Graph neural networks for simulating crack coalescence and propagation in brittle materials
نویسندگان
چکیده
High-fidelity fracture mechanics simulations of multiple microcracks interaction via physics-based models can become computationally demanding as the number increases. This work develops a Graph Neural Network (GNN) based framework to simulate and stress evolution in brittle materials due microcracks’ interaction. The GNN is trained on dataset generated by XFEM-based simulator. Our achieves high prediction accuracy test set (compared an simulator) engineering sequence GNN-based predictions. first stage determines Mode-I Mode-II intensity factors (which be used compute LEFM), second which will propagate, final actually propagates crack-tip positions for selected next time instant. capable simulating crack propagation, coalescence corresponding distribution wide range initial microcrack configurations (from 5 19 microcracks) without any additional modification. Lastly, framework’s simulation shows speed-ups 6x–25x faster compared These characteristics, make our attractive approach propagation with microcracks.
منابع مشابه
Anisotropic mesh adaptation for quasistatic crack propagation in brittle materials
The numerical simulation of quasistatic fracturing of brittle material, where no predefined crack path is imposed, is a challenging problem. In particular, we deal with the Francfort-Marigo model which requires the minimization of the well-known nonconvex and nonsmooth Mumford-Shah functional. To deal with a smoother problem which eases the minimization process of the energy, we consider the Γ-...
متن کاملCrack initiation in brittle materials
In this paper we study the crack initiation in a hyper-elastic body governed by a Griffith’s type energy. We prove that, during a load process through a time dependent boundary datum of the type t → tg(x) and in absence of strong singularities (this is the case of homogeneous isotropic materials) the crack initiation is brutal, i.e., a big crack appears after a positive time ti > 0. On the cont...
متن کاملCrack path prediction in anisotropic brittle materials.
A force balance condition to predict quasistatic crack paths in anisotropic brittle materials is derived from an analysis of diffuse interface continuum models that describe both short-scale failure and macroscopic linear elasticity. The path is uniquely determined by the directional anisotropy of the fracture energy, independent of details of the failure process. The derivation exploits the gr...
متن کاملAn Einstein Model of Brittle Crack Propagation
We propose a minimal nonlinear model of brittle crack propagation by considering only the motion of the crack-tip atom. The model captures many essential features of steady-state crack velocity and is in excellent quantitative agreement with many-body dynamical simulations. The model exhibits lattice-trapping. For loads just above this, the crack velocity rises sharply, reaching a limiting valu...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computer Methods in Applied Mechanics and Engineering
سال: 2022
ISSN: ['0045-7825', '1879-2138']
DOI: https://doi.org/10.1016/j.cma.2022.115021