Seeking UMAT Implementation for Hyperelastic Marlow-like Material Model

1

Hello CalCuliX community,

I’m currently working on implementing a UMAT for hyperelastic materials in ccx, specifically a model similar to the Marlow model in Abaqus (Abaqus docs). The goal is to use experimental uniaxial tension data to define the stress-strain response.

Goals:

  • Implement a FORTRAN UMAT that:
    • Uses linear interpolation or cubic spline interpolation to extract stress and tangent modulus from tabulated data.
    • Computes shear modulus (G) and bulk modulus (K) dynamically from the experimental data.
    • Ensures proper 3D implementation for nearly incompressible hyperelastic materials.
  • Develop a unit test program to verify interpolation accuracy.

Request for Help:

Has anyone here developed a similar UMAT? If so:

  1. Would you be willing to share your approach or code?
  2. What methods did you use to validate the UMAT against experimental or analytical results?
  3. Did you encounter any challenges with stability, convergence, or large deformations?

I’d love to exchange ideas and see if anyone has tackled something similar before. Any insights or shared experiences would be greatly appreciated!

Thanks in advance! :rocket:

2

You can find some Abaqus UMAT examples for other hyperelastic models such as Neo-Hookean.

You could use Treloar data and benchmarks from Abaqus documentation. But first try with a single element test.

Also, here’s a nice example with validation: GitHub - Sina-Taghizadeh/UMAT_Hyperelastic: Comparison of Abaqus Library, UMAT, and Analytical Solution for Neo-Hookean Hyperelastic Material Under Large Deformations

If in doubt, R. Marlow works for DS and is quite active in the Simulia User Community.

1 Like
3

Hi,
If you implement a model with user parameters, I can help you to run calibration towards exp data. I have PIDO software (Optimus).

1 Like
4

Thank you @Calc_em and @Kankreu ! I will work on the implementation and provide an update in a few weeks.