Assessment of nTRACER and PARCS Performance for VVER Configurations

Abstract

The current standard for computational neutronic analysis of nuclear power plants (NPPs) is the so-called conventional approach, which relies on few-group, coarse-mesh diffusion calculations. The recent evolution of computing clusters and computational techniques gives the opportunity to use codes that perform first principles–based multiphysics simulations, allowing high resolution of the calculated parameters. The goal of this work is to assess the performance of the deterministic high-resolution transport code nTRACER and the nodal code PARCS on the basis of VVER core configurations. The V1000-2D benchmarks of the NUclear REactor SIMulation (NURESIM) project framework are used to provide the neutronic and modeling data as well as reference solutions for both codes. A reference solution is also generated using Serpent2. The accuracy and limitations of the codes are illustrated together with their computational requirements. PARCS shows good agreement with the reference solutions although the results present some discrepancies due to the provided discontinuity factors. nTRACER is capable of producing high-accuracy and high-resolution solutions in a fraction of the time required by the Monte Carlo solver.

Keywords:

• High-resolution solvers
• nTRACER
• PARCS
• Serpent2
• VVER

Acknowledgments

This work was conducted in the framework of the Steady-State and Transient Analysis Research for the Swiss Reactors (STARS) program (http://www.psi.ch/stars) with the support of the Swiss National Supercomputing Centre (CSCS) (https://www.cscs.ch/).