CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

Wall Deformation and swelling of an artery

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

Reply
 
LinkBack Thread Tools Display Modes
Old   January 17, 2014, 09:54
Default Wall Deformation and swelling of an artery
  #1
New Member
 
Michael
Join Date: Jan 2014
Posts: 5
Rep Power: 3
MichiMichi is on a distinguished road
Hi all,

I am trying to incorporate a mechanism in Ansys CFX which can predict arteriosclerotic formation in an artery.
My idea is the following:
I have got a cylindrical shaped artery with a bifurcation. As the blood flows through this bifurcation artery several regions of the wall (which are imposed to a threshold value of Wall Shear Stress)become inflamed and start to swell. This results in a formation of a bump (solid material) which reduces the cross sectional area of the flow.

You can see the explanation in the picture below.

I already tried to increase the viscosity of the blood to extreme high values (from 3.6e-3 Pa s to 1000 Pa s), but this leads to very high pressures of the high viscous parts of my fluid. I want this parts to "rest". I looked through the multi phase tutorials but couldn't find anything which really fits my problem. Maybe it is possible to some kind deform the surface mesh of the artery...

In general it should be a solidification process, I've read that this is possible to implement in FLUENT.
Do you have some ideas how to do this in Ansys CFX? Or maybe some keywords to look at.

Thank you very much!
Michael
Attached Images
File Type: jpg aretriosclerosis.jpg (14.9 KB, 22 views)
MichiMichi is offline   Reply With Quote

Old   January 18, 2014, 05:08
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,825
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
You can probably do this in CFX either with a moving mesh or an immersed solid approach. Either should work. You could probably also use a momentum source to drive the flow in the restricted region to zero (which is effectively what the immersed solid approach does).

Generating massive variations in viscosity over short distances is bound to lead to convergence problems. From the little I know of this flow I do not think viscosity is equivalent to what is physically happening - so of course the solver has problems making it converge.
ghorrocks is offline   Reply With Quote

Old   March 14, 2014, 03:58
Default
  #3
Senior Member
 
Ashkan Javadzadegan
Join Date: Sep 2010
Posts: 234
Rep Power: 7
ashtonJ is on a distinguished road
Dear all,
I have the same problem, I have a pipe that fluid is passing through unimpeded, the pipe then narrows to form the constriction shown in the attached image .
Is it possible to model this using moving mesh approach? If yes, please let me know how it is doable.

Thank you.
AshtonJ
ashtonJ is offline   Reply With Quote

Old   March 14, 2014, 03:58
Default
  #4
Senior Member
 
Ashkan Javadzadegan
Join Date: Sep 2010
Posts: 234
Rep Power: 7
ashtonJ is on a distinguished road
This is the image link
http://tinypic.com/r/25s4rhs/8
ashtonJ is offline   Reply With Quote

Old   March 14, 2014, 04:53
Default
  #5
Senior Member
 
Matthias Voß
Join Date: Mar 2009
Location: Berlin, Germany
Posts: 446
Rep Power: 11
mvoss is on a distinguished road
You can define expressions in CFX on how far to move the mesh during a simulation by either applying a total deformation or the deformation per TimeStep/Iteration.
Create an expression where you feed in x,y,z-coordinates and then simply "return" the desired end position/total mesh deformation in [m].

Code:
TotalMeshDeformation Y= RampeY 
RampLength=1cm
RampeY = (-2*(min((x)/RampLength,1))^3+3*(min((x)/RampLength,1))^2)
Check the above expression with the ploting feature in CFX-Pre. Make sure to get the desired behavior for the mesh movement (e.g. MeshMovement for Y ramped wrt. x-coord).

Just an idea: If you have a tube - you can also convert the movement to radius/theta/z by referencing the whole simulation on a new coordinate system.
mvoss is offline   Reply With Quote

Old   March 14, 2014, 06:25
Default
  #6
Senior Member
 
Ashkan Javadzadegan
Join Date: Sep 2010
Posts: 234
Rep Power: 7
ashtonJ is on a distinguished road
Dear mvoss,

Thank you. I tried but the following error was given;
At least one highly skewed element has been detected on a wall
boundary, leading to an unreliable near-wall distance calculation
for the turbulent wall functions. The solver will continue
to execute, but convergence and/or accuracy may be affected.
Please consider improving the mesh quality.


Actually, I do not understand the expression you defined, I need to create a narrowing only in the middle section of the pipe. Where in your expression, this is included. I would be too pleased if you explain a bit more moving mesh modelling, especially for my case.

I've done FSI analysis before, but this is first time I do moving mesh stuff. More details about moving mesh settings in CFX will help me a lot.

Thank you.
Regards,
ashtonJ
ashtonJ is offline   Reply With Quote

Old   March 14, 2014, 06:32
Default
  #7
Senior Member
 
Matthias Voß
Join Date: Mar 2009
Location: Berlin, Germany
Posts: 446
Rep Power: 11
mvoss is on a distinguished road
Did you check the expression in the plot window? There you can see pretty good what the expression is doing. (It is a smooth cubic curve changing from 0 to 1 over the desired RampLength) - the shape of the rampe is what the tube will look like. You really need to understand whats going on there.
The expression from above is moving the elements outwards and is just an example - you need to define what fits your task the best.
How come you did FSI without dealing with moving meshes - One-Way-FSI?
Did you do the tutorials on the moving mesh feature? If not, then i would highly recommend it.
What ANSYS version are you on?
mvoss is offline   Reply With Quote

Old   March 14, 2014, 07:40
Default
  #8
Senior Member
 
Ashkan Javadzadegan
Join Date: Sep 2010
Posts: 234
Rep Power: 7
ashtonJ is on a distinguished road
Yes, I did check the expression shape. Now I guess I understand what the expression is doing, the x in the expression should represent the longitudinal axis, in my case, it should be changed to z.
I've done two-way FSI in elastic arteries, which were deformed due to the pressure wave. In the FSI modellings, I had no control on the mesh displacement.
As suggested, it's better to go through the moving mesh tutorials.
Thank you very much for your help.
ashtonJ is offline   Reply With Quote

Old   March 14, 2014, 07:49
Default
  #9
Senior Member
 
Ashkan Javadzadegan
Join Date: Sep 2010
Posts: 234
Rep Power: 7
ashtonJ is on a distinguished road
Sorry, one more question. In two-dimensional case, the x in your expression shows the longitudinal axis and rampe represents the radius of the tube. Am I correct? If yes, how come the rampe start from zero? Does this mean that the radius at that point is zero?
ashtonJ is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On



All times are GMT -4. The time now is 03:35.