Quantifying Safety Margin Using the Risk-Informed Safety Margin Characterization (RISMC)

Conference · Sun Apr 26 00:00:00 EDT 2015

OSTI ID:1344423

Grabaskas, David; Bucknor, Matthew; Brunett, Acacia; Nakayama, Marvin

The Risk-Informed Safety Margin Characterization (RISMC), developed by Idaho National Laboratory as part of the Light-Water Reactor Sustainability Project, utilizes a probabilistic safety margin comparison between a load and capacity distribution, rather than a deterministic comparison between two values, as is usually done in best-estimate plus uncertainty analyses. The goal is to determine the failure probability, or in other words, the probability of the system load equaling or exceeding the system capacity. While this method has been used in pilot studies, there has been little work conducted investigating the statistical significance of the resulting failure probability. In particular, it is difficult to determine how many simulations are necessary to properly characterize the failure probability. This work uses classical (frequentist) statistics and confidence intervals to examine the impact in statistical accuracy when the number of simulations is varied. Two methods are proposed to establish confidence intervals related to the failure probability established using a RISMC analysis. The confidence interval provides information about the statistical accuracy of the method utilized to explore the uncertainty space, and offers a quantitative method to gauge the increase in statistical accuracy due to performing additional simulations.

• Research Organization: Argonne National Lab. (ANL), Argonne, IL (United States)
• Sponsoring Organization: USDOE Office of Nuclear Energy
• DOE Contract Number: AC02-06CH11357
• OSTI ID: 1344423
• Resource Relation: Conference: 2015 International Topical Meeting on Probabilistic Safety Assessment and Analysis (PSA 2015), 04/26/15 - 04/30/15, Sun Valley, ID, USA
• Country of Publication: United States
• Language: English