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.

Estimation and control of the volume change with integration

Please login with a confirmed email address before reporting spam

Hi everyone. I am trying to calculate the volume of a sphere. I used Integration in Definition > nonlocal coupling>integration. Then, in derived values I try both Volume intergration and Surface integration. In volume integration, I got an unit = m^6, so I think I made something wrong along the way. In surface Integration, I though it was important to consider the fully geometry in space (I made a 2Axisymmetric model), so I typed 2pirad*intop1(1). Also, in this case the unit is wrong. (rad is the radius of the sphere, I attached below my model).

Can somebody gime me some feedback, or knows if there are another ways to make an estimation of the volume, and eventually, how to control the volume change of the object?

Thanks to all who can help me to understand better. Chiara



7 Replies Last Post 23 juin 2020, 08:46 UTC−4
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 12 juin 2020, 15:40 UTC−4
Updated: 4 years ago 12 juin 2020, 15:49 UTC−4

Hello Chiara,

1/ The integration coupling operator already performs an integration. You don't need to further integrate in Derived Values, instead in Derived Values, do a Global Evaluation of intop1(1). The "perform integral in revolved geometry" checkbox takes care of the 2pir term in the integral, BTW.

2/ If it's a sphere you want, in 2D axisymmetry you need to draw a half-circle with its straight side on the r=0 axis. What you drew corresponds to a torus of zero inner radius.

To evaluate the change in volume, you'll want to integrate the trace of the strain tensor, assuming you're operating under small strains assumptions.

Best,

Jeff

-------------------
Jeff Hiller
Hello Chiara, 1/ The integration coupling operator already performs an integration. You don't need to further integrate in Derived Values, instead in Derived Values, do a Global Evaluation of intop1(1). The "perform integral in revolved geometry" checkbox takes care of the 2*pi*r term in the integral, BTW. 2/ If it's a sphere you want, in 2D axisymmetry you need to draw a half-circle with its straight side on the r=0 axis. What you drew corresponds to a torus of zero inner radius. To evaluate the change in volume, you'll want to integrate the trace of the strain tensor, assuming you're operating under small strains assumptions. Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 15 juin 2020, 13:04 UTC−4

Hi Jeff, thanks for your answers. I, previously, tried the Global Evaluation but I didn't get the right volume so I thought that to be wrong. But, I was not considering to build the circle with its straight side on the r=0 axis. Now, I got the right volume value. Thank you.

To evaluate the change in volume, you'll want to integrate the trace of the strain tensor, assuming you're operating under small strains assumptions.

As for the part about the evaluation of the volume change, I don't understand exactly what to do. Is there some documentation that I can refer to??

Chiara

Hi Jeff, thanks for your answers. I, previously, tried the Global Evaluation but I didn't get the right volume so I thought that to be wrong. But, I was not considering to build the circle with its straight side on the r=0 axis. Now, I got the right volume value. Thank you. >To evaluate the change in volume, you'll want to integrate the trace of the strain tensor, assuming you're operating under small strains assumptions. As for the part about the evaluation of the volume change, I don't understand exactly what to do. Is there some documentation that I can refer to?? Chiara

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 15 juin 2020, 13:46 UTC−4

Hello Chiara,

See the attached model. I set it up so that the sphere is in a state of pure pressure and therefore shrinks into a smaller sphere. You can see from one of the plots that its diameter is reduced by ~5.45e-6 meters. The global evaluation node computes the change in volume as ~-3.427e-5 cubic meters. You can check that that value matches the analytical value for the change in volume of a sphere going from a diameter of 2 meters to a diameter of 2-5.45e-6 meters.

Best regrds,

Jeff

-------------------
Jeff Hiller
Hello Chiara, See the attached model. I set it up so that the sphere is in a state of pure pressure and therefore shrinks into a smaller sphere. You can see from one of the plots that its diameter is reduced by ~5.45e-6 meters. The global evaluation node computes the change in volume as ~-3.427e-5 cubic meters. You can check that that value matches the analytical value for the change in volume of a sphere going from a diameter of 2 meters to a diameter of 2-5.45e-6 meters. Best regrds, Jeff


Please login with a confirmed email address before reporting spam

Posted: 4 years ago 15 juin 2020, 17:11 UTC−4

Hi Jeff,

you have been very kind to attach the model. I really appreciate the help. Thank you!!

Best, Chiara

Hi Jeff, you have been very kind to attach the model. I really appreciate the help. Thank you!! Best, Chiara

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 19 juin 2020, 09:58 UTC−4

Hi Jeff, sorry to bother you again but I was wondering about to build the model in the reverse way. I will explain myself better. Is it possible to impose directly a volume variation on the sphere without using the pressure? 1) For example, I have the initial volume, V1, and a final volume, V2 and I want that my sphere (or some other geometry) moves from V1 to V2. I was thinking something about using the interpolation function with the two points: (V1,t1) and (V2,t2), but I am not sure how to impose it on the domains or the boundaries of the sphere. t1 and t2 are for the time. 2) Is there the chance to do it without using a time dependent condition? Thanks in advance.

Best, Chiara

Hi Jeff, sorry to bother you again but I was wondering about to build the model in the reverse way. I will explain myself better. Is it possible to impose directly a volume variation on the sphere without using the pressure? 1) For example, I have the initial volume, V1, and a final volume, V2 and I want that my sphere (or some other geometry) moves from V1 to V2. I was thinking something about using the interpolation function with the two points: (V1,t1) and (V2,t2), but I am not sure how to impose it on the domains or the boundaries of the sphere. t1 and t2 are for the time. 2) Is there the chance to do it without using a time dependent condition? Thanks in advance. Best, Chiara

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 19 juin 2020, 12:05 UTC−4
Updated: 4 years ago 19 juin 2020, 14:05 UTC−4

I don't understand what you are trying to achieve. There are many different deformation fields that would result in an object going from its initial volume V1 to a pre-selected other volume V2. If you know the entire deformation field, not just the overall volume change, then read into the use of moving geometry and moving mesh interfaces, this may be what you are after.

Best,

Jeff

-------------------
Jeff Hiller
I don't understand what you are trying to achieve. There are many different deformation fields that would result in an object going from its initial volume V1 to a pre-selected other volume V2. If you know the entire deformation field, not just the overall volume change, then read into the use of moving geometry and moving mesh interfaces, this may be what you are after. Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 4 years ago 23 juin 2020, 08:46 UTC−4

I don't understand what you are trying to achieve. There are many different deformation fields that would result in an object going from its initial volume V1 to a pre-selected other volume V2. If you know the entire deformation field, not just the overall volume change, then read into the use of moving geometry and moving mesh interfaces, this may be what you are after.

Best,

Jeff

Hi Jeff, thanks for your answer. At the present moment, I was wondering if it possible only to apply a volume change on an object (and not to apply the pressure to see the change in the original shape) and, eventually to control this change thorough time. I think I have only the overall volume change, but thanks for the suggestion about the moving mesh and geometry. I will check on that.

Best, Chiara

>I don't understand what you are trying to achieve. There are many different deformation fields that would result in an object going from its initial volume V1 to a pre-selected other volume V2. If you know the entire deformation field, not just the overall volume change, then read into the use of moving geometry and moving mesh interfaces, this may be what you are after. > >Best, > >Jeff Hi Jeff, thanks for your answer. At the present moment, I was wondering if it possible only to apply a volume change on an object (and not to apply the pressure to see the change in the original shape) and, eventually to control this change thorough time. I think I have only the overall volume change, but thanks for the suggestion about the moving mesh and geometry. I will check on that. Best, Chiara

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.