| Popis: |
Simulation of fracture in ductile materials is a challenging problem, since it typically occurs at length scales that are orders of magnitude smaller than that of the structures in which the fracture is occurring and, hence, difficult to resolve . One approach is to avoid modelling the micro-mechanics of ductile fracture by describing the macroscopic effects of fracture using damage parameters. Damage in metals can be defined as a measure of discontinuous deformation of a body. Many numerical models include some measure of damage to predict when a material will fracture under certain conditions, however there is little consensus as to what measures and parameters will accurately predict the onset of fracture. Most notably, the effect of strain rate at the point of fracture is significant and must be taken into account. The literature indicates that in the quasistatic regime where inertial effects are negligible, an increase in strain rate increases the strain at fracture. However, the research conducted in this dissertation suggests the opposite is true. The aim of this research is to conduct further high strain rate ductile damage experiments so as to extend the available data set, and develop a pragmatic damage model to relate the plastic strain at fracture to material parameters such as triaxiality, lode angle and strain rate in a specimen, which is verified using experiments performed under various loading conditions and strain rates. |
