Monte Carlo codes benchmarking on sub-critical fuel debris particles system for neutronic analysis

Abstract

Fuel debris removal is the most challenging part of damaged nuclear power station decommissioning. It is important to carry out nuclear safety calculations accurately and quickly enough. Here, it was clarified that modern codes based on the Monte Carlo method were capable of performing neutronic analysis with the same accuracy and without significant differences in the results. The benchmark calculations were performed using three codes: MVP, Serpent, and MCU. In this study, the comparison focused on multiplication factor, neutron fluxes and reaction rates relative difference, and calculation time of many fuel debris particles system. Then the calculation results were used when codes comparing. It was shown that the calculation results showed good agreement between all codes. It was assumed that minor differences in the thermal range of neutron fluxes can be caused by different thermal neutrons scattering treatment for all codes. The study also showed that solving such problems requires significant computing power and time. It has been proven that the statistical geometry model in the MVP and the explicit stochastic geometry model in the Serpent have the possibility to provide solutions with the same accuracy, but much faster.