Damage mechanisms and rapid damage assessment of RC columns subjected to near-field explosions

As critical load-bearing members, reinforced concrete (RC) columns may suffer severe damage under near-field explosions, directly threatening structural stability and potentially triggering progressive collapse. Therefore, this study identified the damage-mode transition and developed a rapid damage-assessment method for RC columns subjected to near-field explosions. Initially, two field blast tests were conducted sequentially on two columns in a two-story prototype RC frame structure, targeting a corner column and a middle-edge column on the ground floor. The tests were performed at scaled distances of Z = 0.14 and 0.23 m/kg¹ / ³ , respectively, to investigate their damage modes and dynamic responses. Subsequently, parametric analyses were conducted to evaluate the effects of axial compressive strength of concrete f c , yield strength of longitudinal reinforcement f s , transverse reinforcement ratio ρ v , axial load ratio α , and slenderness ratio β under various Z . The numerical results indicated that, for Z ≤ 0.20 m/kg 1/3 , the failure mode was dominated by localized damage, and the parameter sensitivity was ranked as α > β > ρ v > f c > f s . In contrast, at Z = 0.4 m/kg 1/3 , the failure mode was dominated by global flexural deformation, and the parameter sensitivity was ranked as β > f s > α > ρ v > f c . Moreover, because slight variations in key f actors under near-field explosions may lead to markedly different damage modes, a rapid damage assessment method based on the mass loss ratio M s of RC columns was proposed. The method establishes a link between local material loss and global residual axial capacity and is applicable to both local and global damage modes. Based on regression analyses of 60 cases, damage thresholds for M s were calibrated against residual bearing-capacity criteria, and a curve-fitting formula was developed as a function of Z and structural parameters. The validation results agreed well with the experimental and numerical data, with prediction deviations ranging from 1.13% to 13.08%.