Inverse estimation of material model parameters using Bayesian data assimilation

Abstract. This study proposes a new method for the inverse estimation of the parameters included in material models from full-field measurement data that are obtained using the digital image correlation method. This approach is based on data assimilation according to the Bayes’ theorem (Bayesian data assimilation). In this study, we demonstrate the assimilation of experimental data obtained from uniaxial tensile, forming, and fracture tests of aluminum alloys into elastoplastic finite element and phase-field crack propagation simulations. The proposed method allows the simultaneous estimation of multiple material model parameters. The Bayesian data assimilation is a promising methodology for estimating the parameters of different material models and constructing digital twins of material deformation.