Validation of self-adjoint angular flux neutron transport code with TAKEDA benchmark

View at publisher
Authors: Duoyu Jiang, Peng Fei Xu, Jiang Xinbiao, Tianliang Hu, L. Wang, Lu Cao, Huaqi LI, Da Li, Lixin Chen
Cited by: 0
References: 12

Abstract

The rapid development of advanced small modular reactors—including modular high-temperature gas-cooled reactors and mobile microreactors—demands neutron transport codes capable of accurately resolving both thermal and fast neutron spectra, as well as complex geometries involving structured and unstructured meshes with rectangular or hexagonal boundary conditions. To rigorously validate such capabilities, we employ the three-dimensional TAKEDA benchmark, comprising four configurations: a light-water reactor with square lattice geometry, two fast breeder reactors (FBRs) in square arrangements, and a hexagonally configured FBR - encompassing a representative range of spectral and geometric challenges. Using SAAFCGSN, a second-order self-adjoint angular flux neutron transport code based on the finite element method, we perform full-core simulations of all four TAKEDA benchmark models. The computed effective neutron multiplication factors (keff) agree closely with reference solutions, exhibiting maximum relative deviations of 70pcm, 90pcm, 140pcm and 80pcm (1pcm=10−5) across the configurations. Control rod worth predictions show maximum relative discrepancies of -4.42%, 2.50%, 1.02% and 1.60%, respectively. These results demonstrate high fidelity in both eigenvalue and reactivity coefficient predictions, validating the code’s accuracy and robustness in modeling complex, three-dimensional reactor systems with mixed spectral and geometric features. Our findings establish SAAFCGSN as a reliable tool for high-fidelity neutronics analysis in next-generation small modular reactors.

Fields of Study & Subjects

Nuclear reactor physics and engineering Nuclear Physics and Applications Nuclear Engineering Thermal-Hydraulics

Publication details

Journal / source
Journal of Nuclear Science and Technology
Volume
63
Issue
8
Pages
901–911
Publication date
16 January 2026
Language
EN