IAEA GT-MHR benchmark calculations by using the HELIOS/MASTER physics analysis procedure and the MCNP Monte Carlo code

Abstract

The IAEA's gas-cooled reactor program has coordinated international cooperation for an evaluation of a high temperature gas-cooled reactor's performance, which includes a validation of the physics analysis codes and the performance models for the proposed GT-MHR. This benchmark problem consists of the pin and block calculations and the reactor physics of the control rod worth for the GT-MHR with a weapon grade plutonium fuel. Benchmark analysis has been performed by using the HELIOS/MASTER deterministic code package and the MCNP Monte Carlo code. The deterministic code package adopts a conventional 2-step procedure in which a few group constants are generated by a transport lattice calculation, and the reactor physics analysis is performed by a 3-dimensional diffusion calculation. In order to solve particular modeling issues in GT-MHR, recently developed technologies were utilized and new analysis procedure was devised. Double heterogeneity effect could be covered by using the reactivity-equivalent physical transformation (RPT) method. Strong core–reflector interaction could be resolved by applying an equivalence theory to the generation of the reflector cross sections. In order to accurately handle with very large control rods which are asymmetrically located in a fuel and a reflector block, the surface dependent discontinuity factors (SDFs) were considered in applying an equivalence theory. A new method has been devised to consider SDFs without any modification of the nodal solver in MASTER. All computational results of the HELIOS/MASTER code package were compared with those of MCNP. The multiplication factors of HELIOS for the pin cells are in very good agreement with those of MCNP to within a maximum error of 693 pcm Δ ρ. The maximum differences of the multiplication factors for the fuel blocks are about 457 pcm Δ ρ and the control rod worths of HELIOS are consistent with those of MCNP to within a maximum error of 3.09%. On considering a SDF in the core calculations, the maximum differences of the control rod worths are significantly decreased to be 7.7% from 21.5%. It is showed that there are good consistencies between the deterministic code package and the Monte Carlo code from the results of these benchmark calculations. Therefore, the HELIOS/MASTER 2-step procedure can be used as a standard reactor physics analysis tool for a prismatic VHTR.