Use of Response Surface for Evaluation of Functional Failure of Passive Safety System

Abstract

Passive safety systems are more vulnerable to their environment and initial condition due to the typical low driving forces, e.g., that of the natural circulation. We investigate the merits of different methods for analysis of the probability of passive safety system “functional failure”. Variation of the coolant flow condition due to complex thermal hydraulic phenomena may cause a passive safety system to be unable to perform its function. In this report the RELAP5 code is used with normally distributed input parameters to estimate the functional failure probability of the passive system. Response surfaces are generated from RELAP5 results using different sampling techniques. Comparison between response surface and RELAP5 shows that the standard deviations are different. We identify sufficient levels of simulation effort required for accurate estimates.