Prediction of void evolution in sheet bending based on statistically representative microstructural data for the Gurson-Tvergaard-Needleman model
Ductile damage in sheet steels is caused by voids. It is crucial for product design to predict the distribution of voids in bent components. Since the void volume fraction is a state variable in the Gurson-Tvergaard-Needleman (GTN) model, it is applied to predict the evolution of voids in bending. Material parameters are identified based on force-displacement curves of a dual phase steel and also through statistical microstructural information obtained from panoramic scanning-electron microscopy images. The void volume fraction and particular void populations of GTN-model are determined with a recently proposed scheme, which involves machine learning algorithms.