ML25206A071

From kanterella
Jump to navigation Jump to search
Advancing IST with Digital Twins and Condition Monitoring - OM Symposium 2025 July
ML25206A071
Person / Time
Issue date: 07/25/2025
From: Ricardo Torres
NRC/NRR/DEX/EMIB
To:
References
Download: ML25206A071 (1)


Text

Advancing IST with Digital Twins and Condition Monitoring Roberto L Torres Davis Office of Nuclear Reactor Regulation, DEX Mechanical Engineering and Inservice Testing Branch ASME/NRC OM Symposium 2025 1

Disclaimer This presentation was prepared by staff of the U.S. Nuclear Regulatory Commission (NRC). It may present information that does not currently represent an agreed upon NRC staff position. NRC has neither approved nor disapproved the technical content.

2

NRC on Digital Twins 3

  • NRC staff are actively evaluating Digital Twin (DT) technologies as part of our broader assessment of advanced condition-based monitoring (CBM) for inservice testing (IST).
  • Regulatory guides such as RG 1.192 (OMN-29) already support technology-inclusive, performance-based IST strategies.

What Were Exploring 4

  • NRC Office of Nuclear Regulatory Research has evaluated hypothetical case studies using synthetic data and simulation tools:
  • Reactor Coolant Pump (RCP): Simulated thermal barrier leak detection with machine learning-enabled Digital Twins.
  • Heat Pipe Monitoring: Simulated passive component diagnostics in advanced reactor designs (e.g., microreactors).
  • These are not field-deployed but help us understand potential benefits, limitations, and code implications.

Key Challenges Under Review

  • VVUQ (Verification, Validation, and Uncertainty Quantification) for AI/ML models.
  • Sensor integrity and long-term reliability in nuclear environments.
  • Model explainability to support transparency and regulatory confidence.

5 Source: Adadi, A., & Berrada, M. (2018). Peeking inside the black-box: A survey on explainable artificial intelligence (XAI).

IEEE Access, 6, 52138-52160.

ASME Code Implications

  • OM-2 Code supports performance-based IST - ready for DT/CBM
  • Define: sensor criteria, model metrics, VVUQ expectations 6

Recommendations

  • Develop guidance on integrating DTs into IST programs.
  • Support pilot projects to benchmark performance and understand implementation gaps.
  • Form a working group to coordinate DT applications across OMN-29, OM-2, and future OM Code initiatives.

7 Image by pch.vector on Freepik

QUESTIONS?

8