Introduction of material length scales through damage percolation modelling

摘要

Ductile fracture prediction considers the nucleation, growth and coalescence of micro-voids leading to fracture. A major weakness of continuum-level finite element predictions of micro-void damage is a lack of material length scale within damage-based constitutive models. One technique to introduce material length scale at the micro-structural level utilises a damage percolation approach. Measured second phase particle fields are acquired using digital image analysis and tessellated to determine nearest neighbour and particle clustering statistics. Void nucleation and growth models are then applied at the individual void level which consider particle/void size and orientation, as well as cluster membership, to determine damage rates. Coalescence or linking is considered between individual voids as well as between clusters of voids to form cracks. Damage is shown to develop in a stable fashion until large strains are reached. The critical event appears to be the coalescence of two or more cracks (groups of coalesced voids) at which point a chain reaction ensues with catastrophic void-crack linkage leading to fracture.