Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Prescribed Displacement or Fixed Constraint

Please login with a confirmed email address before reporting spam

Hello, I'm using Comsol Multiphysics for my master thesis to realize a breast model. Now I'm studying the effects of the gravity load on the breast tissues, but I have some problems to reproduce the same results which are reported in my reference (an article that I found in literature). I've created a hemisperical model of four concentric tissue layers with uniform thickness; then I chose my boundary condition, by imposing a fixed contraint on the chest wall. (the article says that "An additional boundary condition was assumed that no displacement occurs on the chest wall interface"). However, i didn't obtain the same results.

Could this be related to the boundary condition? Is there any difference between "fixed contraint" and "prescribed displacement (where I impose displacement equal to zero in all directions)?

Thanking you in advance, Maria


16 Replies Last Post 17 déc. 2018, 15:16 UTC−5
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 14 déc. 2018, 13:59 UTC−5

No, there is no difference. A fixed constraint is the same as prescribing all displacements to be zero.

Jeff

-------------------
Jeff Hiller
No, there is no difference. A fixed constraint is the same as prescribing all displacements to be zero. Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 15 déc. 2018, 01:48 UTC−5

Thank you so much Jeff! So, which kind of boundary condition do you suggest me in order to replicate what the article says?

Thank you so much Jeff! So, which kind of boundary condition do you suggest me in order to replicate what the article says?

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 15 déc. 2018, 07:14 UTC−5

It does not sound like the discrepancy between your results and the reference are due to that boundary condition.

Jeff

-------------------
Jeff Hiller
It does not sound like the discrepancy between your results and the reference are due to that boundary condition. Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 15 déc. 2018, 09:05 UTC−5

Yes, I know. The problem is that the only boundary condition is the one that I already said, so I don't understand why my results are different. I've created my geometry according to the model represented in the reference, then I've applied the gravity load, but the problem is that in my work the total displacement increases with the Poisson's ratio (I've used a Parametric study, as in literature), while the correct result shows an opposite trend, so the displacement decreases when the Poisson's reation increases. I'm not able to understand which is the error. Could you help me?

M.

Yes, I know. The problem is that the only boundary condition is the one that I already said, so I don't understand why my results are different. I've created my geometry according to the model represented in the reference, then I've applied the gravity load, but the problem is that in my work the total displacement increases with the Poisson's ratio (I've used a Parametric study, as in literature), while the correct result shows an opposite trend, so the displacement decreases when the Poisson's reation increases. I'm not able to understand which is the error. Could you help me? M.

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 08:35 UTC−5

Hi Maria,

Nothing jumps at me in your description as being the source of the discrepancy, and I am sure that you have triple-checked your model set up against the reference already. If it was my model, I would take one more look at the material assignments (I am presuming that you have four materials, one per layer, and there's always the chance that you set them up differently than the authors of the original study), to make sure they replicate the paper exactly, and if it all checks out I would contact the authors of the reference paper to see if they could discuss any assumptions they may have made and not documented in the paper (Are they for instance assuming large deformations while you're assuming small deformations? ot the other way round? or something else altogether).

Best,

Jeff

-------------------
Jeff Hiller
Hi Maria, Nothing jumps at me in your description as being the source of the discrepancy, and I am sure that you have triple-checked your model set up against the reference already. If it was my model, I would take one more look at the material assignments (I am presuming that you have four materials, one per layer, and there's always the chance that you set them up differently than the authors of the original study), to make sure they replicate the paper exactly, and if it all checks out I would contact the authors of the reference paper to see if they could discuss any assumptions they may have made and not documented in the paper (Are they for instance assuming large deformations while you're assuming small deformations? ot the other way round? or something else altogether). Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 08:44 UTC−5

Thank you again Jeff. Unfortunately in my reference there are not informations about the material assignments: I have three different types of tissue (skin, fat and fibroglandular tissue) whose properties are not explained. Only the range of Young modulus is reported, so for the adipose tissue a young modulus between 0.5 e 25 kPa, for the fibroglandular tissue the range 2-45 kPa, and for the skin a Young modulus of 20 kPa. Furthermore the Poisson ration is set 0.485; nevertheless, we've no informations about density (for this reason I've used the values which I found in literature). Maybe this can be the error, but I don't know how to solve this. I've written to the author and I hope he will answer me as soon as possible.

Thank you so much for your patience, Maria

Thank you again Jeff. Unfortunately in my reference there are not informations about the material assignments: I have three different types of tissue (skin, fat and fibroglandular tissue) whose properties are not explained. Only the range of Young modulus is reported, so for the adipose tissue a young modulus between 0.5 e 25 kPa, for the fibroglandular tissue the range 2-45 kPa, and for the skin a Young modulus of 20 kPa. Furthermore the Poisson ration is set 0.485; nevertheless, we've no informations about density (for this reason I've used the values which I found in literature). Maybe this can be the error, but I don't know how to solve this. I've written to the author and I hope he will answer me as soon as possible. Thank you so much for your patience, Maria

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 09:06 UTC−5

Given that the paper is not telling you exactly what values they are using for the Young's moduli, it is quite possible that there's absolutely nothing wrong with your model. It may well be that you are just getting different trends with regards to the Poisson ratio because you are using a different combination of Young's moduli.

