Prediction of ductile failure using a local strain-to-failure criterion

In this article, we provide the details of the predictive simulations performed by the University of Texas team in response to the 2012 Sandia Fracture Challenge (Boyce et al. in The Sandia Fracture challenge: blind predictions of ductile tearing. Int J Fract. doi: 10.1007/10704-013-9904-6, 2013). The material constitutive model was calibrated using the tensile test data through an optimization scheme. A modified Johnson–Cook failure criterion was also partially calibrated using the material characterization data obtained from a tension test and a compact-tension fracture test. These models are then embedded in a highly refined finite element simulation to perform a blind prediction of the failure behavior of the Sandia Fracture Challenge geometry. These results are compared with experiments performed by Sandia National Laboratories and additional experiments that were performed at the University of Texas at Austin with full-field three-dimensional digital image correlation in order to explore the different failure modes. It is demonstrated that a well-calibrated model that captures the essential elastic–plastic constitutive behavior is necessary to confidently capture the elasto-plastic response of challenging structural geometries; it is also shown that a simple ductile failure model can be used to predict ductile failure correctly, when proper calibration of the material model is established.