Perimeters and areas of open holes

mbeauvaismbeauvais
I have a hyperelastic cylinder full of noncircular holes. The cylinder is being compressed along a single axis in the cross-sectional direction, and the cylinder and the holes inside it deform as the compression occurs.

Along with the surface area (which Mecway has a tool for), I would like to be able to find out the two dimensional area of the empty holes as well as the total perimeter of the compressed cylinder and the holes (in a 2-dimensional cross section). Is this possible? I'm trying to avoid having to export a million times as I need these values tabulated as the compression occurs.

Comments

  • disladisla
    edited June 2022
    Hi mbeauvais,

    I would create a very thin and soft (E=0.01 Pa) shells to cover the holes. Those shells could provide the area if the nodes are merged with the main body.
    By other hand, if the thickness of that shell is known and is very thin, perimeter could be computed as the side area of the shell / thickness.

    To get the time evolution of those surfaces you will probably need a script or modify an existing one.

    EDITED: Mmmhhh. To be accurate, that would provide the 3D surface and perimeter not really the 2D flat section cut. Would work in a plane strain section and starting with the compressed model to avoid shells buckling.



  • disladisla
    If you try this method refine enough the shells and check stresses in the area to discard any shell shear locking that could change the hyper elastic behavior.
  • disladisla
    edited June 2022
    Hi,

    That's a very interesting problem. ¿Are you designing a Control valve?

    I have tried with orthotropic shells with high E on thickness direction and they could be very useful as they preserve the thickness of the shell (needed to properly extract the perimeter). See example.
    These are just some ideas. Use it at your own risk.

  • disladisla
    edited July 2022
    Thin and low E linear truss elements work perfect to follow the change in the perimeter length.
    Merge with the desired perimeter to control.
    Stresses (which can be tabulated) should be directly proportional to the change in length of each individual truss.




Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register