Material constitutive parameters are usually identified from standard tests exhibiting homogeneous strain states. However, recent improvements in inverse identification methods call upon the use of heterogeneous tests. The purpose of the present work is to develop an inverse identification procedure based on the Finite Element Updating method. Kinematic and thermal full-fields measurements are thus captured from increasingly heterogeneous tests, and provide the input data of the identification process. Various planar tests with commercially pure titanium have been performed and a new non planar experimental test has been designed and set up. This latter allows the identification of tensile, shear and expansion behaviours from a single test conducted on a standard tensile device. Digital Image Correlation along with Stereo Image Correlation is used for measurement purposes. Comparisons are made between measured data and retrieved data from a numerical model of the experiment. An optimization algorithm is used to update the constitutive parameters of the finite element model. The number of identified parameters varies with the chosen constitutive model namely : orthotropic elasticity, then anisotropic plasticity. Several validation protocols highlight the advantages of using heterogeneous rather than homogeneous tests. Moreover it has been shown that the choice of the elastic model (isotopic or ortho
展开▼
