About this Abstract |
Meeting |
2022 TMS Annual Meeting & Exhibition
|
Symposium
|
Materials Design and Processing Optimization for Advanced Manufacturing: From Fundamentals to Application
|
Presentation Title |
New Methodologies for Grain Boundary Detection in EBSD Data of Microstructures |
Author(s) |
Richard Catania |
On-Site Speaker (Planned) |
Richard Catania |
Abstract Scope |
This work discusses new methodologies for identifying the grain boundaries in color images of metallic microstructures and the quantification of their grain topology. This work employs the experimental microstructure data of Titanium-Aluminum alloys, which are used for various aerospace components, owing to their mechanical performance in elevated temperatures. The grain topology of these metallic microstructures is quantified using the concept of shape moment invariants (SMI). First, it is necessary to identify the grain boundaries and separate them from their respective grains. The two methods this work investigates are tolerance-based neighbor analysis and Euclidean distance analysis to separate different grains. Additionally, since the grain boundaries do not possess the same material properties as the grain itself, this work investigates the effect of including the grain boundaries when determining the overall microstructure properties. Once the topology is quantified, a machine learning algorithm will be used to obtain material properties from the SMI. |
Proceedings Inclusion? |
Planned: |
Keywords |
Mechanical Properties, Machine Learning, Titanium |