AI-ENABLED MONITORING OPTIONS TOWARDS SECURE MICROREACTOR
DEPLOYMENT AND OPERATIONS

Year
2023
Author(s)
Pavel V. Tsvetkov - Texas A&M University
Anna S. Erickson - Georgia Institute of Technology
Piyush Sabharwall - Idaho National Laboratory (INL) BEA
Gustavo Reyes - Idaho National Laboratory
Mario Mendoza - Texas A&M University
Miguel Avalos - Texas A&M University
File Attachment
Abstract
Global nuclear energy deployment scenarios suggest favorable economics for smaller, more versatile, and self-contained reactor technologies. Recognizing their key features as integrated, autonomous and either semi-remotely operated or fully remotely operated systems, microreactors are expected to be deployed in large numbers servicing off/micro-grids, many in geographically remote locations. Integrated nature of these systems as well as ease of their transportation as complete units, simplified installation and relocation/decommissioning challenge traditional continuity-of-knowledge practices used for conventional light water reactors where refueling is done onsite at designated times only replacing portions of their cores and only after their full commissioning for operations including completion of their containment building with security and safeguards measures in place. This effort is exploring AI (artificial intelligence)-enabled monitoring options that would be design agnostic and would assure secure unit deployment and operations. The principle is to maintain situational awareness via real-time evaluations of simultaneous and remotely transmitted monitoring data capturing key safeguards attributes including such characteristics as temperature, radiation, vibrational signals (inter alia), and others. The key principle is to provide reliable and resilient security options while maintaining simplified and economical deployment. The paper will review feasible options for AI-enabled solutions for such evaluations.