A HYBRID MODEL OF DUCTILE FAILURE ACCOUNTING FOR STRAIN HARDENING [Poster #8]
Sahil WajidGrand Ballroom D
Existing ductile failure models such as the Gurson-Tvergaard-Needleman (GTN) model as well as more recent physics-based models (for instance, the Benzerga-Leblond coalescence model from 2014) were all derived for perfectly plastic porous materials using classical limit analysis, with plastic flow in the matrix being described by J2 flow theory. When extended heuristically to hardenable materials, these models do not account for the heterogeneity of plastic strain in the matrix, and are unable to capture the effect of hardening on the evolution of porosity, the primary damage variable.
This work uses “sequential limit analysis” (SLA) to first derive a hardening-sensitive void coalescence criterion for a cylindrical cell containing a coaxial cylindrical void of finite height, by discretizing the intervoid ligament into a finite number of shells in each of which the quantities characterizing isotropic hardening are considered to be homogeneous. Next, this new criterion is combined with a recently formulated hardening-sensitive void growth criterion (also derived using SLA) to obtain a hybrid model of ductile failure. The new constitutive formulation’s ability to remedy the two aforementioned shortcomings of existing models is examined, and a set of finite-element micromechanical unit cell calculations is used to further assess the model's predictive capabilities.