Anisotropic Mechanical Behavior and Localized Deformation Evolution in Q420 High-Strength Steel

Q420 high-strength steel exhibits pronounced anisotropy due to its rolling process, and conventional uniaxial tensile testing is incapable of acquiring strain field evolution information during the local necking stage. In this study, quasi-static uniaxial tensile tests were conducted on Q420 cold-rolled high-strength steel sheets at six orientations (0°, 15°, 30°, 45°, 60°, and 90°) using Digital Image Correlation (DIC) technology. The evolution of the strain field and the corresponding stress–strain responses at different orientations were systematically investigated. The results show that the DIC technique effectively captured the full-field strain evolution of the specimens from uniform deformation to local necking and final fracture in all directions. Taking the 0° direction as an example, the local maximum engineering strain prior to fracture reached 35.866%, whereas the average fracture strain within the gauge section was only approximately 22.5%, corresponding to a ratio of approximately 1.6 and clearly demonstrating the severe strain concentration within the necking zone. The stress–strain curves corresponding to different rolling directions exhibited pronounced anisotropy. The tensile strength was highest in the 90° direction and lowest in the 0° direction; however, the 0° direction exhibited the best ductility, whereas the 45° direction showed the poorest ductility. Among the six orientations, the midpoint transverse engineering strain exhibited the largest absolute value in the 45° direction, further indicating that this orientation is the most susceptible to plastic instability. In this work, DIC-based full-field measurement was combined with multi-directional tensile testing to quantitatively characterize the relationship between local strain concentration and anisotropy. The findings provide high-precision experimental data for the calibration of anisotropic constitutive models and the optimization of forming processes.