| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Elucidating Microstructural Evolution Under Extreme Environments
|
| Presentation Title |
Deep-Learning Driven Pt Particle Analysis for BWR Corrosion Insights |
| Author(s) |
Txai Sibley, Kevin Field, Elizabeth Holm |
| On-Site Speaker (Planned) |
Txai Sibley |
| Abstract Scope |
This study explores the application of a deep learning-driven image analysis model to identify platinum particles on components of Boiling Water Reactor (BWR) systems. These particles, introduced to mitigate stress corrosion through NMCA (noble metal chemical addition), play a crucial role in reactor part longevity via water chemistry control. Our research employs a segmentation model trained on a small image dataset to characterize platinum particles adhering to reactor surfaces. By correlating microstructural features with electrochemical potential (ECP), we can gain insights into corrosion behavior and streamline analysis of NMCA efficacy. This exploration provides a foundation for optimizing segmentation quality, minimizing the expense of data collection and annotation, and ultimately enhancing our understanding of the interplay between platinum particle characteristics and their impact on BWR nuclear reactor functionality. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Characterization, Machine Learning, Nuclear Materials |