Benchmark study of failure criteria for ship collision modeling using purpose-designed tensile specimen geometries

Finite element analysis is extensively employed to predict the structural response of marine structures when subjected to operational loads. The quality of the obtained results is to a high degree dependent on the models used to describe the mechanical behavior of the materials involved. The present study reports on an extensive experimental program setup to evaluate constitutive models for shipbuilding materials. Models used for the elasto-plastic behavior, strain rate sensitivity and material failure are presented for a carbon steel sheet. Four failure criteria developed particularly for ship collision modeling are critically assessed: the Equivalent Plastic Strain criterion (EPS), the Germanischer Lloyd criterion (GL), the Rice-Tracey-Crockroft-Latham criterion (RTCL) and the Bressan-Williams-Hill criterion (BWH). To evaluate these failure criteria performance over a wide range of triaxialities, tensile experiments are carried out on six purpose-designed coupon geometries. The plastic strains, strain rates and triaxialities are evaluated using experimental full-field photogrammetric techniques and numerical finite element simulations. Comparison of the experimental and numerical results shows a good performance of the RTCL, BWH and EPS criteria.