Extension of the sensitivity-based virtual fields to large deformation anisotropic plasticity

The virtual fields method is an approach to inversely identify material parameters using full-field deformation data. In this work, we extend the sensitivity-based virtual fields to large deformation anisotropic plasticity. The method is firstly generalized to the finite deformation framework and then tested on numerical data obtained from a finite element model of a deep-notched specimen subjected to a tensile loading. We demonstrated the feasibility of the method for two anisotropic plasticity models: Hill48 and Yld2000-2D, and showed that all the parameters could be characterise from such a test. The sensitivity-based virtual fields performed better than the currently accepted standard approach of user-defined ones in terms of accuracy and robustness. The main advantage of the sensitivity-based virtual fields comes from the automation of virtual fields generation. The process can be applied to any geometry and any constitutive law.