| About this Abstract |
| Meeting |
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Symposium
|
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Presentation Title |
Now On-Demand Only- Exploring the Role of Uncertainty Quantification in Thermodynamic Data and Models |
| Author(s) |
Noah H. Paulson, Joshua J Gabriel, Thien C. Duong, Marius Stan |
| On-Site Speaker (Planned) |
Noah H. Paulson |
| Abstract Scope |
The thermodynamic and kinetic properties of materials are the glue that binds together discrete elements of physics-based computational workflows for materials design. In recent decades, these composition, temperature, and pressure dependent properties have been encoded in semi-empirical expressions, as exemplified by the calculation of phase diagrams (CALPHAD) approach. Traditionally, these expressions are calibrated with experimental and/or atomistic simulation results without regard to the uncertainty introduced from data and models and propagated into further models (e.g. phase field modeling) and decision-making processes. The need for quantified uncertainty has gained increasing attention in the research community and triggered the interest of computational tool developers. We present recent Argonne efforts on uncertainty quantification in thermodynamic data and models using a variety of approaches spanning atomistic data, from machine-learned force fields to Bayesian methods for parameter inference, uncertainty quantification, and automated data weighting. |
| Proceedings Inclusion? |
Definite: Other |