Constitutive Modeling of High Strength Steel Sheets

Abstract

The development of constitutive models for two high strength steels (HSS), TWIP steel with a FCC structure and ferritic stainless steel with a BCC structure, was carried out in this work. Advanced yield functions, Yld2000-2d for plane stress and Yld2004-18p for general stress states, which were developed originally for aluminum alloys, were evaluated for predicting the forming performance in the cases of steels.In order to characterize the strain hardening in balanced biaxial tension, the hydraulic bulge and disk compression tests were carried out for a ferritic stainless steel sheet sample. The “frictionless” disk compression strain hardening curve was found to be in reasonable agreement with the hydraulic bulge strain hardening curve.The macroscopic constitutive modeling of two TWIP steel sheets, TWIP940 and TWIP1000, was studied. Yield function Yld2000-2d was shown to capture the uniaxial tension flow stresses and values better than von Mises and Hill 1948. In addition, the results of the disk compression tests demonstrated that the friction has little effect on plastic deformation and anisotropy, but significant influence on the stress-strain curve. Moreover, in order to investigate the stretch flangeability of TWIP steel, the flat-bottom hole expansion test was performed for TWIP1000 experimentally and numerically. The results showed that the 3D model with solid element can predict the hole radius and thickness strain better than the 2D model with shell element. It was also demonstrated that only Yld2004-18p predicted the plastic anisotropy during the hole expansion. Constitutive modeling of the two ferritic stainless steel sheets, AISI409L and AISI430, was investigated at the macroscopic and mesoscopic scales. At the macroscopic scale, Yld2000-2d and Yld2004-18p were shown to describe the flow behavior and plastic anisotropy better than other anisotropic yield functions considered. Using crystal plasticity, the viscoplastic self-consistent (VPSC) model demonstrated that, for AISI409L, texture evolution was mainly responsible for the hardening difference between the uniaxial tension and balanced biaxial tension. For AISI430, though, texture evolution may not be the only reason for anisotropic strain hardening. Moreover, the forming limit diagrams (FLDs) of both materials were predicted with the Marciniak and Kuczynski (MK) and Bressan-William-Hill (BWH) models associated with the yield function Yld2000-2d and the Swift hardening law. For both materials, the BWH model led to a good prediction of the experimental limit strains. The MK model was not so successful, no matter whether the strain rate sensitivity effect was considered or not. The results of the modeling work and micrograph observations both indicated that the failure mechanism was localized necking in uniaxial and plane strain tension, and localized shearing in balanced biaxial tension. In conclusion, the advanced yield functions, Yld2000-2d and Yld2004-18p, are suitable for evaluating the forming performance of steels, either with a FCC or BCC structure.