Validation of homogeneous anisotropic hardening model using non-linear strain path experiments

Abstract

We conducted non-linear strain path experiments on a dual-phase steel-sheet sample. The first strain path in each of these experiments, called pre-strain, was always uniaxial tension in the direction transverse to the sheet (90 degrees from the rolling direction). For this purpose, we clamped large tensile specimens in a suitable gripping system and deformed them to produce a sufficiently large area with a uniform strain distribution. We machined new specimens from this area, and we subjected them to different modes of deformation, namely uniaxial tension in different directions from the transverse direction, simple shear, and biaxial tension with different load ratios. In addition, we performed uniaxial tension-compression tests consisting of either one or three full cycles. We obtained the coefficients of a distortional plasticity model -the homogeneous anisotropic hardening (HAH) model -using an inverse analytical identification tool. First, we calibrated two sets of reverse loading coefficients using the tension-compression stress-strain curves obtained with either one or three full cycles. Then we used the 90 degrees tension-45 degrees tension and the 90 degrees tension-0 degrees simple-shear sequences to determine two sets of cross-loading coefficients. We compared the other independent experimental flow curves with simulations using the HAH model with combinations of the different sets of coefficients. These comparisons showed the influence of the selected input data on the calculations of the coefficients calibrated and demonstrated the advantages and limitations of the present HAH model.