|About this Abstract
||2018 TMS Annual Meeting & Exhibition
||Algorithm Development in Materials Science and Engineering
||Simulation of multi-component microstructure evolution coupling
phase field and tensor decomposition techniques
||Yuan Yuan, Fusheng Pan, Nico Vervliet, Lieven Delathauwer, Nele Moelans
|On-Site Speaker (Planned)
Phase-field method is versatile and powerful for simulating the microstructural evolution in variety of material processes. However, simulations for multi-component systems often require large memory storage and long processing time due to huge data input and a lot of non-linear calculations for the thermodynamic properties. Therefore, a tensor decomposition approach has been introduced to conquer this problem.
A polyadic tensor decomposition method has been applied to both thermodynamic and kinetic properties of substitutional solutions, followed by applying the decomposed data in phase field simulations of homogenization process.
It is shown that the results obtained from Dictra simulations and those from phase field simulations using the original CALPHAD thermodynamic functions are identical. The phase field simulation results obtained using the tensor models are also consistent with those obtained using original thermodynamic functions. The effect of the rank of the tensor model on the accuracy of the simulation results was evaluated.
||Planned: Supplemental Proceedings volume