Validation of the DYN3D-Serpent code system for SFR cores using selected BFS experiments. Part II: DYN3D calculations.

The capability of the DYN3D-Serpent codes system to simulate highly heterogeneous sodium-cooled fast reactor cores has been studied. The BFS-73-1 and the BFS-62-3A critical assemblies were chosen for the investigation. The study was performed in two parts. In the first part of the study, a 3D full model of each of the assemblies was simulated using the Serpent Monte-Carlo (MC) code, and the basic neutronic characteristics were evaluated and compared against experimental values. In the second part of the study, which is the subject of this paper, the assemblies were modeled using the DYN3D nodal diffusion code. The few-group cross-sections for the DYN3D analysis were generated using the Serpent MC code. The generation of effective few-group cross-sections of such assemblies is quite a challenge due to the substantial heterogeneity of the assemblies configuration. Therefore, the use of homogenization techniques was considered and evaluated. Initially, the GET and SPH techniques were applied for the analysis of the BFS-73-1 assembly core fuel rods, and of selected fuel rods from the BFS-62-3A assembly. Then, the SPH method was implemented and demonstrated for a pin-by-pin calculation of the BFS-73-1 assembly. It was shown that the GET and the SPH method noticeably improve the prediction accuracy of the DYN3D code. The results of the DYN3D pin-by-pin calculation with the SPH correction agree very well with that of the full assembly Serpent results, which in turn agree very well with the experimental data.