Profile based stochastic hybrid automata for a training use with external user
Résumé
Electrical energy is one of the most closely watched topics due to its growing need and its current and future issues. To cope, France has a nuclear fleet of 56 reactors [1,2]. To manage this fleet and to train operators, high performance tools are needed to both maintain and improve the know how of operating these reactors. To help preserve and improve this technical knowledge, a project has been launched, regrouping many French actors in the nuclear field, the “Digital Reactor” (“Réacteur Numérique”, in French) [3]. The objective of the "Digital Reactor" project is to create a digital twin of a reactor of a nuclear power plant to fulfil 3 main goals, being: a training tool; a following tool; and a tool to anticipate the future. In this context, research work is being carried out at CRAN to develop a dysfunctional model that considers failures, degradation, and ageing of systems.