A comparative study of three forms of an uncoupled damage model as fracture judgment for thin-walled metal sheets

Ductile fracture always occurs in plastic forming processes of thin-walled metal sheets. Damage models, both coupled and uncoupled ones, have been extensively developed to predict fracture of the sheets. Among these, uncoupled damage models, owing to their simple form are widely used. While for an uncoupled damage model, there exist three forms for its fracture prediction: (1) damage threshold (DT), which is related to the damage accumulation; (2) fracture strain (FS), which is always a function of stress triaxiality and Lode parameter; (3) fracture forming limit diagram (FFLD), which is always related to the principal strains. However, for an uncoupled damage model, its three forms (DT, FS and FFLD) for fracture prediction have never been compared before, and thus scholars have always been confused about how to choose the right form of the three in different loading processes. To clarify this confusion, two popular uncoupled damage models of the MMC4 and DF2016 are embedded into an anisotropic yield function first. Then the above-mentioned three forms of the two models are separately applied for fracture prediction in shear test, dog-bone specimen tensile test, hole specimen tensile test, notched specimen tensile test, bulging test in proportional loadings, and a complex non-proportional loading process of spin forming, respectively. Finally, the predicted results with these three forms are systematically compared and a new explanation for their applicability under different stress states is displayed.