The Gurson-Tveergard-Needleman (GTN) model has been widely used to describe ductile fracture. In this paper, a series of tensile tests were carried out on notched specimens to assess the GTN model. The GTN model parameters were calibrated from a smooth tensile specimen by a hybrid particle swarm optimization, and the reliability of the calibrated parameters was verified by the profile of the smooth tensile specimen. The calibrated parameters were used to predict the ductile fracture of notched specimens. A comparison of fracture initiation sites between simulations and experiments indicates that the GTN model has a good performance on predicting fracture initiation site but fails at predicting fracture moment. The assessment of the transformability of the GTN model parameters was performed by comparing the load-displacement curves between simulations and experiments. It is observed that the GTN model parameters are material constant, except the critical void volume fraction fc. The influence of stress triaxiality on the critical void volume fraction fc is also discussed.

This content is only available via PDF.
You do not currently have access to this content.