements.

ashtonJ · September 18, 2012, 04:25

Thanks Singer. Could you please clarify what you meant in your last post. What do you mean by remove the displacement.

Regards,
AshtonJ

singer1812 · September 18, 2012, 09:20

When you are talking FSI, are you talking a fluid model coupled with a structural model? Or are you just modeling the fluid flow due to wall movement? If it is the latter, that is not an FSI, just a simple moving mesh problem. Use CEL to prescribe wall motion.

Or are you actually modeling the wall motion, due to fluid and heart contraction? For example, your have the heart wall modeled with some material properties. In relaxed state there is no stress in the heart walls, the muscle contracts and moves to a set positions defined by expereriment, and induces stress in the heart walls. You remove the displacement and let the heart wall move back in according to fluids and wall properties. This is FSI.

ashtonJ · September 19, 2012, 02:24

Quote:

Originally Posted by singer1812

When you are talking FSI, are you talking a fluid model coupled with a structural model? Or are you just modeling the fluid flow due to wall movement? If it is the latter, that is not an FSI, just a simple moving mesh problem. Use CEL to prescribe wall motion.

Or are you actually modeling the wall motion, due to fluid and heart contraction? For example, your have the heart wall modeled with some material properties. In relaxed state there is no stress in the heart walls, the muscle contracts and moves to a set positions defined by expereriment, and induces stress in the heart walls. You remove the displacement and let the heart wall move back in according to fluids and wall properties. This is FSI.

Thanks Singer for your kind comments. Please let me explain what I am going to do.
I have got geometry of a coronary artery and measured the pressure waveforms at inlet and outlet of that. What I did, I modelled the artery using the two way FSI and simulated the motion resulted from the pressure waveform which we call it pulsatility motion

As you know, the artery also experiences another motion due to the movement (contraction) of the heart which we call it cardiac motion. I measured the average displacement of the artery at the different times during the cardiac cycle; I mean I know the averaged displacement of the artery due to the cardiac motion. Now, I want to combine two aforementioned motions, one due to the pressure waveform and one due to the cardiac motion. I just want to know whether it is possible to include the measured artery displacement due to the cardiac motion into the FSI model, in this case, I can see the effects of both motions on flow hemodynamic simultaneously.

Kind regards,
AshtonJ

singer1812 · September 19, 2012, 09:13

Ok, so I take it that your displacement measurements of your test data are isolated to the cardiac portion only and do not include displacement portion due to the fluid motion?

If this is the case, you will need to cause the cardiac displacement in some other manner than specified displacements in the FEA portion of the model. If you specify the displacements, your fluid will no longer work to cause motion on the boundarys.

One way I can think of doing this is to run a separate FEA model of the artery, perhaps using a force BC, and adjust the force inputs until you get the same motion as your cardiac portion.

With the force BC knowledge, you can rerurn the model in FSI with the force BC and the fluid flowing to provide the motion due to the combined effects.

What it really comes down to is that you need to change the method to cause displacement due to the cardiac event (something other than prescribed displacement).

I must state, I do not have experience in the medical field, and perhaps this is not the common way they do these things, but it is a way I can think of that might do what you are attempting. Someone else with more relevant experience might provide a better solution.

ashtonJ · September 20, 2012, 20:50

Thanks again. It is really a good idea, however, I am not too familiar with structural analysis. I tried to measure the artery displacement by applying a force BC as you suggested. I checked out all possible results that I can get from FEA analysis, I can just measure the deformation which is quite different from the displacement!

Do you have any idea how to measure the artery displacement not the deformation.

Regards,
ashtonJ

singer1812 · September 21, 2012, 09:07

Not sure I follow. Unless I am missing something, or we are losing some meaning across the thread, those are the same.

