Dynamic PRA Methods to Evaluate the Impact on Accident Progression of Accident Tolerant Fuels

被引:8
作者
Mandelli, Diego [1 ]
Parisi, Carlo [1 ]
Anderson, Nolan [1 ]
Ma, Zhegang [1 ]
Zhang, Hongbin [1 ]
机构
[1] Idaho Natl Lab, 1955 Fremt Ave, Idaho Falls, ID 83415 USA
关键词
Accident tolerant fuel; dynamic probabilistic risk assessment; probabilistic risk assessment; RELIABILITY;
D O I
10.1080/00295450.2020.1794234
中图分类号
TL [原子能技术]; O571 [原子核物理学];
学科分类号
0827 ; 082701 ;
摘要
Accident tolerant fuels (ATFs) are new nuclear fuels developed in response to the accident at the Fukushima power station in March 2011. The goal of ATFs is to withstand accident scenarios through better performance compared to currently employed fuels (e.g., small-scale hydrogen generation). This paper targets a method for evaluating and comparing ATF performance from a probabilistic risk assessment (PRA) perspective by employing a newly developed combination of event trees and dynamic PRA methods. Compared to classical PRA methods based on event trees and fault trees, dynamic PRA can evaluate with higher resolution the safety impacts of physics dynamics and the timing/sequencing of events on the accident progression without the need to introduce overly conservative modeling assumptions and success criteria. In this paper, we analyze the impact on the accident progression of three different cladding configurations for two initiating events [a large break loss-of-coolant accident (LB-LOCA) and a station blackout (SBO)] by employing dynamic PRA methods. The goal is to compare the safety performance of ATFs (FeCrAl and Cr-coated cladding) and the currently employed Zr-based clad fuel. We employ two different strategies. The first focuses on the identification of success criteria discrepancies between the accident sequences generated by the classical PRA model and the set of simulation runs generated by dynamic PRA using ATF. The second one, on the other hand, directly uses dynamic PRA to evaluate the impact of timing of events (e.g., recovery actions) on accident progression. By applying these methods to the LB-LOCA and SBO initiating events, we show how dynamic PRA methods can provide analysts with detailed and quantitative information on the safety impact of ATFs.
引用
收藏
页码:389 / 405
页数:17
相关论文
共 28 条
[1]   REDUCED ORDER MODELING FOR NONLINEAR MULTI-COMPONENT MODELS [J].
Abdel-Khalik, Hany S. ;
Bang, Youngsuk ;
Kennedy, Christopher ;
Hite, Jason .
INTERNATIONAL JOURNAL FOR UNCERTAINTY QUANTIFICATION, 2012, 2 (04) :341-361
[2]  
ALDEMIR T., 1991, P PROBABILISTIC SAFE
[3]  
ALFONSI A., 2014, P EUR SAF REL C ESRE, DOI [10.1201/b17399-106, DOI 10.1201/B17399-106]
[4]  
[Anonymous], 2005, NUREGCR1150 US NUCL
[5]  
[Anonymous], 2012, NUREGCR6753 SAND NAT
[6]  
[Anonymous], 2012, 1206 NEI
[7]   A review of nuclear fuel performance codes [J].
Aybar, HS ;
Ortego, P .
PROGRESS IN NUCLEAR ENERGY, 2005, 46 (02) :127-141
[8]  
BORING R.L, 2014, INLEXT1433903
[9]   The DYLAM approach for the dynamic reliability analysis of systems [J].
Cojazzi, G .
RELIABILITY ENGINEERING & SYSTEM SAFETY, 1996, 52 (03) :279-296
[10]   SUPPORT-VECTOR NETWORKS [J].
CORTES, C ;
VAPNIK, V .
MACHINE LEARNING, 1995, 20 (03) :273-297