Aside from that, since you are mentioning a Poisson ratio of 0.485 (i.e. a near incompressible material), if you have not done so already you can turn on the mixed u-p formulation (found on the "Linear Elastic Material" node in your model). See the User's Guide for the Structural Mechanics Module, version 5.4, page 296, and this older thread .

Jeff

-------------------
Jeff Hiller
Given that the paper is not telling you exactly what values they are using for the Young's moduli, it is quite possible that there's absolutely nothing wrong with your model. It may well be that you are just getting different trends with regards to the Poisson ratio because you are using a different combination of Young's moduli. Aside from that, since you are mentioning a Poisson ratio of 0.485 (i.e. a near incompressible material), if you have not done so already you can turn on the mixed u-p formulation (found on the "Linear Elastic Material" node in your model). See the User's Guide for the Structural Mechanics Module, version 5.4, page 296, and [this older thread](https://www.comsol.com/forum/thread/10958/poisson-ratio-v-0-5?last=2010-11-06T06:20:56Z) . Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 09:25 UTC−5

Yes, i'm already done it. Furthermore, I tried to use a Parametric study by varying the Young's moduli in the range which are explained in the reference, and considering all the possible combinations, but my results remain different. I don't know how to solve this problem; now i think that the only solution is to have information from the authors-

Yes, i'm already done it. Furthermore, I tried to use a Parametric study by varying the Young's moduli in the range which are explained in the reference, and considering all the possible combinations, but my results remain different. I don't know how to solve this problem; now i think that the only solution is to have information from the authors-

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 09:36 UTC−5

It makes sense. It's certainly possible that it's the reference that's wrong...

Jeff

-------------------
Jeff Hiller
It makes sense. It's certainly possible that it's the reference that's wrong... Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 12:01 UTC−5

Sorry Jef, I try to ask you another question. I don't understand why, after having chosen material properties and computing my study, I obtain a total displacement equal to 49 mm, while when I go on the "Results" node and I try to plot the displacement field in the z direction I obtain a value around 5*10^-8. Considering that I'm applying a gravity load, the most significant deformations occur in the inferior-superior direction. So how is it possible to obtain these values?

Sorry Jef, I try to ask you another question. I don't understand why, after having chosen material properties and computing my study, I obtain a total displacement equal to 49 mm, while when I go on the "Results" node and I try to plot the displacement field in the z direction I obtain a value around 5*10^-8. Considering that I'm applying a gravity load, the most significant deformations occur in the inferior-superior direction. So how is it possible to obtain these values?

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 13:17 UTC−5

It sounds like your displacements are mostly in the x and y directions.

-------------------
Jeff Hiller
It sounds like your displacements are mostly in the x and y directions.

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 13:21 UTC−5

Yes, it seems. But this is not possible, of course. I notice that if I right-click the 3D Plot Group 3 node and select More Plots > Max/Min Volume I obtain a correct value . So now, Which is the different between this method and the one that I used before (click the Derived Values and select Maximum > Maximum Volume)?

Yes, it seems. But this is not possible, of course. I notice that if I right-click the 3D Plot Group 3 node and select More Plots > Max/Min Volume I obtain a correct value . So now, Which is the different between this method and the one that I used before (click the Derived Values and select Maximum > Maximum Volume)?

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 13:30 UTC−5
Updated: 6 years ago 17 déc. 2018, 16:13 UTC−5

Please upload your mph file.

-------------------
Jeff Hiller
Please upload your mph file.

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 14:05 UTC−5

Yes, of course. Thank you in advance.

Yes, of course. Thank you in advance.


Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 14:27 UTC−5
Updated: 6 years ago 17 déc. 2018, 15:00 UTC−5

The issue is one of sign. Your z displacements are mostly negative. They have a maximum absolute value of ~49mm but a maximum value of ~0. So, under Derived Values, when using w as the expression, you need to do a Volume Minimum, not a Volume Maximum, if your goal is to capture where the z displacement has the greatest magnitude.

Jeff

-------------------
Jeff Hiller
The issue is one of sign. Your z displacements are mostly negative. They have a maximum absolute value of ~49mm but a maximum value of ~0. So, under Derived Values, when using w as the expression, you need to do a Volume Minimum, not a Volume Maximum, if your goal is to capture where the z displacement has the greatest magnitude. Jeff

Please login with a confirmed email address before reporting spam

Posted: 6 years ago 17 déc. 2018, 15:16 UTC−5

Thank you Jess, you're right. I didn't see that error,

Thank you Jess, you're right. I didn't see that error,

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.