PROBLEMS OF ATOMIC SCIENCE AND TECHNOLOGY
Series: Nuclear and Reactor Constants

since 1971

Русский (РФ)

ISSN 2414-1038 (online)

NEUTRON DIFFUSION POWER ITERATIONS ACCELERATED BY PACA METHOD

EDN: LGJQLS

Authors & Affiliations

Suslov I.R.
Private Institution of the State Atomic Energy Corporation Rosatom “Innovation and Technology Center of the Breakthrough Project”, Moscow, Russia

Suslov I.R. – Leading Expert, Cand. Sci. (Phys.-Math.), Private Institution of the State Atomic Energy Corporation Rosatom “Innovation and Technology Center of the Breakthrough Project”. Contacts: 1, bldg. 7, office 307, pl. Akademika Dollezhalya, Moscow, Russia, 107140. Tel.: +7 (903) 814-35-81; e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it..

Abstract

The task of accelerating external iterations, despite the large number of acceleration methods already developed, is very relevant for cores with a dominant ratio close to unity. In particular, for cores of high-power fast reactors with flattened zones, in particular with axial heterogeneity, and is extremely important in automatic optimization methods. Neutron-physical modules of optimization complexes require a large number of sequential calculations on keff. A dominant ratio value close to unity means slow convergence of external iterations (iterations of fission neutron sources) and, as a consequence, significant calculation times.
The paper presents a formulation of the PACA (Pointwise Algebraic Collapsing Acceleration) method for accelerating external iterations in diffusion calculations, and discusses the development, implementation, and testing of the algorithm to improve the performance of the main MAG diffusion code. The iterative method is based on constructing an auxiliary single-group problem obtained by pointwise algebraic convolution (PACA) of matrices on the current solution of the original multi-group problem has keff and the spatial distribution of fission neutrons (the main mode) exactly coinciding with the corresponding values of the original problem.
The method consists of alternating multi-group and single-group iterations. Alternating multi-group and single-group iterations allows to significantly speed up obtaining a solution. Numerical examples show a decrease in the number of multi-group iterations by approximately an order of magnitude. The method is applicable to any method for solving a conditionally critical diffusion equation or a transport equation formulated in a linear-algebraic form: for codes, in particular for those using the finite difference method and the finite element method.

Keywords
neutrons diffusion equation, conditional criticality problem, iterations convergence acceleration, multi-group task, single-group task

Article Text (PDF, in Russian)

References

UDC 621.039.51.12

Problems of Atomic Science and Technology. Series: Nuclear and Reactor Constants, 2025, no. 2, 2:9