A digital twin of the AGN-201 reactor to simulate nuclear

Ryan Stewart - Idaho National Laboratory (INL) BEA
Ashley Shields - Idaho National Laboratory (INL) BEA
Chad Pope - Idaho State University
John Darrington - Idaho National Laboratory
Katherine Wilsdon - Idaho National Laboratory (INL) BEA
samuel Bays - Idaho National Laboratory (INL) BEA
Kolton Heaps - Idaho National Laboratory
Nathan Woodruff - Idaho State University
Gustavo Reyes - Idaho National Laboratory (INL) - BEA
Mark Schanfein - Idaho National Laboratory
Eduardo Trevino - Idaho National Laboratory
Jaden Palmer
Christopher Ritter - Idaho National Laboratory
File Attachment
Nonproliferation organizations must understand the potential proliferation pathways in a facility that could lead to weaponizable nuclear material. The International Atomic Energy Agency (IAEA) provides a source of confidence in the safe growth in global nuclear power through its assurance that states are continuing to abide by their international commitments while dampening the risk of undetected instances of proliferation. To succeed in this critical endeavor, the physics, design features, and proliferation indicators must be understood, and proliferation pathways mitigated. This requires a fundamental understanding of all aspects of an operating facility at the design level to conduct a diversion pathway analysis. Modeling breakthroughs, along with emerging technologies, offer a unique opportunity to develop digital twins (DTs) to accurately perform a diversion pathway analysis. Indicators can then be analyzed using artificial intelligence and machine learning engines, and the same DT can be turned into a real-time monitoring system. In the end, a DT is an important approach for nonproliferation knowledge retention and remains a foundation capability that any organization can maintain and expand to other nuclear fuel cycles over time. This paper focuses on the development of a DT of the Idaho State University AGN-201 reactor. The DT will be fed real-time data streams from an operating facility to detect early indicators in a timely manner so countermeasures can be taken. The AGN-201 virtual testbed will generate data representing activities at the physical testbed. This data will be used for algorithm development, virtual replay of campaigns, and prospective inspector training. It will also indicate which data feeds, both nuclear and non-nuclear, are most practical for successfully detecting errant reactor facility behavior. This will be a critical capability as the IAEA currently safeguards over 200 reactors around the world, a number expected to grow substantially in the near future. As it stands, 50 additional member states have expressed interest in pursuing nuclear power and roughly half of those states are in the pre-nuclear planning stage. If development of a digital twin is performed during the reactor design phase, the result can be the inclusion of timely safeguards by design features to ensure effective and efficient safeguards. Demonstrating a DT of a real reactor is the natural technology maturation stepping stone to a highpower reactor DT. Basic nuclear behaviors are similar at 1 W to behaviors at 1,000 MW. The basic process, previously demonstrated with a virtual DT of a sodium-cooled fast reactor, and the AGN-201 physical twin is highly tractable to a much larger scale for real-world reactors.