Abstract: Abstract: To systematically characterize the dynamic mechanical behavior of 30CrMnSiNi2A ultra-high-strength steel under different stress states, strain rates, and temperatures, and to evaluate its ballistic resistance and failure characteristics under flat-nosed projectile impact, a comprehensive experimental and numerical investigation was conducted. A modified Johnson-Cook (MJC) plasticity model was employed to describe the plastic response, in which the strain-hardening term was represented by a linear combination of the Ludwik and Voce laws, together with strain-rate hardening and temperature-softening terms. An extended Xue-Wierzbicki (XW) fracture criterion incorporating stress triaxiality, the normalized third invariant of the deviatoric stress tensor (Lode parameter), strain rate, and temperature was adopted to characterize ductile fracture. The material parameters were calibrated using a combined experimental and finite-element-based identification procedure. Quasi-static tests on smooth round-bar, notched round-bar, flat shear, and plane-strain specimens were performed to cover different stress states. Split Hopkinson pressure bar compression tests at strain rates of 1063.8~4753.8 s⁻¹, high-rate tensile tests at 25~1000 s⁻¹, and elevated-temperature tensile tests from 200 to 800 °C were further conducted to characterize strain-rate and temperature effects. Digital image correlation was used to measure local deformation and strain fields in the shear, plane-strain, and high-rate tensile specimens. The quasi-static hardening parameters were determined from the smooth round-bar tensile response, with one weighting parameter identified through inverse optimization by coupling Isight with ABAQUS. The strain-rate and thermal-softening parameters were calibrated from the measured yield stresses at different strain rates and temperatures. For fracture calibration, finite element simulations of the smooth and notched round-bar, shear, and plane-strain tests were used to extract the histories of equivalent plastic strain, stress triaxiality, and Lode parameter at the fracture location, from which the stress-state-dependent fracture locus was established. The strain-rate- and temperature-dependent fracture parameters were subsequently determined from the high-rate and elevated-temperature tensile tests. Ballistic resistance was evaluated through 13 impact tests using nominally 6-mm-diameter flat-nosed projectiles against 4-mm-thick steel targets at impact velocities of 196.2~413.9 m/s. The ballistic limit was obtained by fitting the initial–residual velocity data using the Lambert-Jonas relationship. Axisymmetric finite element simulations were finally performed in ABAQUS to evaluate the predictive capability of the calibrated material model and to quantify the influence of local element size. The results show pronounced strain-rate strengthening and thermal softening of 30CrMnSiNi2A steel. As the strain rate increased from the quasi-static regime to 4547.8 s⁻¹, the yield strength increased from 1095.6 to 1497.0 MPa, whereas a marked loss of strength occurred at temperatures of 600 °C and above. The ductile fracture response exhibited a pronounced dependence on stress state and the Lode parameter. Relative to uniaxial tension, the fracture strains under shear and plane-strain tension decreased by 35.8% and 63.9%, respectively. The Lambert-Jonas fit to the ballistic-test data yielded a ballistic limit of 209.9 m/s, and shear plugging was identified as the dominant target failure mode. The numerical ballistic limit showed significant mesh-size dependence: local element sizes of 75 and 50 μm resulted in predicted ballistic limits of 213.7 and 205.0 m/s, corresponding to deviations of +1.8% and −2.4% from the experimental fit, respectively. The dominant shear-plugging failure mode was also reproduced numerically. The results demonstrate that the systematically calibrated MJC plasticity model combined with the extended XW fracture criterion provides an effective description of the dynamic plasticity and stress-state-dependent ductile fracture of 30CrMnSiNi2A steel. With an appropriately refined local mesh, the modeling framework can reasonably predict the ballistic limit and dominant shear-plugging failure of thin 30CrMnSiNi2A steel targets subjected to flat-nosed projectile impact.