|About this Abstract
||2017 TMS Annual Meeting & Exhibition
||Defects and Properties of Cast Metals
||Modelling the Effects of Fluid Flow on Microstructure Evolution at the Component Scale
||Matthaios Alexandrakis, Andrew Kao, Koulis Pericleous
|On-Site Speaker (Planned)
Predicting the formation of defects and macroscopic material properties of components is very important to modern industry. Many factors affect the formation of defects including fluid flow and the geometry of the component. Through increases in high performance computing (HPC), the ability to predict these features on the component level, but from a microstructural perspective is becoming a reality. This work presents a massively parallel implementation of directional solidification using a Cellular Automata (CA) based method that employs a modified decentred octahedron method. In contrast to phase field methods, the CA method allows the use of larger cell sizes (O10microns), compared to (O1micron) and coupled with massive parallelisation allows for the simulation of billions of computational cells, approaching the scale of small components. The model has been coupled to both an electromagnetic and fluid flow solvers that has allowed the investigation of novel fluid effects such as Thermoelectric Magnetohydrodynamics.
||Planned: Supplemental Proceedings volume