Integrated Framework for Propagation of Uncertainties in Nuclear Cross-Sections in CANDU Steady-State and Transient Reactor Physics Simulations

Abstract

This paper describes the approach for implementation of a reduced Uncertainty Characterization Framework (UFC) which is focused on developing an integrated automated capability for uncertainty analysis for the core simulator NESTLE-C in both steady state and transient CANDU core calculations. This work is carried out under a research project initiated by the Canadian Nuclear Safety Commission (CNSC) to enhance the capability for regulatory independent evaluation of best estimate reactor analysis simulations with consideration of uncertainties. The end result is a complete library of diffusion group cross-sections and associated uncertainty distribution function and covariance matrix and an integrated platform which includes execution scripts, sampler software, NESTLE-C, and processing software with an input flag that allows the user to estimate the joint probability distribution for all output results. The output results are also customized to include typical statistical quantities such as means, variances, and covariance information for user-selected core attributes. The main computer codes employed in the generation of the diffusion cross-sections library are SERPENT, TRITON, T2N, and CRANE.