Details

Autor(en) / Beteiligte

• Queral, C.
• Gómez-Magán, J.
• París, C.
• Rivas-Lewicky, J.
• Sánchez-Perea, M.
• Gil, J.
• Mula, J.
• Meléndez, E.
• Hortal, J.
• Izquierdo, J.M.
• Fernández, I.

Titel

Dynamic event trees without success criteria for full spectrum LOCA sequences applying the integrated safety assessment (ISA) methodology

Ist Teil von

• Reliability engineering & system safety, 2018-03, Vol.171, p.152-168

Ort / Verlag

Barking: Elsevier Ltd

Erscheinungsjahr

2018

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• •ISA methodology allows an accurate verification of the classical results of LOCA PSA analyses, such as success criteria, available times and damage exceedance frequency (DEF).•The dynamic event tree approach allows determining the success criteria variability with the break size.•The comparison between classical PSA and ISA methodology allows assessing the impact of classical PSA modeling assumptions.•The ISA application to Full Spectrum LOCA sequences shows that classical PSA methodologies significantly overestimate the LOCA DEF value. The integrated safety assessment (ISA) methodology, developed by the Spanish nuclear safety council (CSN), has been applied to the analysis of full spectrum loss of coolant accident (FSLOCA) sequences in a 3-loop pressurized water reactor (PWR). The ISA methodology proposal starts from the unfolding of the dynamic event tree (DET), focusing on the uncertainty of a reduced set of sequence parameters. Outcomes from this step allow assessing the sequence delineation of standard probabilistic safety analysis (PSA) results. For some sequences of interest of the outlined DET, the following ISA methodology steps involve the identification of the damage domain (DD). This is the region of main uncertain parameters space where a safety limit is exceeded during a given sequence. This analysis illustrates the application of this concept, based on transient simulations using MAAP. From the information obtained from the DDs, and considering the time-density probability distributions of human actions and stochastic phenomena occurrence, ISA integrates the dynamic reliability equations proposed to obtain each sequence contribution to the damage exceedance frequency (DEF). The study is then extended to include the uncertainty of subsidiary parameters and, finally, a comparison between the ISA methodology application to FSLOCA and the classical PSA methodology is established.