SIGNAL GROUPING FOR CONDITION MONITORING OF NUCLEAR POWER PLANT COMPONENTS
| Autor: | Baraldi, P., Canesi, R., Enrico Zio, Seraoui, R., Chevalier, R. |
|---|---|
| Přispěvatelé: | Dipartimento di Energia [Milano] (DENG), Politecnico di Milano [Milan] (POLIMI), Chaire Sciences des Systèmes et Défis Energétiques EDF/ECP/Supélec (SSEC), Ecole Centrale Paris-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-CentraleSupélec-EDF R&D (EDF R&D), EDF (EDF)-EDF (EDF), Laboratoire Génie Industriel - EA 2606 (LGI), CentraleSupélec, EDF (EDF) |
| Jazyk: | angličtina |
| Rok vydání: | 2011 |
| Předmět: | |
| Zdroj: | Reliability and Risk Management ESREL 2011 Advances in Safety, Reliability and Risk Management: ESREL 2011 Advances in Safety, Reliability and Risk Management: ESREL 2011, Sep 2011, France. pp.1-13 Scopus-Elsevier |
| Popis: | International audience; The present work investigates the possibility of building a condition monitoring model by splitting the usually very large number of signals measured by the sensors into subgroups and building a specialized model for each subgroup. Different criteria are considered for selecting the signal groups, such as the location of the measurements (i.e., signals measured in the same area of the plant belong to the same group) and their correlation (i.e., correlated signals are grouped together). A real case study concerning 48 signals selected between those used to monitor the reactor coolant pump of a Pressurized Water Reactor (PWR) is considered in order to verify the monitoring performance of different grouping criteria. Performance metrics measuring accuracy, robustness and spill-over effect have been considered in the evaluation. Key Words: Condition Monitoring, Empirical Modeling, Power Plants, Safety Critical Nuclear Instrumentation, Autoassociative models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|