Model-free data-driven inelasticity in Haigh-Westergaard space - a study how to obtain data points from measurements

Model-free data-driven computational mechanics replace phenomenological models with numerical simulations in strain-stress space based on sample data sets. The approach has recently been extended to inelasticity problems employing structured data sets, tangent space information, and transition rules. Coverage of qualified data states and calculation of the related tangent space is critical from the standpoint of actual application. This study applies the data-driven paradigm to elasto-plasticity with isotropic hardening. We develop our method using Haigh-Westergaard coordinates, which provide information on the underlying material yield surface. Based on this, we employ a combined tension-torsion test to cover the yield surface knowledge and a single tensile test to determine the tangent space. Under compatibility and equilibrium requirements, the resulting data-driven technique minimizes the distance over the Haigh-Westergaard space augmented by directions in the tangent space.