| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Bridging Scale Gaps in Multiscale Materials Modeling in the Age of Artificial Intelligence
|
| Presentation Title |
Realizing high-throughput multi-scale simulations of materials through machine learning |
| Author(s) |
Shyue Ping Ong |
| On-Site Speaker (Planned) |
Shyue Ping Ong |
| Abstract Scope |
A fundamental challenge in computational materials science is bridging length and time scales from quantum to atomistic to continuum. In this talk, I will present a roadmap to attacking this challenge in an automated fashion through machine learning (ML). In the past decade, ML interatomic potentials (MLPs) have emerged as a robust approach to bridge the quantum and atomistic scales by learning the potential energy surface from high-throughput ab initio calculations. In turn, simulations from MLPs can be used to learn statistical parameters for continuum scale models. I will illustrate this process using several prototypical examples from multi-principal element alloys and solid ceramic electrolytes for lithium-ion batteries. Finally, I will outline the major gaps that remain to the realization of high-throughput multi-scale simulations of materials. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Mechanical Properties |