| About this Abstract |
| Meeting |
2023 TMS Annual Meeting & Exhibition
|
| Symposium
|
Algorithm Development in Materials Science and Engineering
|
| Presentation Title |
Development of Structure-property Linkages for Damage in Crystalline Microstructures Using Bayesian Inference and Unsupervised Learning |
| Author(s) |
David Montes De Oca Zapiain, Anh Tran, Hojun Lim |
| On-Site Speaker (Planned) |
David Montes De Oca Zapiain |
| Abstract Scope |
Crystal Plasticity Finite Element Method (CPFEM) is a robust tool that accurately captures the effects the orientation of the crystal lattice has on the deformation of a material, which can be used to assess damage performance. Nevertheless, the high computational cost of CPFEM makes intractable the usage of this technique in industry. Therefore, there is a critical need for an accurate and computationally-efficient linkage between the internal crystalline texture of a material and its damage performance. This work addresses this critical need by developing an accurate reduced-order model capable of predicting the damage performance of a material from its internal crystalline texture. Furthermore, this work leverages Bayesian inference to optimally select the crystallographic orientations to build an accurate linkage given the fact that each training point requires the evaluation of an expensive high-fidelity CPFEM-simulation.
SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.SAND No: SAND2022-8914 A |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Mechanical Properties, Machine Learning |