Implementation and Comparison of Super Homogenization Method Based on Monte Carlo and Deterministic Codes

Abstract Monte Carlo homogenization method is increasingly used to replace conventional deterministic lattice codes to generate few-group or multi-group constants for novel nuclear systems like fast reactors and high temperature gas-cooled reactors, owing to its flexibility in geometry modeling and its wide applicability to reactors with arbitrary spectrum. In this paper, RMC, a Monte Carlo Code used for reactor physics analysis, was used to generate multi-group constants and super-homogenization (SPH) method was further used to correct cross sections for the conservation of reaction rates. In order to improve SPH calculation efficiency, SaraGR, a transport core code based on SN nodal method, was chosen to perform the iterative multi-group homogenous calculations instead of using Monte Carlo multi-group calculations. And a comparison between SPH process with SaraGR and SPH process with RMC multi-group calculations was conducted based on the very high temperature reactor critical assembly (VHTRC) benchmark. The results indicate that both SPH processes are feasible and the results agree well. The SPH process with SaraGR is much more efficient because of the inherent efficiency advantage of the deterministic method. Therefore, it is a more promising practice to generate group constants with a Monte Carlo code and perform SPH calculations with a deterministic code. Furthermore, it would be helpful to choose proper convergence criteria and introduce a proper relaxation factor into the iterative formula.