skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Statistically based uncertainty analysis for ranking of component importance in the thermal-hydraulic safety analysis of the Advanced Neutron Source Reactor

Technical Report ·
DOI:https://doi.org/10.2172/5485664· OSTI ID:5485664

The Analytic Hierarchy Process (AHP) has been used to help determine the importance of components and phenomena in thermal-hydraulic safety analyses of nuclear reactors. The AHP results are based, in part on expert opinion. Therefore, it is prudent to evaluate the uncertainty of the AHP ranks of importance. Prior applications have addressed uncertainty with experimental data comparisons and bounding sensitivity calculations. These methods work well when a sufficient experimental data base exists to justify the comparisons. However, in the case of limited or no experimental data the size of the uncertainty is normally made conservatively large. Accordingly, the author has taken another approach, that of performing a statistically based uncertainty analysis. The new work is based on prior evaluations of the importance of components and phenomena in the thermal-hydraulic safety analysis of the Advanced Neutron Source Reactor (ANSR), a new facility now in the design phase. The uncertainty during large break loss of coolant, and decay heat removal scenarios is estimated by assigning a probability distribution function (pdf) to the potential error in the initial expert estimates of pair-wise importance between the components. Using a Monte Carlo sampling technique, the error pdfs are propagated through the AHP software solutions to determine a pdf of uncertainty in the system wide importance of each component. To enhance the generality of the results, study of one other problem having different number of elements is reported, as are the effects of a larger assumed pdf error in the expert ranks. Validation of the Monte Carlo sample size and repeatability are also documented.

Research Organization:
EG and G Idaho, Inc., Idaho Falls, ID (United States)
Sponsoring Organization:
USDOE; USDOE, Washington, DC (United States)
DOE Contract Number:
AC07-76ID01570
OSTI ID:
5485664
Report Number(s):
EGG-NE-10078; ON: DE92012311
Country of Publication:
United States
Language:
English