Analysis of the 2-D C5G7 MOX Benchmark with VISWAM Code

Abstract

The Indian nuclear power programme is being augmented with a variety of imported light water reactors (LWRs). Close interaction between the national labs, plant operators and educational institutions is required to meet the human resource requirements of this program. A neutronics code VISWAM for physics analysis of current and future power reactors is being developed as a part of the above long term goal. The overall plan is to develop a lattice code for generating few group cross sections and a diffusion theory based few group core solver for steady state and transient core calculations. As a first step towards this goal, we have taken up the development of a lattice burnup code. Initially, we used a combination of 1‐D multigroup transport for pincell and supercell calculations and 2‐D few‐group diffusion theory for assembly calculations. Although this method was fast and reasonably accurate for simple fuel assembly designs, we found that the power distribution errors were large in the vicinity of strong absorbers like burnable poison and control rod pins. We improved the solution method by using the 2‐D collision probability (CP) method for calculating the fluxes in each region and currents at each surface of a cell. Adjacent cells are coupled using interface currents at cell boundaries with double P2 (DP2) expansion of angular flux. The advantage of this method is that it can been extended for 2‐D full core solution of the neutron transport equation without spatial homogenization. We tested this code using the OECD/NEA 2‐D C5G7 MOX fuel assembly benchmark. The eigenvalue for the core calculation lies within 0.05% of the reference result. The average power distribution error is less than 1.0%.