Volume of a Cavity

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • Sanaz
    Junior Member
    • Apr 2016
    • 14

    Volume of a Cavity

    Hi Forum Members

    I need to find the volume of a cavity (an empty space within a hollow cylinder) in each time step, to plot Pressure-Volume curve. Is it possible in Febio/Postview? Any suggestions?
    I had the same issue with Abaqus. I solved it by defining a cavity as a new part. Then I found the volume of that part in each step.

    Thank You
    Sanaz
    Attached Files
  • maas
    Lead Code Developer
    • Nov 2007
    • 3400

    #2
    Hi Sanaz,

    Unfortunately, I can't think of a way to do this with any of the current features in FEBio or PostView. However, if this something you really need, I'd be happy to take a closer look and add a new feature. Would you please be willing to make a feature request in the appropriate forum Project PostView so that I stays on my radar? Let us know if you would prefer the feature in FEBio or PostView (or both?). If you need this quickly, then PostView is probably the way to go.

    Thanks,

    Steve
    Department of Bioengineering, University of Utah
    Scientific Computing and Imaging institute, University of Utah

    Comment

    • Sanaz
      Junior Member
      • Apr 2016
      • 14

      #3
      Hi Steve

      I made a feature request in PostView.

      Thanks,
      Sanaz

      Comment

      • jonasj17
        Junior Member
        • May 2017
        • 6

        #4
        I'm interested in the same functionality. I've tried to calculate the inner volume by two metods in PostView but get conflicting results:

        1. PostView/Tools/Measure Volume by selecting the inner surface (faces). This must, however, be done for each time step and it seems to be impossible to generate a plot directly.
        2. Selecting the inner surface (faces) and intergrating over total displacement to see the change in volume

        I've been trying these alternatives in i hollow sphere of Mooney-Rivlin material inflated by an increasing internal surface pressure.

        By method 1. I get a volume of 520 ml at P=0 and 820 ml at P=max. This equals a difference of 300 ml. This must be correct since the radius increases from 50 mm to 58 mm.
        By method 2. I get a volume difference of 400 ml.

        Metod 2 is preferable since it is possible to get a plot of volume with time in PostView. I do not understand why the integration of total face displacement overestimates the volume change. Any ideas?

        A direct way to plot inner volume vs pressure would be even better if it could be implemented. Perhaps through a custom x-y plot, where other parameters than time could be chosen as x?

        Regards

        Jonas

        Comment

        • ateshian
          Developer
          • Dec 2007
          • 1821

          #5
          Hi Jonas,

          I implemented a plot variable in FEBio called "enclosed volume", which can be used with a named surface. It evaluates the volume enclosed by a surface at each time point and associates it with any/all of the faces of that surface.

          The named surface is declared in the <Geometry> section, using <Surface name="unique_name">. The plot variable is declared in the <plotfile> section as <var type="enclosed volume" surface="unique_name"/>. This way you can create scatter plots in PostView that use this variable along either of the plot axes. Multiple declarations of this variable may be used to evaluate multiple enclosed volumes.

          This feature will become available in the next release.

          Best,

          Gerard

          Comment

          • jonasj17
            Junior Member
            • May 2017
            • 6

            #6
            Excellent. Looking forward to the next release then. In the current release you can not access named selections as defined in preview, right? They do not seem to show up under Model/Mesh in PostView.

            Regards,

            Jonas

            Comment

            • ateshian
              Developer
              • Dec 2007
              • 1821

              #7
              Hi Jonas,

              That's correct. The latest version of PreView will automatically save named selections and provide the option to plot "enclosed volume" for each named surface in the model.

              Best,

              Gerard

              Comment

              • Fisher
                Member
                • Feb 2017
                • 38

                #8
                Has this feature been implemented yet? When I tried it in 2.6.4 I got the error 'output variable "enclosed volume" is not defined'.
                I have my own matlab method to manually calculate volume using node position data from the logfile, but I want to verify this against your own implementation because I am getting some inconsistent results when using the 'volume constraint' feature.

                Comment

                • ateshian
                  Developer
                  • Dec 2007
                  • 1821

                  #9
                  It will be released shortly in FEBio 2.7.

                  Best,

                  Gerard

                  Comment

                  • asixbabak
                    Member
                    • Jan 2015
                    • 39

                    #10
                    Originally posted by ateshian View Post
                    Hi Jonas,

                    I implemented a plot variable in FEBio called "enclosed volume", which can be used with a named surface. It evaluates the volume enclosed by a surface at each time point and associates it with any/all of the faces of that surface.

                    The named surface is declared in the <Geometry> section, using <Surface name="unique_name">. The plot variable is declared in the <plotfile> section as <var type="enclosed volume" surface="unique_name"/>. This way you can create scatter plots in PostView that use this variable along either of the plot axes. Multiple declarations of this variable may be used to evaluate multiple enclosed volumes.

                    This feature will become available in the next release.

                    Best,

                    Gerard
                    Hi Gerard,

                    I have a question about the enclosed volume plot variable. I am trying to get the "volume" enclosed by a portion of semi-spherical pie geometry (image attached; axis of symmetry Z), and I wonder if the enclosed volume would work with this kind of geometry. How does this method calculate the volume?


                    Best,
                    Babak
                    You do not have permission to view this gallery.
                    This gallery has 1 photos.

                    Comment

                    • ateshian
                      Developer
                      • Dec 2007
                      • 1821

                      #11
                      Hi Babak,

                      The "enclosed volume" plot variable is calculated from a surface integral using the divergence theorem:
                      EnclosedVolume.png
                      Here, V is the enclosed volume, ∂V is the surface boundary of V (the user-selected surface), x is the position of a point on the surface boundary and n is the unit normal at that point.

                      The boundary ∂V is determined entirely by the surface(s) selected by the user. If the surface is not closed, the calculation proceeds according to the above integration scheme. In that case, the location of the surface relative to the coordinate axes will affect the calculated enclosed volume. In your model, if the two corners of the spherical wedge fall along one of the coordinate axes, the enclosed volume will be the expected value of the spherical wedge volume. However, if those corner points are located elsewhere in 3D space, the calculation will not give you what you expect. In that case you would have to select a closed surface (implying that you would need to model additional surface elements to enclose the desired volume).

                      Best,

                      Gerard

                      Comment

                      • asixbabak
                        Member
                        • Jan 2015
                        • 39

                        #12
                        Perfect! It makes sense. Because on the surface of the wedge planes x is perpendicular to the surface normals the surface integrals are zero and the curved surface gives the non-zero value.
                        I get a negative volume value though, which is due to the inward normal of the curved surface.

                        Thanks Gerard!

                        Comment

                        Working...
                        X
                        😀
                        😂
                        🥰
                        😘
                        🤢
                        😎
                        😞
                        😡
                        👍
                        👎