Experimental and numerical studies on dynamic behavior of reinforced UHPC panel under medium-range explosions

Highlights

• A box-like blast loading apparatus is designed to provide the one-way simply-supported boundary constraints and avoid the effect of the diffracted blast waves.

• Blast-induced incident and reflected overpressure-time histories, and dynamic responses of panels were experimentally obtained under medium-range explosions.

• The effectiveness of CONWEP method in predicting the blast loadings for the present scaled distances and detonation point was proved.

• KCC model parameters of UHPC were comprehensively calibrated and fully validated.

Abstract

This paper aims to study the blast-resistant behavior of one-way simply-supported reinforced ultra-high performance concrete (UHPC) panels through field tests and numerical simulations. Firstly, by designing a box-like blast loading apparatus, both three reinforced UHPC panels and three reinforced normal strength concrete (NSC) control panels were fabricated and tested under medium-range explosions from end-detonated cylindrical charges with different scaled distances (0.5~1.0 m/kg^1/3). The valuable data including the explosion-induced incident and reflected overpressure-time histories, deflection- and acceleration-time histories, as well as the post-blast damage of panels were obtained and assessed experimentally. The superiority of UHPC as a blast-resistant material was proved quantitatively by comparison with NSC. Then, the corresponding 3D finite element (FE) model was established by adopting the program LS-DYNA, both the blast loadings induced by the medium-range explosions and the constitutive model parameters of UHPC were mainly concerned. The applicability of the CONWEP method in predicting the blast overpressures was verified, but the prediction precision declines with the decrease of scaled distance since the influences of charge shape and detonation point enhanced accordingly. Moreover, based on the systematic static and dynamic mechanical tests data, the parameters of Karagozian & Case concrete (KCC) model describing the strength surface, equation of state (EOS), damage evolution, and strain rate effect of UHPC were calibrated. Finally, the present FE model, numerical algorithm and the calibrated model parameters were fully verified by comparing the numerical results with the test data, which could provide reference for the evaluation and design of UHPC structures under blast loadings.

Introduction

Over the past several decades, the explosive effects that stem from terrorist attacks and industrial accidents have attracted increasingly more public concerns. The occurrence of explosions will threaten the safety of the surrounding building structures and lead to huge economic losses and extensive casualties. Therefore, it is of great significance to improve the blast-resistant performance of civil structures, especially for the strategically vital infrastructures, landmark buildings, and military or civil shelters, etc. Concrete is a widely applied construction material and the traditional normal strength concrete (NSC) structures have limited blast-resistance attributed to its low tensile capacity and energy absorption capacity. Comparably, ultra-high performance concrete (UHPC) is a relatively new type of cementitious material mixing with a very low water-to-binder ratio, high amount of high-range water reducer (HRWR), fine aggregates and high-strength steel or organic fibers. Due to its prominent mechanical properties, i.e., high compressive and tensile strengths, high ductility as well as the high fracture energy, UHPC is regarded as the most prominent construction materials for both civil and military structures to resist the intensive loadings, e.g., high-speed projectile penetration [1], [2], [1], low-velocity drop hammer impact [2], [3], [4], [5], [6], [7] and blast [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]. The blast-resistant performance of reinforced UHPC panels is mainly concerned in the current study.

From the detailed review of the existing related work in Section 2, generally, the existing related work still has the following limitations: (i) Based on the plenty of the test data, Orton et al. [26] defined the scaled distance Z≤0.4m/kg^1/3, 0.4<Z≤1.0m/kg^1/3 and Z>1m/kg^1/3 (Z=R/W^1/3 is the scaled distance, R is the standoff distance from the explosive center to the structure surface, and W is the charge weight of the equivalent TNT explosive) as the close-, medium- and far-range explosion scenarios, respectively. Most of the existing work is either concerning the local damage and failure of UHPC panels under contact/close-in explosions or focusing on the dynamic performance of UHPC panels under far-range explosions, the related studies for the medium-range explosions are therefore insufficient; (ii) For the design of test setup, the existing test UHPC panels were usually constrained by a ground frame which is completely exposed to air and could not eliminate the effect of the diffracted blast waves on the rear face of the test panel. The boundary conditions of the test panel provided by the steel angles and wood stripes are not clear. Besides, attributed to the damage of the measuring sensors under intensive blast loadings, the systematic dynamic responses of panels, e.g., deflection- and acceleration-time histories, as well as the instantaneous incident and reflected blast overpressures are limited; (iii) KCC model [27] (*MAT_CONCRETE_DAMAGE_REL3, *MAT_72R3) embedded in LS-DYNA [28] is validated for capturing the complex concrete behaviors and predicting the structural dynamic responses under blast loadings with good accuracy [27], [28], [29]. The most significant user improvement provided by *MAT_72R3 is the model parameters auto-generated algorithm based solely on the unconfined compression strength of concrete. However, this parameter generation algorithm is proposed for NSC with the uniaxial compressive strength of 45 MPa. Therefore, the corresponding KCC constitutive model parameters for UHPC should be comprehensively calibrated, however, the related work is very lacked.

At present, in Section 3, three reinforced UHPC panels and three NSC control panels were fabricated and tested on a specially designed box-like blast loading apparatus to avoid the effect of the diffracted blast waves and provide the simply-supported boundary constraints, and the scaled distances of 4 kg TNT charges were ranged from 0.5 to 1.0 m/kg^1/3. Various sensors were arranged to record the incident and reflected overpressure-time histories, deflection-and acceleration-time histories. Besides, by examining the damage of the post-blast panels, the blast-resistance of NSC and UHPC panels are evaluated quantitatively. Then, in Section 4, the refined FE models of the present test were established, and the KCC model parameters for UHPC were comprehensively calibrated. Finally, in Section 5, the applicability of the CONWEP method to predict the blast overpressures was firstly discussed, and then the simulated dynamic responses and damage of panels were compared systematically with the test results. The present conclusions can provide references in the design and analyses of UHPC protective structures under blast loadings.