LAUSR

Abstract Concrete structures are commonly exposed to low velocity impact loads originating from windborne/waterborne debris, vehicle/vessel collision, and rock fall. For the performance assessment of concrete structures under such loads, several constitutive models have been developed to date. There was, however, no holistic study to compare the accuracy of the available models for practical applications. To address this gap, the current study evaluates four constitutive models, i.e., continuous surface cap model (CSCM), elasto-plastic damage cap (EPDC) model, Karagozian and Case concrete (KCC) model, and Winfrith concrete model, in a systematic way. For this purpose, the constitutive models are first examined at the material level through single element simulations under basic stress paths, such as uniaxial compression and tension, as well as triaxial compression. A range of measures, such as post-peak softening, shear dilation, and confinement effect, are extracted and compared. Out of the four models, the KCC model is found to provide the most realistic response. The investigation is then extended to understand how the concrete constitutive models perform at the structure level. This is achieved by replicating drop hammer tests on reinforced concrete (RC) and concrete filled steel tube (CFST) beams. Investigation of these two structural categories provides a unique opportunity to further evaluate the accuracy of the concrete constitutive models in interaction with the most common reinforcement details. To achieve this goal, the impact responses of RC and CFST beams are compared with full-scale experimental test data. Upon understanding the capabilities of each constitutive model in predicting the structural behavior, a parametric study is carried out to examine the most important modeling parameters. The outcome of this study is expected to facilitate the selection and use of the concrete constitutive models for the design and assessment of concrete structures subjected to various low velocity impact loads.