Curve fitting of hyperelastic materials

[lowellsmoger]

ANSYS seems to curve fit biaxial stress-strain data very well up to 90% (and calculate representative coefficients for the Mooney Rivlin model).

Yet when I extract data from the equibiaxial region of my planar biaxial simulation, ANSYS estimates stresses not in line with the curve I used to generate the coefficients.

The extracted data looks more like the typical hyperelastic material (

http://www.axelproducts.com/images/3_strains.gif)

but as you can see in the attachment, the curve does not stiffen before 90%.

 

[Elabbasi]

The Mooney Rivlin material model does not stiffen at high strains when loaded in plane strain or uniaxial tension. You have to select a material model that has that stifferning feature (which is very common for elastomers). Try the Arruda-Boyce model for example.

Also, when curve fitting stress-strain data, I believe it is better to use all your experimental data (uniaxial, biaxial, etc.) and find the best fit for all the curves combined.

Nagi Elabbasi
Veryst Engineering