Orthogonal Test and Mesoscopic Numerical Simulation of Dynamic Compression Performance of Ultra-High Performance Concrete at Elevated Temperatures

Abstract

To evaluate the influence of fiber content, temperature, and loading rate on the dynamic compression performance of ultra-high performance concrete (UHPC), orthogonal Split Hopkinson Pressure Bar (SHPB) tests were performed on 120 MPa UHPC specimens. Range analysis of the test data reveals that the steel fiber content exerts the largest range on dynamic compressive strength, with loading rate ranking second and temperature having the least effect. A three-dimensional mesoscopic finite element model that accounts for temperature-dependent behavior was developed using a modified Karagozian & Case (K&C) constitutive model together with high-temperature bond–slip degradation curves. The simulated peak stresses are generally higher than the experimental values, with a Root Mean Square Error of 9.02 MPa, a Normalized Root Mean Square Error of 4.65%, and a maximum discrepancy of 10.07%, while the major experimental failure characteristics are reasonably reproduced. Additional numerical simulations indicate that the influence of steel-fiber content becomes increasingly temperature-dependent. Within the experimentally investigated range up to 300 °C, higher fiber content generally improves dynamic response and specimen integrity. At 600~800 °C, the numerical extrapolations suggest that the reinforcing efficiency of steel fibers may be substantially reduced under the assumed temperature-dependent degradation conditions. These high-temperature trends require further experimental validation.