Automated state identification systems facilitate reactor monitoring and control of nuclear systems by consolidating information collected by deployed sensors. In the current paper, we present the use of relevance vector machines (RVM) for real-time state identification of boiling water reactors (BWR). In particular, RVM models utilize the incoming signals of interest and identify in real time the state of the BWR either as normal or as one of the transition states. Each of the RVM models is assigned to a single signal; it receives the measured value at each instance and outputs the identified BWR state. The state that has been designated by the majority of the signals is displayed to the human operator as the identified BWR state. The proposed methodology is applied and tested on a set of signals taken from the FIX-II experimental facility that is a scaled representation of a BWR.

Real-time state identification of boiling water reactors using relevance vector machines

Cappelli M.
2016-01-01

Abstract

Automated state identification systems facilitate reactor monitoring and control of nuclear systems by consolidating information collected by deployed sensors. In the current paper, we present the use of relevance vector machines (RVM) for real-time state identification of boiling water reactors (BWR). In particular, RVM models utilize the incoming signals of interest and identify in real time the state of the BWR either as normal or as one of the transition states. Each of the RVM models is assigned to a single signal; it receives the measured value at each instance and outputs the identified BWR state. The state that has been designated by the majority of the signals is displayed to the human operator as the identified BWR state. The proposed methodology is applied and tested on a set of signals taken from the FIX-II experimental facility that is a scaled representation of a BWR.
2016
978-0-7918-5001-5
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11697/242040
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
social impact