ML24276A001

From kanterella
Jump to navigation Jump to search
Regulatory Implications of Advanced Technologies for Component Condition Monitoring in Nuclear Energy Systems
ML24276A001
Person / Time
Issue date: 10/02/2024
From: Raj Iyengar
NRC/RES/DE
To:
Raj Iyengar 301-415-0770
References
Download: ML24276A001 (11)


Text

Regulatory Implications of Advanced Technologies for Component Condition Monitoring in Nuclear Energy Systems Raj Iyengar Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission The views expressed are those of the authors and do not reflect the views of the U.S. Nuclear Regulatory Commission. This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for public release; distribution is unlimited.

OBJECTIVES

  • Demonstrate the application of digital twins (DT) for mechanical system condition monitoring in nuclear power plants (NPPs)
  • Relevance to ASME OM and OM-2 code requirements
  • Gain knowledge and confidence
  • Modeling and simulation for digital twins
  • Obtain insights
  • Support future regulatory application reviews and develop guidance, as needed

DIGITAL TWIN SYSTEM - NPP SCENARIO

  • Four main elements to an NPP-DT system
1. An NPP
2. A Digital Twin (DT)
3. Sensor and performance data flowing from NPP to DT
4. Actions and recommendations flowing from DT to NPP

DIGITAL TWIN CHARACTERISTICS & CAPABILITIES

ANOMALY DETECTION CASE STUDY: Simulator synthetic data

  • Full scope BWR: Represents the full set of plant components and operations
  • Physics-based model: Parameters are calculated using first principles
  • Real-time data: Operates in real time just like an actual plant
  • Initial conditions: Can set the initial conditions of the plant
  • Malfunctions: Variety of malfunctions that can be inserted at a specified time and severity
  • Output: Records time-series data of selected plant parameters and malfunctions Probabilistic Safety Assessment and Management (PSAM) 2023 Topical Conference on Artificial Intelligence (AI) and Risk Analysis, October 2325, 2023.

https://www.ideals.illinois.edu/items/128985

ANOMALY DETECTION CASE STUDY: Trained LSTM autoencoder Probabilistic Safety Assessment and Management (PSAM) 2023 Topical Conference on Artificial Intelligence (AI) and Risk Analysis, October 2325, 2023.

https://www.ideals.illinois.edu/items/128985

ANOMALY DETECTION CASE STUDY: Identified inserted anomalies Probabilistic Safety Assessment and Management (PSAM) 2023 Topical Conference on Artificial Intelligence (AI) and Risk Analysis, October 2325, 2023.

https://www.ideals.illinois.edu/items/128985 Malfunctions Modeled Recirc Pump A Runaway Recirc Pump B Runaway Recirc Pump A Seal #1 Failure Recirc Pump B Seal #1 Failure Recirc Pump A Seal #2 Failure Recirc Pump B Seal #2 Failure RBCCW Heat Exchanger Tube Leak

IST AND CONDITION MONITORING CASE STUDIES

  • Case studies for condition monitoring of mechanical systems and components
  • Heat Pipe
  • AI/ML explainability
  • Influence of input features to model output INL/EXT-20-60782 INL/EXT-17-43212 Insights from these case studies will help prepare NRC staff to evaluate applications of condition monitoring to meet IST requirements

DIGITAL TWINS PROJECT ACCOMPLISHMENTS 8

6

  • 6 Technical Letter Reports
  • 2 Public workshops
  • Ongoing research Future Focused Research Project ML22192A046

Technical Letter Reports ML23058A085 ML22235A643 ML21361A261

ML21160A074

ML21348A020

ML2108A132 Information Letters Follow-on Report ML23271A055

ML24065A049

INNOVATION AT THE NRC:

Digital Twins for Condition Monitoring

THANK YOU https://www.nrc.gov/reactors/power/digital-twins.html