Research on the ODR-VS Method Applicable to Various Geometric Models

Abstract To improve fuel performance and overcome the lack of thermal conductivity and ductility of ceramic nuclear fuel, metal-based fuel elements are used in some high-flux research and test reactors, high-power power reactors and long-life power reactors. However, the stochastic dispersion of fuel particles presents a challenge for accurate modeling and neutron transport analysis. Two commonly employed explicit modeling approaches for stochastic media are the Random Sequential Addition (RSA) method and the Distinct Element Method (DEM). However, the RSA method is constrained by a maximum packing fraction (38.4%). While the DEM method overcomes the limitation of the packing fraction, its inherent complexity, substantial computational requirements, and low packing efficiency remain significant challenges. Moreover, different shapes of fuel elements impose higher demands on geometric universality. Building upon the proposed ODR-VS (Optimized Dropping and Rolling and Virtual Surface) method, we further extend its geometric applicability to include shapes such as cubes, sector cylinders, annular cylinders, and spheres, addressing the geometric modeling needs of a wide range of fuel elements. By employing the enhanced ODR-VS method, we model fuel elements of various shapes and conduct an analysis of the packing efficiency for each type.