I'm doing a simulation for a thin plate polycrystalline subjected in pure bending, and I consider the grain boundaries by the surface interaction using the cohesive surface behaviour in Abaqus. But, as the plate is bent, there is a portion of the cohesive surface in traction and the other portion in compression, I don't know how the surface will react before the compression. Otherwise, should I define the hard contact property at the same time? Thank you very much.
Since there is nobody who replies to this question. I'd like to talk about my own opinions. I think there are two solutions. The first is to set up the general contact for the solid elements, as the cohesive elements have zero thickness, when their two surfaces tend to interpenetrate wthin each other, the contact works.The second one is to complete the cohesive law by ourselves, which means writting a UMAT subroutine in which we define a very large compressive stiffness. What do you think? Thank you.
- Badges
- Science topic