About this Abstract |
Meeting |
2021 TMS Annual Meeting & Exhibition
|
Symposium
|
Advanced Characterization Techniques for Quantifying and Modeling Deformation
|
Presentation Title |
Grain Boundary Slip Transfer Classification and Metric Selection with Artificial Neural Networks |
Author(s) |
Zhuowen Zhao, Thomas R. Bieler, Javier LLorca, Philip Eisenlohr |
On-Site Speaker (Planned) |
Zhuowen Zhao |
Abstract Scope |
The accurate prediction of the interaction between dislocation slip and grain boundaries is a long-standing challenge in the field of crystal plasticity. An artificial neural network (ANN) is used to evaluate the effectiveness of six metrics and their combinations to assess whether instances of slip transfer happen across grain boundaries in coarse-grained oligocrystalline Al foils. This approach extends the one- or two-dimensional projections that were formerly applied to analyze slip transfer data. We observe that the maximum attainable classification accuracy is limited to about 90% and does not substantially increase between low-dimensional projections and the full six-dimensional picture. The most effective metrics reflect the geometric relationship between grains sharing a boundary. |
Proceedings Inclusion? |
Planned: |
Keywords |
Machine Learning, Computational Materials Science & Engineering, |