CODE VERIFICATION OF MPACT USING GANAPOL CRITICAL ROD BENCHMARK

Abstract

This work is dedicated to the code verification of MPACT, which is developed under the Consortium for Advanced Simulation of Light Water Reactors by the University of Michigan and Oak Ridge National Laboratory, where the numerical solution is compared to the reference solution of a benchmark problem with a known analytical solution. In this work, Benchmark Problem 3.4 in Barry Ganapol’s benchmark book was chosen as an MOC code verification test problem. Problem 3.4 is a bare cylinder of infinite height, which is an excellent benchmark problem for 2D MOC. To ensure that this benchmark problem exercised the same code as typically used by MPACT, the bare rod configuration was surrounded by a bounding box filled with a non-scattering material. To avoid implementing a critical rod search in the MPACT code, the MPACT analysis was performed using cross sections that yielded the given c, the average number of secondary neutrons per collision, and a rod radius that was the corresponding critical rod radius. MPACT agreed with all cases to within a few pcm. The convergence behavior was studied. The results show a 2nd order radial convergence, consistent with flat-source approximation. The convergence curves with respect to ray spacing and polar angle quadrature set order were also obtained. The other quantity of interest tabulated for Problem 3.4 was the radial distribution of the scalar flux. Two configurations were analyzed, and the resultant radial flux profiles agreed very well with the tabulated results. The verification of the production neutronics code MPACT has been augmented by the addition of the analytical solutions for an infinite cylinder from the Ganapol benchmark book. These test cases can be included in the regression suite for MPACT.