change fracture criterion in XFEM

[wsfsz123]

Dear all,

I am trying to embed anisotropic fracture toughness property into XFEM, the crack direction is determined by the maximum energy release rate criterion (max(G(θ)/Gc(θ)), have anyone tried this? The current examples in XFEM modules are using maximal principle stress, can anyone help me to change the criterion?

Thanks!

[jiangwen84]

Are you planning to use a domain integral to compute the energy release rate G?

[wsfsz123]

Yes, the main purpose is to introduce the anisotropic fracture energy (Gc), then compare it to G to predict the crack direction. Thus I need to compute the instant integral firstly.

Thanks,
Shizhe

[jiangwen84]

We have published a paper before using K as the criterion for crack propagation. https://www.sciencedirect.com/science/article/pii/S0013794419307568

However, the codes have not merged into the MOOSE yet. We have post processor to compute K/J, and what is missing is to hook the post processor to XFEM crack cut object.

If we can find some time, I am very interested to add such capability in MOOSE.

[wsfsz123]

Thank you for your attention. I have read your paper and see the examples on MOOSE. It would be grateful if you could take time to do it, then we can use it with anisotropic fracture energy of material, up to now, there are rare investigations about it, and a good simulation tool with suitable criterion is required.

Thanks,
Shizhe

[jiangwen84]

Are you primarily interested in 2D or 3D? single crack or multiple cracks?

[wsfsz123]

2D primarily, and a single crack is enough for now, but we are also interested in the dynamic crack in the future. Is it possible to input Gc as a function of theta (the format of anisotropic fracture energy)?