Hi,

I also had similar problem. The linear elastic problem converges quite easily whereas the problem fails to converge completely when elastic-plastic material is defined. The XFEM ABAQUS is formulated assuming linear elastic problem so due to the non-linear material behavior, it suffers convergence problem.

Instead of using elastic-plastic model directly, you can first try using bilinear material model (perfectly plastic) and try to check how the model behaves. Then you can jump into the elastic-plastic model.

I hope you have adjusted step parameters (reduce step sizes, increase no. of steps,..) and solution parameters (discontinuous analysis and allow higher number of cutbacks (IA)) to allow nonlinear problem solution. Further, you can add damping and stabilization to check how it helps to converge the problem.

One of the previous post (https://www.researchgate.net/post/How_to_converge_XFEM_crack_propagation_for_crack_growth_towards_and_beyond_inclusion) could be of your interest.

Also do check the capabilities and limitations of the current XFEM formulation in ABAQUS, as in my case the limitation of initiating only one crack per enrichment region if damage initiation criteria reached at different time step, was the major problem stopping further crack growth.

Cheers,

Thank you for your answer. The problem was I had a false value for Fracture Energy criteria.

Regards,

Hi, Surendra Sujakhu and Yann Zimmermann

I have been simulating a similar model, so can you share your model document because I am confused about how to set the damage criteria.

My email address is [email protected]

Thanks.