PMAXS Library Generation for the Benchmark on Rostov-2 VVER-1000 Reactor

Recently, the OECD/NEA organization developed a benchmark problem based on a specific experiment conducted during the Rostov-2 first start-up procedure. The first phase of this benchmark gives an option to participants to develop their own two-group, assembly-wise homogenized cross sections. The main objective of this research is description of different techniques and methods employed during our own PMAXS libraries generation as well as their verification. The sophisticated SERPENT-2 Monte Carlo code has been employed for neutron transport in five distinct types of fuel assemblies and their corresponding group constants generation in a burnup and branch hierarchical structure. For each assembly 7 burnup states, 3 fuel temperature points, 2 moderator temperature points, 2 boron concentrations and 2 control rod states have been considered. The output files have been converted to PMAXS format using GENPMAXS interface. The verification of generated libraries has been conducted with two independent approaches. The first one is based on the comparison in assembly multiplication factor resulted from SERPENT and GENPMAXS codes. The results showed that the maximum reactivity difference between two codes is only 29 pcm. In second approach, the generated libraries have been fed to the PARCS code and neutronics results compared with reference data. The results showed that at different burnup states, the radial relative power and burnup distributions resulted from PARCS calculations are in good agreement with corresponding reference values. The maximum relative difference was less than 4% and the location of "hot assembly" was same.