Fuzzy Uncertainty Evaluation for Fault Tree Analysis
- Authors
- Kim, Ki Beom; Shim, Hyung Jin; Jae, Moo Sung
- Issue Date
- May-2015
- Publisher
- 한국원자력학회
- Keywords
- COMPUTER CALCULATIONS; DAMAGE; ERRORS; FAULT TREE ANALYSIS; FUZZY LOGIC; LOSS OF COOLANT; PROBABILITY
- Citation
- 한국원자력학회 2015춘계학술발표회논문집, pp.1 - 3
- Indexed
- OTHER
- Journal Title
- 한국원자력학회 2015춘계학술발표회논문집
- Start Page
- 1
- End Page
- 3
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/157158
- Abstract
- This traditional probabilistic approach can calculate relatively accurate results. However it requires a long time because of repetitive computation due to the MC method. In addition, when informative data for statistical analysis are not sufficient or some events are mainly caused by human error, the probabilistic approach may not be possible because uncertainties of these events are difficult to be expressed by probabilistic distributions. In order to reduce the computation time and quantify uncertainties of top events when basic events whose uncertainties are difficult to be expressed by probabilistic distributions exist, the fuzzy uncertainty propagation based on fuzzy set theory can be applied. In this paper, we develop a fuzzy uncertainty propagation code and apply the fault tree of the core damage accident after the large loss of coolant accident (LLOCA). The fuzzy uncertainty propagation code is implemented and tested for the fault tree of the radiation release accident. We apply this code to the fault tree of the core damage accident after the LLOCA in three cases and compare the results with those computed by the probabilistic uncertainty propagation using the MC method. The results obtained by the fuzzy uncertainty propagation can be calculated in relatively short time, covering the results obtained by the probabilistic uncertainty propagation.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 원자력공학과 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.