Demonstration of the Serpent Monte-Carlo code applicability to Few-group Constants Generation for Existing and Advanced Reactor concepts


Demonstration of the Serpent Monte-Carlo code applicability to Few-group Constants Generation for Existing and Advanced Reactor concepts

Fridman, E.; Shwageraus, E.

Serpent is a continuous-energy Monte Carlo (MC) reactor physics code recently developed at VTT Technical Research Centre of Finland. Serpent can be used for 2D fuel lattice calculations as well as for 3D full core simulations. Due to its built-in decay and burnup routine Serpent can perform depletion and decay analysis to provide time-dependent isotopic compositions and spent fuel characteristics including radioactivity and decay heat. Serpent uses matrix exponential method to solve the Bateman decay and depletion equations while the solution of the matrix exponential relies on the Chebyshev Rational Approximation Method (CRAM). Serpent runs significantly faster than other MC codes due to the two main reasons: 1) the use of the Woodcock delta-tracking in a combination with a typical surface-to-surface ray-tracing in a geometry routine, and 2) the use of the unionized energy grid for all point-wise reaction cross sections. The later, however, considerably increases the memory requirements and can be a bottleneck in simulations with a large number of involved nuclides.
Serpent is especially designed to generate homogenized constants for deterministic 3D core analysis. For any region of interest the code automatically calculates homogenized few-group cross sections, group-to-group scattering matrices, diffusion coefficients, assembly discontinuity factors, kinetics parameters, etc. More details can be found in Serpent User's Manual. Recently some new calculation methods related to the production of homogenized few-group constants were implemented in the Serpent code including homogenization in leakage-corrected criticality spectrum, group constant generation in reflectors and other non-fissile regions, and improved treatment of neutron-multiplying scattering reactions.
The capability to generate homogenized few-group constants can be considered as one of the most attractive features of Serpent. Being a MC code, Serpent is capable of handling complex geometries without any major approximations and can be used for producing cross section data for virtually any fuel or reactor type. The demonstration of the Serpent capability to generate few-group cross sections for different reactor systems is the main topic of this paper.

  • Lecture (Conference)
    26th Conference of the Nuclear Societies in Israel, 21.-23.02.2012, Dead Sea, Israel
  • Contribution to proceedings
    26th Conference of the Nuclear Societies in Israel, 21.-23.02.2012, Dead Sea, Israel

Permalink: https://www.hzdr.de/publications/Publ-16568
Publ.-Id: 16568