Development and Assessment of Data-Driven Digital Twins in a Nearly Autonomous Management and Control System for Advanced Reactors

Lin, Linyu; Rouxelin, Pascal; Athe, Paridhi; Lane, Jeff; Dinh, Nam

doi:10.1115/1.0000482v

Cited by 7 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To better extract and store knowledge, NAMAC uses the DT technology, where we rely on the power of data-driven methods and machine learning algorithms in expressing highly nonlinear behaviors and heterogenous information during nominal and accident transients [14]. In NAMAC, the DTs are defined as a knowledge acquisition system from the knowledge base to support the intended uses.…”

Section: Nearly Autonomous Management and Control Systemmentioning

confidence: 99%

Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system

Lin

Athe

Rouxelin

et al. 2022

Annals of Nuclear Energy

Self Cite

View full text Add to dashboard Cite

Section: Nearly Autonomous Management and Control Systemmentioning

confidence: 99%

Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system

Lin

Athe

Rouxelin

et al. 2022

Annals of Nuclear Energy

Self Cite

View full text Add to dashboard Cite

“…A DT is a digital representation of a physical asset or system that relies on real-time and history data for inferring complete reactor states, finding available control actions, predicting future transients, and identifying the most preferred actions [6]. During NPP operation, DTs can help decision-makers maintain an accurate understanding of the physical system, improve the effectiveness of operation, and avoid operator errors.…”

Section: Digital Twin Technologymentioning

confidence: 99%

“…Lin et al [1] developed a nearly autonomous management and control (NAMAC) system for restoring core flow during a partial loss-of-flow accident (LOFA) in the Experimental Breeder Reactor II (EBR-II) located at Idaho National Laboratory (INL). A class of DTs is developed with different algorithms for inferring the SSFs, finding available control actions, predicting consequences, and identifying the most preferred actions [6].…”

Section: Introductionmentioning

confidence: 99%

Uncertainty quantification and software risk analysis for digital twins in the nearly autonomous management and control systems: A review

Lin

Bao

Dinh

2021

Annals of Nuclear Energy

Self Cite

View full text Add to dashboard Cite

Time-aware deep reinforcement learning with multi-temporal abstraction

Kim

Chi

2023

Appl Intell

View full text Add to dashboard Cite

KIM, YEO JIN. Time-aware Deep Reinforcement Learning with Multi-Temporal Abstraction. (Under the direction of Dr. Min Chi).Reinforcement learning (RL) is an effective learning mechanism for handling sequential decisionmaking tasks. Specifically, deep reinforcement learning (DRL) has achieved great success in diverse complex tasks such as games and robotics. While a conventional DRL framework premises 1) complete data, 2) regular time intervals, and 3) full observations to induce an optimal policy, such premises often do not hold in many real-world tasks with offline irregular time series. In this dissertation, we tackled the three major challenges: 1) missing data, 2) temporal irregularity, and 3) partial observations in offline sequential data.First, heterogeneous features having different observation frequencies cause a high rate of missing data, which deteriorates the data quality and increases the uncertainty of decisions. To impute missing values, we propose Temporal Belief Memory (TBM), which is a bio-inspired missing data imputation method based on the temporal belief of observations. Our results show that TBM outperforms Missing Indicators, the state-of-the-art missing data handling method, on our septic shock prediction task.Second, real-world sequential data are often collected at irregular time intervals. Without consideration of continuous time intervals on discrete time steps, the agent is deficient to discern different temporal contexts and often fails to accurately estimate states and rewards. To address this issue, we propose a Time-aware DRL framework, which estimates states and rewards by explicitly representing time intervals between consecutive observations in the learning framework. We evaluated the proposed framework on two classic RL problems, Atari games, and two real-world tasks including nuclear reactor operation and septic shock prevention, and the results show the Time-aware DRL framework significantly improves performance in all the tasks. Third, in partially observable environments, a single view of temporal sequences generally limits observation experience and is biased to a specific series of observations, while a multi-view learns a more comprehensive representation than a single view and thus improves generalization performance. Also, in long-horizon decision making problems, temporal abstraction enables the agent to effectively learn long-range dependency by skipping over the high-frequency details. To take each advantage of multi-view and temporal abstraction, we propose a Multi-Temporal Abstraction (MTA) mechanism, which provides diverse temporal views over various periods of dependencies for state approximation. MTA provides three temporal modes: Retrospective, Concurrent, and Proactive modes, conditional on the temporal properties of target problems. Our results demonstrate that the Time-aware DRL framework with MTA (T-MTA) significantly increases training efficiency and effectiveness on two aforementioned real-world tasks.

show abstract

Development and Assessment of Data-Driven Digital Twins in a Nearly Autonomous Management and Control System for Advanced Reactors

Cited by 7 publications

References 0 publications

Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system

Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system

Uncertainty quantification and software risk analysis for digital twins in the nearly autonomous management and control systems: A review

Time-aware deep reinforcement learning with multi-temporal abstraction

Contact Info

Product

Resources

About