About this Abstract |
Meeting |
2020 TMS Annual Meeting & Exhibition
|
Symposium
|
Materials Design Approaches and Experiences V
|
Presentation Title |
Strength Prediction in a Quaternary Mg Alloy System Using a Multi-scale Optimization Framework |
Author(s) |
Stephen J. DeWitt, Brian Puchala, Qianying Shi, Anirudh Raju Natarajan, Chaoming Yang, Anton Van der Ven, Liang Qi, John Allison |
On-Site Speaker (Planned) |
Stephen J. DeWitt |
Abstract Scope |
When developing a multi-scale model for a material property, many of the steps performed are similar regardless of specific material property or system. These steps can be loosely classified as steps fitting existing data (e.g. with machine learning), steps deciding where new data should be collected, steps linking model components with each other or with data repositories, and steps deploying the multi-scale model (e.g. optimization, inverse parameter estimation, uncertainty quantification). We have created a new open-source Python package, called prisms.multiscale, that performs these common operations. Here, we describe the use of prisms.multiscale for strength prediction in a quaternary Mg alloy system. This optimization problem involves the integration of first-principles statistical mechanics calculations (CASM), a thermodynamic database (ThermoCalc), KWN simulations (TC-PRISMA), dislocation dynamics simulations (ParaDiS), and a solute strengthening database. We demonstrate the effects of alloy composition and aging time on the alloy strength and validate the predictions against experiments. |
Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |