Whole-Core Transport Solutions to A Stylized CANDU-6 Core Problem

Abstract

A stylized CANDU half core with adjuster rods was used to test the accuracy and efficiency of the coarse mesh transport (COMET) code. The benchmark problem consists of 2280 fuel bundles in 190 fuel channels with 8 different burnups. The model includes a total of 21 adjuster rods of 1 type perpendicular to the fuel channels and located on three axial locations. A modified version of the MCNP code was used to generate the response function library in the pre-computation phase, and the COMET was used to compute the global eigenvalue and bundle/pin power distribution. In these calculations, a set of tensor products of orthogonal polynomials were used to expand the neutron angular phase space distributions on the interfaces between coarse meshes. The COMET calculations were compared with the MCNP reference solutions. The comparison showed both the global eigenvalue and bundle/pin power distributions predicated by COMET agrees very well with the MCNP reference solution if the orders of expansion in the two spatial variables and the polar and azimuth angles are 4, 4, 2 and 2. These comparisons indicate that COMET can achieve accuracy comparable to Monte Carlo methods while maintaining computational efficiency significantly (orders of magnitude) faster than that of MCNP.