August 23, 2012, 20:48	Applying an External Displacement in FSI Modelling	#1
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 16	Dear all, I am trying to model the motion of the coronary arteries while the heart is beating. During 1 second, I measured the average displacement of the artery, for example, at *t = 0: displacement=2mm* t= 0.1: displacement=2.4 mm and so on. Now, I am trying to apply the above-mentioned time dependant displacement in the FSI modelling. I would be too pleased if anybody can help me. Kind regards, AshtonJ

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CGNS Compiling	Diego	Main CFD Forum	17	December 21, 2014 01:40
Boundary Condition Enquiry in FSI Modelling	ashtonJ	CFX	0	April 15, 2012 00:28
Help with FSI modelling in arteries	ashtonJ	CFX	2	April 3, 2012 18:06
FSI - Specified Mesh Displacement	Vinzent	CFX	2	September 17, 2010 07:09

August 23, 2012, 22:53		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	If you already know the displacement then it is not FSI. You can model it simply with moving mesh and this can be done entirely in CFX.

August 23, 2012, 23:20		#3
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 16	Thanks for your help. I am trying to combine the measured displacement with displacement resulted from FSI modelling. As you know, the first displacement is due to the heart beating and the second one is due to the flow pulsatility, so, I need to see the effect of both motions simultaneously. Kind regards, AshtonJ

August 24, 2012, 07:37		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	How do you propose to apply a measured displacement, but let it move? Then you aren't applying a measured displacement then, are you?

August 27, 2012, 04:40		#5
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 16	I am trying to combine to different motions in coronary arteries, one due to the pressure waveform and the other due to the heart contraction. I already modelled the motion due to the pressure waveform using FSI modelling. Using some experiments, I also measured the displacement of the artery due to the heart contraction. Now, I am trying to include the experimentally measured displacement in the FSI model that I already set up for the motion due to the pressure waveform. I am not sure whether it is feasible or not.

August 27, 2012, 06:22		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	If you apply a displacement to the boundary to model the pressure waveform deflection then you have specified the position and it will not move further for other loading. Hence my question in post #4. There is no straight forward way of doing this. You are going to have to think up how the physics of what you are tryign to do will work and then try to get the simulation to model it. Until you know the physics you cannot model it.

August 27, 2012, 09:59		#7
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 20	You need to apply the fixed displacement on the ANSYS side not the CFX side. Once you set the displacement, you can remove it and the FSI simulation will commence with calcualted displacements.

September 18, 2012, 04:25		#8
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 16	Thanks Singer. Could you please clarify what you meant in your last post. What do you mean by remove the displacement. Regards, AshtonJ

September 18, 2012, 09:20		#9
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 20	When you are talking FSI, are you talking a fluid model coupled with a structural model? Or are you just modeling the fluid flow due to wall movement? If it is the latter, that is not an FSI, just a simple moving mesh problem. Use CEL to prescribe wall motion. Or are you actually modeling the wall motion, due to fluid and heart contraction? For example, your have the heart wall modeled with some material properties. In relaxed state there is no stress in the heart walls, the muscle contracts and moves to a set positions defined by expereriment, and induces stress in the heart walls. You remove the displacement and let the heart wall move back in according to fluids and wall properties. This is FSI.

September 19, 2012, 09:13		#11
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 20	Ok, so I take it that your displacement measurements of your test data are isolated to the cardiac portion only and do not include displacement portion due to the fluid motion? If this is the case, you will need to cause the cardiac displacement in some other manner than specified displacements in the FEA portion of the model. If you specify the displacements, your fluid will no longer work to cause motion on the boundarys. One way I can think of doing this is to run a separate FEA model of the artery, perhaps using a force BC, and adjust the force inputs until you get the same motion as your cardiac portion. With the force BC knowledge, you can rerurn the model in FSI with the force BC and the fluid flowing to provide the motion due to the combined effects. What it really comes down to is that you need to change the method to cause displacement due to the cardiac event (something other than prescribed displacement). I must state, I do not have experience in the medical field, and perhaps this is not the common way they do these things, but it is a way I can think of that might do what you are attempting. Someone else with more relevant experience might provide a better solution.

September 20, 2012, 20:50		#12
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 16	Thanks again. It is really a good idea, however, I am not too familiar with structural analysis. I tried to measure the artery displacement by applying a force BC as you suggested. I checked out all possible results that I can get from FEA analysis, I can just measure the deformation which is quite different from the displacement! Do you have any idea how to measure the artery displacement not the deformation. Regards, ashtonJ

September 21, 2012, 09:07		#13
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 20	Not sure I follow. Unless I am missing something, or we are losing some meaning across the thread, those are the same.