ID:
publications-4905
Type:
Conference paper
Year:
2024
Authors:
Ross J.; Collins B.; Clarno K.
Title:
State Estimation and Control in Digital Twins for TRIGA Research Reactors
Venue/Journal:
Proceedings of the International Conference on Physics of Reactors, PHYSOR 2024
DOI:
10.13182/PHYSOR24-43709
Research type:
Water System:
Technical Focus:
Abstract:
An initial study was conducted to test the viability of digital twins in TRIGA research reactors for state estimation and automated control.The tranport code MPACT was adopted as a high order model, while point kinetics and a novel LSTM function served as a physics-enhanced machine learning method for the low order model.The PK-LSTM workflow was able to predict the flux at a specific location with an rms error of 1.3E-2 over an entire transient.Additionally, the low order model was used to create a pseudo-optimal control scheme with an rms error of 1.8E-2 for flux step changes. Β© 2024 AMERICAN NUCLEAR SOCIETY. All rights reserved.
Link with Projects:
Link with Tools:
Related policies:
ID:
