Integrated Computational Materials Engineering in Multiphysics Software

Our team will investigate a multiscale modeling framework to predict the final microstructure of a dual phase Ti 6-4 component as a function of thermo-mechanical processing. The microstructure properties will vary locally throughout the part, and will include features such as grain size, phase volume fraction, morphology of grains and phases, and the data structure representations will be scalable and expandable as the need for future microstructure modeling descriptors arise. The local evolution of microstructural features will be tightly coupled to the computation of a component wide process simulation, and will be based upon fundamental physics, not simple phenomenological models. The proposed work will investigate expanding existing single-phase physics based constitutive models into dual-phase models for titanium alloys. Crystallographic texture information will be stored locally at each element of the FEM simulation, and will evolve via physics-based texture evolution models. Typical crystal plasticity formulations including the Taylor model and the Visco Plastic Self Consistent (VPSC) model will be studied in Phase I. A method to represent crystal texture at every element within a macroscopic component utilizing the Rodrigues texture representation will be used. Texture representations that consider grain size and morphology (Grain Size Orientation Distribution Function) will also be investigated.