SMART_TC: an R&D Programme on uses of artificial intelligence techniques for tritium monitoring in complex ITER-like tritium plant systems

The realization of nuclear fusion energy is nowadays based on the concept of tritium breeding and the success of the ITER experiment. The latter relies today on a static monitoring approach to fulfill the emission limits imposed by the regulatory institutions. Artificial intelligence applications fo...

ver descrição completa

Detalhes bibliográficos
Autores: Iraola de Acevedo, Eduardo|||0000-0002-0837-7086, Sedano Miguel, Luis Ángel|||0000-0002-3629-8297, Nougués, José María, Feliu, Josep Anton, Coya, Bruno, Batet Miracle, Lluís|||0000-0003-1882-6313
Formato: artículo
Fecha de publicación:2021
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/367794
Acesso em linha:https://hdl.handle.net/2117/367794
https://dx.doi.org/10.1016/j.fusengdes.2021.112409
Access Level:acceso abierto
Palavra-chave:Tritium
Artificial intelligence
Machine learning
ITER
Fault detection and diagnosis
Triti
Intel·ligència artificial
Aprenentatge automàtic
Àrees temàtiques de la UPC::Física
Descrição
Resumo:The realization of nuclear fusion energy is nowadays based on the concept of tritium breeding and the success of the ITER experiment. The latter relies today on a static monitoring approach to fulfill the emission limits imposed by the regulatory institutions. Artificial intelligence applications for fault diagnosis and process monitoring anticipate potential for the dynamic management of tritium in complex plant systems. This paper explores the dynamic tritium inventory management issue in complex systems, reviews the diverse artificial intelligence techniques and discusses the most promising approaches for ITER-like plant system match balance monitoring.