Hello, everyone. I have encountered an issue during the simulation of triaxial compression of rocks using the perfectly elasto-plastic Mohr-Coulomb model in Abaqus. In my simulation, the stress-strain curve consists of two inclined straight lines (Fig. 1). But in theoretical, the stress-strain curve should be composed of an inclined straight line followed by a horizontal line. Could anyone explain why this inconsistency occurs?
Here are the details of my setup: the height of the cylinder is 100mm and the radius is 50mm (Fig. 2), the cohesion and friction angle are 21.569MPa and 35.42°, both transformed from the rock parameters, and the flow rule is associated. In addition, I would like to know how to obtain the yield curve based on the results of the triaxial compression test?
Dear Hu Shihong,
It seems to me that there might be a confusion between the constitutive model and the model behavior. Each element within your model must behave according to the MC model you defined. However, the behavior of the model in total may not be elastic – perfectly plastic. To verify ths, you may load the model up to a higher strain value.
Can you share also the value of Poisson’s ratio?
Dear Mert Eyupgiller,
First much thank you for your reply. I think I have found the reason why the stress-strain curve appears to show hardening behavior when using an elastic-perfectly plastic constitutive model. The main reason is that before loading, the geostress is not balanced.
Additionally, the value of Poisson’s ratio is 0.25.
Dear Hu Shihong
I can add the following: The built in Abaqus Mohr-Coulomb model can never use associated plasticity, as the yield surface is different from the plastic potential.
I have the “classic” Mohr-Coulomb as an Abaqus UMAT. Feel free to reach out to me, if you are interested.
Best regards
Johan