A Generic, Time-Resolved, Integrated Digital Image Correlation, Identification Approach

Abstract

A generic one-step Integrated Digital Image Correlation (I-DIC) inverse parameter identification approach is introduced that enables direct identification of constitutive model parameters by intimately integrating a Finite Elements Method (FEM) with Digital Image Correlation (DIC), directly connecting the complete time sequence of experimental images to the sought model parameters. The problem is cast into a transparent single-minimization formulation with explicit expression of the unknowns, being the material properties and, optionally, experimental uncertainties such as misalignments. The tight integration between FEM and DIC creates an information dialogue that yields accurate material parameters while providing necessary regularization to the DIC problem, making the method robust and noise insensitive. Through this method the versatility of the FEM method is translated to the experimental realm, simplifying the existing experiments and creating new experimental possibilities.

A convincing demonstration of the method has been achieved by successful identification of three challenging models from three very different (virtual) experiments, all thoroughly analyzed on accuracy and noise sensitivity: identification of a mixed-mode interface model from a mixed-mode delamination test, a 10-parameter history- and rate-dependent glassy polymer model from a simple tensile test, and simultaneous identification of two material models from a single bulge test of a structured membrane.