Fracture Predictions Using Phase Field Models for Tuned Random Heterogeneous Surrogate Microstructures
Please login to view abstract download link
Microstructures significantly influence the effective material properties of engineering materials, making it crucial to incorporate microstructure and anisotropy into fracture and fatigue modeling of materials and structures. Previous experiments and observations have demonstrated the impact of anisotropy and microstructure on crack nucleation, propagation and failure. However, the coupling of fracture and fatigue properties with the statistics of complex microstructures remains an unexplored area. We use a Gaussian random field-based microstructure generator to create anisotropic microstructures that can be tailored to real-world samples through a stochastic optimisation process. These random microstructures are then considered in the context of fracture problems, where fracture properties are computed using a phase field approach capable of predicting crack initiation and propagation. This stochastic model is used for the propagation of the microstructure uncertainties to make robust failure predictions.
