First-of-a-Kind Risk-Informed Digital Twin for Operational Decision Making

Muhlheim, Michael D.; Ramuhalli, Pradeep; Guler Yigitoglu, Askin; Huning, Alex; Wood, Richard; Narvaez, Jorge L.; Saxena, Abhinav

doi:10.1080/00295639.2025.2471724

First-of-a-Kind Risk-Informed Digital Twin for Operational Decision Making

Journal Article · Wed May 14 00:00:00 EDT 2025 · Nuclear Science and Engineering

DOI:https://doi.org/10.1080/00295639.2025.2471724· OSTI ID:2586968

^[1]; Ramuhalli, Pradeep ^[1]; Guler Yigitoglu, Askin ^[1]; Huning, Alex ^[1]; Wood, Richard ^[2]; Narvaez, Jorge L. ^[2]; Saxena, Abhinav ^[3]

Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Univ. of Tennessee, Knoxville, TN (United States)
GE Research US, Niskayuna, NY (United States)

A digital twin (DT) is a digital model or a collection of models of a physical entity. DTs in the nuclear arena can be used from plant design through decommissioning. Decisions are typically a priori or made offline. Risk-informed decision making is identifying what can go wrong, its frequency, and the consequences of its failure. Ideally risk-informed decision making reflects the current state of the plant and provides a decision in real time. Traditionally, probabilistic risk assessments (PRAs) evaluate the failures of safety systems, the risk of core damage, and the offsite dose as the consequence. However, this DT evaluates the decisions on the control side rather than the protection side. It uses the same risk methods to probabilistically inform the decision-making process but in a different way. Rather than evaluating the risk of core damage, this DT evaluates the likelihood of avoiding a trip set point while maintaining plant safety. Performance-based assessments are identified via its probabilistic evaluation of operational alternatives based on system status. Because the purpose of the control system is to maintain system variables within prescribed operating ranges, upsets or challenges that can exceed a trip set point resulting in a plant transient and a challenge to plant mitigating systems based on actual plant conditions, are evaluated to safely maintain the plant within the operating ranges. The probabilistic portion of the model is autonomously and automatically adjusted, and the metric of interest (i.e. likelihood of avoiding a trip set point) is recalculated. The digital representation of the physical system (i.e. the DT) performs a deterministic performance–based assessment of the probabilistically identified alternatives identified to validate the probabilistic assessment. A decision-making algorithm selects the appropriate option based on the probabilistic and deterministic assessments and transmits a control signal to a component(s) to initiate a corrective action or informs an operator of its decision.

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE Advanced Research Projects Agency - Energy (ARPA-E)

Grant/Contract Number:: AC05-00OR22725; AR0001290

OSTI ID:: 2586968

Journal Information:: Nuclear Science and Engineering, Journal Name: Nuclear Science and Engineering Journal Issue: 11 Vol. 199; ISSN 0029-5639; ISSN 1943-748X

Publisher:: Informa UK LimitedCopyright Statement

Country of Publication:: United States

Language:: English

References (9)

Digital Twin: Generalization, characterization and implementation VanDerHorn, Eric; Mahadevan, Sankaran Decision Support Systems, Vol. 145 https://doi.org/10.1016/j.dss.2021.113524	journal	June 2021
Digital twin in energy industry: Proposed robust digital twin for power plant and other complex capital-intensive large engineering systems Sleiti, Ahmad K.; Kapat, Jayanta S.; Vesely, Ladislav Energy Reports, Vol. 8 https://doi.org/10.1016/j.egyr.2022.02.305	journal	November 2022
Data fusion and machine learning for industrial prognosis: Trends and perspectives towards Industry 4.0 Diez-Olivan, Alberto; Del Ser, Javier; Galar, Diego Information Fusion, Vol. 50 https://doi.org/10.1016/j.inffus.2018.10.005	journal	October 2019
The application of machine learning for the prognostics and health management of control element drive system Oluwasegun, Adebena; Jung, Jae-Cheon Nuclear Engineering and Technology, Vol. 52, Issue 10 https://doi.org/10.1016/j.net.2020.03.028	journal	October 2020
Advances of Digital Twins for Predictive Maintenance You, Yingchao; Chen, Chong; Hu, Fu Procedia Computer Science, Vol. 200 https://doi.org/10.1016/j.procs.2022.01.348	journal	January 2022
Machinery health prognostics: A systematic review from data acquisition to RUL prediction Lei, Yaguo; Li, Naipeng; Guo, Liang Mechanical Systems and Signal Processing, Vol. 104 https://doi.org/10.1016/j.ymssp.2017.11.016	journal	May 2018
A particle filtering-based framework for real-time fault diagnosis and failure prognosis in a turbine engine Orchard, Marcos E.; Vachtsevanos, George J. 2007 Mediterranean Conference on Control & Automation https://doi.org/10.1109/MED.2007.4433871	conference	June 2007
Prognostics-enhanced Automated Contingency Management for advanced autonomous systems Tang, Liang; Kacprzynski, Gregory J.; Goebel, Kai 2008 International Conference on Prognostics and Health Management https://doi.org/10.1109/PHM.2008.4711448	conference	October 2008
Optimum Settings for Automatic Controllers Ziegler, J. G.; Nichols, N. B. Journal of Dynamic Systems, Measurement, and Control, Vol. 115, Issue 2B https://doi.org/10.1115/1.2899060	journal	June 1993

Similar Records

Integrated Risk-Informed Decision-Making for an ALMR PRISM

Technical Report · Sun May 01 00:00:00 EDT 2016 · OSTI ID:1329123

Development of a First-of-a-Kind Deterministic Decision-Making Tool for Supervisory Control System

First-of-a-Kind Risk-Informed Digital Twin for Operational Decision Making

Citation Formats

References (9)

Similar Records

Related Subjects