Modelling the movement of an artery

ashtonJ · September 25, 2014, 01:01

Dear all,
I am trying to model the movement of an artery. I have the geometry file of the artery every 0.1[s], therefore in 1[s], I have 11 geometry files.
Geometry 1 @ t=0
Geometry 2 @ t=0.1
.
.
.
Geometry 11 @ t=1
I was just wondering how it is possible to model the movement of the artery such that geometry of artery changes from Geometry 1 to Geometry 11 in one cycle.

Thank you.
AshtonJ

ghorrocks · September 25, 2014, 07:19

Do you want it to interpolate a smooth motion between the files you have? You will probably need this to get a smooth motion.

As long as your motion is not too massive then you can mesh them all with a mesh which is topologically the same and just move the nodes to the new geometry. Then you can do a mesh motion simulation which uses the series of mesh files. You will need user fortran to do this. There used to be a CFX example which did this but it has been removed in recent versions. If you asked ANSYS support they hopefully will still have access to the tutorial example.

ashtonJ · September 25, 2014, 23:49

Thanks very much Glenn.
I contacted ANSYS support regarding that tutorial, but unfortunately they were not aware of that. Would you remember the name of the tutorial?

ghorrocks · September 26, 2014, 06:01

It was the ball valve example from when moving mesh was new. So it was probably from about 8 years ago (!)

ashtonJ · September 26, 2014, 22:23

Thanks Glenn. In ball valve case, the topology does not change – i.e. the ball shape is not changing. However, in my case, the geometry of artery is changing. Do you think it is feasible to model that??

Lesson · September 30, 2014, 06:18

Dear AshtonJ

Every kind of motion that you can define with a function in cartesian coordinate system or cylindrical coordinate system can be done with CEL in CFX.
For example, you can define expansion of a lateral wall of a cylinder.
This is possible if you have not large deformations otherwise you'll have bad mesh.

I read that you have 11 geometries, you can load mesh of this geometries during the simulation ( interpolation of simulation in the new mesh is trilinear ).
Pay attention that you can not do brute change of the geometry.

bye
Fely

September 25, 2014, 01:01	Modelling the movement of an artery	#1
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 17	Dear all, I am trying to model the movement of an artery. I have the geometry file of the artery every 0.1[s], therefore in 1[s], I have 11 geometry files. Geometry 1 @ t=0 Geometry 2 @ t=0.1 . . . Geometry 11 @ t=1 I was just wondering how it is possible to model the movement of the artery such that geometry of artery changes from Geometry 1 to Geometry 11 in one cycle. Thank you. AshtonJ

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Modelling an artery	izzy13	FLUENT	1	September 14, 2017 10:36
error message	cuteapathy	CFX	14	March 20, 2012 07:45
modelling wind movement	charltonjames	STAR-CCM+	8	May 10, 2010 03:52
Modelling the movement of a liquid surface	C. Lucero	FLUENT	4	March 11, 2008 14:56
Modelling Movement of Trawler Nets in water	Chibisi	FLUENT	0	March 29, 2006 10:46

September 25, 2014, 07:19		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	Do you want it to interpolate a smooth motion between the files you have? You will probably need this to get a smooth motion. As long as your motion is not too massive then you can mesh them all with a mesh which is topologically the same and just move the nodes to the new geometry. Then you can do a mesh motion simulation which uses the series of mesh files. You will need user fortran to do this. There used to be a CFX example which did this but it has been removed in recent versions. If you asked ANSYS support they hopefully will still have access to the tutorial example.

September 25, 2014, 23:49		#3
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 17	Thanks very much Glenn. I contacted ANSYS support regarding that tutorial, but unfortunately they were not aware of that. Would you remember the name of the tutorial?

September 26, 2014, 06:01		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	It was the ball valve example from when moving mesh was new. So it was probably from about 8 years ago (!)

September 26, 2014, 22:23		#5
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 17	Thanks Glenn. In ball valve case, the topology does not change – i.e. the ball shape is not changing. However, in my case, the geometry of artery is changing. Do you think it is feasible to model that??

September 30, 2014, 06:18		#6
Lesson New Member Felix Join Date: Dec 2012 Posts: 7 Rep Power: 14	Dear AshtonJ Every kind of motion that you can define with a function in cartesian coordinate system or cylindrical coordinate system can be done with CEL in CFX. For example, you can define expansion of a lateral wall of a cylinder. This is possible if you have not large deformations otherwise you'll have bad mesh. I read that you have 11 geometries, you can load mesh of this geometries during the simulation ( interpolation of simulation in the new mesh is trilinear ). Pay attention that you can not do brute change of the geometry. bye Fely