Long Term analysis

Pled · February 1, 2018, 11:43

How can I make a sim that considers time in long term?
I need to define boundaries related to time as well.

ghorrocks · February 1, 2018, 17:57

Just do a transient analysis with a long total time. Make sure you have a good computer, robust backups and a reliable network/computer before attempting it.

Time is a variable which is available in CEL expressions, so you can make boundary conditions functions of time. Have a look at the CFX tutorials on how to use CEL expressions and the CFX documentation (Reference manual) on available CEL variables and functions.

Pled · February 2, 2018, 05:15

I have considered that, but I need to consider a time step of miliseconds into boundaries
for a heart sim, and evaluate the results in a period of year.
I was thinking of something that could consider a periodicity in values of the first iterations and continue faster.

ghorrocks · February 2, 2018, 18:47

Sounds like you have a model with a wide separation of time scales. Usually this means that you can use CFD to solve the short fluid time scales, but some other approach (possibly a simple ODE solver) to do the long time scales.

If you explained what you are trying to do we might be able to help you.

Gert-Jan · February 3, 2018, 15:25

What questions are your trying to answer when performing the CFD-calculation of single cycle and the CFD-calculation of 1 year?

Pled · February 5, 2018, 04:41

What I am asked to do is to evaluate the plaque creation inside an artery of a heart.

To do that, plaque is expressed through the deformation of the artery wall mesh.
And the plaque creation, which is the mesh deformation, is a function of wall shear stresses.

I wanted to evalute the plaque progression in one, two and three years, considering the artery inlet as a sine function,but I understand this is not possible. So I am going to consider input as a fixed value, based on bibliography.

Even so how can I make a model that converges for a wide variety of artery geometries?

ghorrocks · February 5, 2018, 05:44

This is a classic case where you need to separate out the time scales.

On the time scale of a heart beat the plaque does not change. So simulate the heart beat assuming fixed plaque, but do a series of different plaque levels to simulate the plaque accumulating over time.

Pled · February 6, 2018, 07:47

I'll try that out with different numbers of plaque levels to evaluate the results

February 1, 2018, 11:43	Long Term analysis	#1
Pled Member Dimitrios S. P. Join Date: Jan 2018 Posts: 63 Rep Power: 8	How can I make a sim that considers time in long term? I need to define boundaries related to time as well.

February 1, 2018, 17:57		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Just do a transient analysis with a long total time. Make sure you have a good computer, robust backups and a reliable network/computer before attempting it. Time is a variable which is available in CEL expressions, so you can make boundary conditions functions of time. Have a look at the CFX tutorials on how to use CEL expressions and the CFX documentation (Reference manual) on available CEL variables and functions. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

February 2, 2018, 18:47		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Sounds like you have a model with a wide separation of time scales. Usually this means that you can use CFD to solve the short fluid time scales, but some other approach (possibly a simple ODE solver) to do the long time scales. If you explained what you are trying to do we might be able to help you. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

February 5, 2018, 05:44		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	This is a classic case where you need to separate out the time scales. On the time scale of a heart beat the plaque does not change. So simulate the heart beat assuming fixed plaque, but do a series of different plaque levels to simulate the plaque accumulating over time. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem of SOURCE term gradient in UDS	wind	Fluent UDF and Scheme Programming	6	December 1, 2022 15:21
Flutter analysis of compressor blade	sridharabc228	CFX	1	April 13, 2017 20:10
FSI-system coupling: "...same analysis type"?	rjw	CFX	0	March 26, 2017 08:25
How does fluent discretize the viscous term?	motorbean	FLUENT	3	August 25, 2015 17:38
Normal Stress Term in properties of Standard K-Epsilon model	MissCFD	STAR-CCM+	0	April 8, 2015 16:37

February 2, 2018, 05:15		#3
Pled Member Dimitrios S. P. Join Date: Jan 2018 Posts: 63 Rep Power: 8	I have considered that, but I need to consider a time step of miliseconds into boundaries for a heart sim, and evaluate the results in a period of year. I was thinking of something that could consider a periodicity in values of the first iterations and continue faster.

February 3, 2018, 15:25		#5
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,928 Rep Power: 28	What questions are your trying to answer when performing the CFD-calculation of single cycle and the CFD-calculation of 1 year?

February 5, 2018, 04:41		#6
Pled Member Dimitrios S. P. Join Date: Jan 2018 Posts: 63 Rep Power: 8	What I am asked to do is to evaluate the plaque creation inside an artery of a heart. To do that, plaque is expressed through the deformation of the artery wall mesh. And the plaque creation, which is the mesh deformation, is a function of wall shear stresses. I wanted to evalute the plaque progression in one, two and three years, considering the artery inlet as a sine function,but I understand this is not possible. So I am going to consider input as a fixed value, based on bibliography. Even so how can I make a model that converges for a wide variety of artery geometries?

February 6, 2018, 07:47		#8
Pled Member Dimitrios S. P. Join Date: Jan 2018 Posts: 63 Rep Power: 8	I'll try that out with different numbers of plaque levels to evaluate the results