Prediction of plastic instability in extruded aluminium alloys using shell analysis and a coupled model of elasto-plasticity and damage

A test programme has been carried out to establish the mechanical properties of extruded aluminium alloys. The investigated alloys exhibited significant anisotropy in strength, plastic flow and ductility. The anisotropy is mainly due to crystallographic texture, and an anisotropic yield criterion is needed to describe the mechanical behaviour. Several yield criteria were evaluated against the experimental data. The yield criterion Yld96 proposed by Barlat and co-workers was found to be superior with respect to accuracy for the actual aluminium alloys. In an attempt to model ductile failure of the investigated materials, a coupled model of elasto-plasticity and ductile damage was implemented in LS-DYNA for plane stress analysis with co-rotational shell elements. The material model combines the anisotropic yield criterion Yld96 with the associated flow rule, isotropic strain hardening and isotropic damage. The parameters defining the yield criterion and the strain hardening were determined from tensile tests and pure bending tests, while the damage parameters were identified using inverse modelling of tensile tests performed with purpose-made specimens. The numerical analyses of the uniaxial tensile tests showed good agreement with the experiments. Different failure modes caused by plastic instability mechanisms are well predicted by the model, which is a result of the accurate representation of the plastic anisotropy of the material. A numerical study of a patch of shell elements revealed that the calibrated constitutive model resulted in plastic instability before critical damage also in the stretch-stretch region of the forming limit diagram, given that a very small geometric inhomogeneity was introduced to the elements. The results show that shell element analysis is applicable to predict the important failure mechanism of plastic instability in for instance sheet metal forming.