SERPENT-based Few-Group Cross Section Data for NESTLE-C

Abstract of the technical paper/presentation to be presented at:
M&C 2019 International Conference on Mathematics and Computational Methods Applied to Nuclear Science and Engineering
August 25-29, 2019

Prepared by:
A. Trottier and S. Pfeiffer, Canadian Nuclear Laboratories
D. Serghiuta, Canadian Nuclear Safety Commission

Abstract

To enhance its capability for independent verification of safety cases using more realistic methodologies which rely on complex analytical simulations using 3D-neutronics thermalhydraulic coupled computational procedures, Canadian Nuclear Safety Commission (CNSC) staff have initiated a multi-phase research project to evaluate and implement an integrated Uncertainty Characterization Framework (UCF) with primary application to CANDU neutronics calculations. As a first step in the implementation of the UCF, it was decided to focus on cross-section uncertainties and reduce some of the modelling uncertainties by using a Monte-Carlo method for the generation of diffusion cross sections and implementing multi-groups capability in the NESTLE-C core simulator. The proposed reduced framework assumes that the generic multi-group library along with its covariance matrix developed at Oak Ridge National Laboratory for reactors with thermal spectra can be used for CANDU lattice physics simulation. To remain consistent with this library structure, the SCALE neutron transport codes have been selected to propagate the cross sections uncertainties thru the CANDU lattice simulations.

The reduced UCF also assumes that all NESTLE-C input uncertainties may be described using Gaussian normal distributions. The Gaussian assumption requires both the mean values and covariance matrix for the NESTLE-C input cross-sections. The covariance matrix is generated using SCALE lattice physics codes (TRITON/NEWT and KENO-3D), however the mean values are generated using the Monte-Carlo code SERPENT. The discussion in this paper is limited to the generation of the mean values for the NESTLE-C few-group cross sections using the Monte-Carlo code SERPENT. Specifically, this paper presents the codes and methods used to update and expand the few-group cross section input files for use with the NESTLE-C core simulator using the SERPENT code. These input files or NESTLE-C library of few-group cross sections and kinetics data, provide the necessary cross section data to perform full-core simulations of CANDU-type reactors fuelled with Natural Uranium 37-element fuel bundles. Each file encodes two main sets of group constant tables: one for the fuelled regions of the reactor core, and one for the heavy water reflector. Incremental cross section data for modeling of reactivity devices were also prepared. Group constants were prepared over several branch cases, covering a range of local conditions which deviate from the reference normal operating conditions. The paper starts with a review of the NESTLE-C and the few-group cross section data input, followed by a discussion of the SERPENT models used to prepare the data, and the execution and post-processing. Results for the few-group library of cross section input data for the two-, four-, and eight-group data sets are presented. The two-group data is compared against the HELIOS-based library of two-group data previously used with NESTLE-C core simulator in independent regulatory verification work. Results for a steady-state case for a generic CANDU-6 core are also presented. This work was carried out by Canadian Nuclear Laboratories for the CNSC.

To obtain a copy of the abstract’s document, contact us at cnsc.info.ccsn@cnsc-ccsn.gc.ca or call 613-995-5894 or 1-800-668-5284 (in Canada). When contacting us, please provide the title and date of the abstract.