A stochastic-deterministic approach for evaluation of uncertainty in the predicted maximum fuel bundle enthalpy in a CANDU postulated LBLOCA event

Abstract of a technical presentation presented at:
The 19th Pacific Basin Nuclear Conference (PBNC 2014)
Vancouver, British Columbia, Canada,
August 24–28, 2014.

Prepared by:
Dumitru Serghiuta, John Tholammakkil and Wei Shen
Canadian Nuclear Safety Commission (CNSC)
Ottawa, Ontario, Canada

Abstract

A stochastic-deterministic approach based on representation of uncertainties by subjective probabilities is proposed for evaluation of bounding values of functional failure probability and assessment of probabilistic safety margins. The approach is designed for screening and limited independent review verification. Its application is illustrated for a postulated generic CANDU LBLOCA and evaluation of the possibility distribution function of maximum bundle enthalpy considering the reactor physics part of LBLOCA power pulse simulation only. The computer codes HELIOS and NESTLE-CANDU were used in a stochastic procedure driven by the computer code DAKOTA to simulate the LBLOCA power pulse using combinations of core neutronic characteristics randomly generated from postulated subjective probability distributions with deterministic constraints and fixed transient bundle-wise thermal hydraulic conditions. With this information, a bounding estimate of functional failure probability using the limit for the maximum fuel bundle enthalpy can be derived for use in evaluation of core damage frequency.