| About this Abstract |
| Meeting |
2023 TMS Annual Meeting & Exhibition
|
| Symposium
|
Advanced Characterization Techniques for Quantifying and Modeling Deformation
|
| Presentation Title |
Towards Data-driven In-Situ Materials Testing in SEM |
| Author(s) |
Fang Zhou |
| On-Site Speaker (Planned) |
Fang Zhou |
| Abstract Scope |
In-situ materials testing in SEM is widely used to link materials properties to the microstructures. Machine learning based approach is very promising for finding diverse sets of materials microstructures correlated to certain properties or materials behavior which can accelerate the materials modeling and materials design. However, the machine learning approach is very data hungry and a sophisticated data-driven in-situ materials testing workflow in SEM is hardly available so far. In this work, a well-integrated solution for demanding data-driven in-situ testing in SEM, combining high resolution surface sensitive SEM imaging and EDS/EBSD analytical methods with materials testing stages is introduced. Further advancements such as automated feature tracking, autofocus and multiple regions of interest (ROIs) enable true one-button-start workflows and data-driven experiments. Data-driven automated in-situ materials testing solutions are crucial for collecting reliable experimental datasets for materials modeling via machine learning and, thereafter, predicting materials behavior during the in-situ testing experiment. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Characterization, Mechanical Properties, Computational Materials Science & Engineering |