A Data-driven Approach for Characterizing Elastoplastic Materials with Full-field Displacement Measurements Using Dual Stereo Digital Image Correlation

Abstract

A data-driven identification (DDI) method is developed to evaluate material properties and stress distribution from full-field displacement distributions measured by digital image correlation. It is not possible to obtain stresses directly from strains without assuming constitutive laws, which are often empirical and contain uncertainties. In contrast, the DDI approach enables the computation of reasonable stresses by employing uncertainty-free displacement-strain relations, equilibrium equations, and a distance minimization function. The proposed method is applied to displacement fields of an elastoplastic material obtained through digital image correlation to identify material properties and stress distributions. Results demonstrate that the equivalent stress-strain relationship closely matches the reference value and the appropriate stress distributions are obtained, thereby validating the plausibility of the proposed method.