About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Late News Poster Session
|
| Presentation Title |
K-13: Rapid AI-Driven Acoustic Inspection of Advanced Nuclear Fuel Pebbles |
| Author(s) |
Adrien J. Terricabras, Do-Kyung Pyun, Sangmin Lee, Rajendra Palanisamy, Jacqe Jansen van Vuuren, Danie Jacobs, Renoux Kritzinger, Timothy Coons, Alp Findikoglu |
| On-Site Speaker (Planned) |
Adrien J. Terricabras |
| Abstract Scope |
Graphite pebble composite structures containing TRistructural-ISOtropic (TRISO) particles are being developed as advanced nuclear fuels capable of safe operation at elevated temperatures. Ensuring their structural integrity requires robust non-destructive evaluation (NDE) methods, yet conventional acoustic techniques are limited by the highly attenuative and compositionally complex nature of these materials. This study presents an improved acoustic NDE approach for accurate detection and classification of defects in graphite pebbles using high-resolution acoustic signals and optimized sensor networks. A triangular three-sensor configuration expands inspection from a linear path to a two-dimensional region, enhancing spatial coverage and resolution. Excitation parameters are optimized to maximize signal-to-noise ratio. A machine-learning-based algorithm extracts multi-domain features from the resulting signals, enabling superior defect identification and classification. The proposed method demonstrates strong potential for efficient, reliable diagnostics of next-generation nuclear pebble fuels. |
| Proceedings Inclusion? |
Undecided |
| Keywords |
Characterization, Nuclear Materials, Ceramics |