LAUSR

Abstract A phenomenological failure criterion, which attempts to achieve a tradeoff between simplicity and prediction quality, is developed to predict the fracture behavior of ship structures under quasi-static impact loads. For simplicity, only the dominant physical mechanism remains in the model. In high stress triaxiality regime, it is believed that ductile fracture is mainly influenced by the stress triaxiality, while under low stress triaxiality regime, Lode angle is considered as the key factor in the fracture prediction. Thus, the combined maximum shear stress and Rice Tracey (MSSRT) criterion is developed to cover a wide range of stress states. The calibration parameters in the criterion can be determined from the uniaxial tensile test only, which is suitable for industrial application where available material data are limited. Along with the fracture scaling framework, the developed failure criterion is implemented into the explicit finite element code LS-DYNA, and the performance of the failure criterion is evaluated by making comparisons with a series of indentation tests involving different materials, structural arrangements and indenter sizes. It is shown that the fracture behavior is well predicted by simulations for all cases. Besides, comparisons among different failure criteria are also carried out. It is found that MSSRT criterion could generate more consistent results when compared with other commonly used failure criteria.