Validation of Stress-Strain curves for Large Deformation analysis

[bedyhope]

I am using Abaqus to simulate tensile test beyond necking in order to validate the Stress-strain curve through the comparison of numerical and experimental Load-Displacement curves. Due to asymmetry, I model one eight of flat dog bone specimen, the boundary condition are shown in the figure below. I gave density, young’s modulus, poison ratio, and point wise true stress-plastic strain curve as the mechanical material properties. Also I have selected type of nonlinear geometry (NLGEOM) to account for large deformation ,yield function criterion as Von Mises and Hardening rule as isotropic. I have solved the problem , and got the load (from sum of nodal reaction force (RF) at the plane of applied displacement, at the same time I obtain numerical displacement of a node that distanced 25 mm (half of gage length) from the fixed edge.

My problem is that when I compared Experimental and Numerical load-displacement I found that, the Numerical load-displacement corresponds very well to the Experimental one just little beyond the necking point as shown in the figure below. I tried to adjust the material beyond necking and also to use fine mesh to get better correspondence but the problem still the same.



If anyone wants to see the input file I can up load it.

Please if someone has a solution or hint let me know