2014SNA + MC 2013 - Joint International Conference on Supercomputing in Nuclear Applications + Monte Carlo
|Number of page(s)||11|
|Section||1. Computational Nuclear Applications: a. Nuclear Reactor Analysis|
|Published online||06 June 2014|
Applicability of 3D Monte Carlo simulations for local values calculations in a PWR core
1 Institut de Radioprotection et de Sûreté Nucléaire (IRSN), PSN-EXP/SNC, Fontenay-aux-Roses, 92262, France
2 External Consultant at IRSN, Salignac, 33240, France
* Corresponding Author: E-mail: firstname.lastname@example.org
As technical support of the French Nuclear Safety Authority, IRSN has been developing the MORET Monte Carlo code for many years in the framework of criticality safety assessment and is now working to extend its application to reactor physics. For that purpose, beside the validation for criticality safety (more than 2000 benchmarks from the ICSBEP Handbook have been modeled and analyzed), a complementary validation phase for reactor physics has been started, with benchmarks from IRPHEP Handbook and others.
In particular, to evaluate the applicability of MORET and other Monte Carlo codes for local flux or power density calculations in large power reactors, it has been decided to contribute to the “Monte Carlo Performance Benchmark” (hosted by OECD/NEA). The aim of this benchmark is to monitor, in forthcoming decades, the performance progress of detailed Monte Carlo full core calculations. More precisely, it measures their advancement towards achieving high statistical accuracy in reasonable computation time for local power at fuel pellet level. A full PWR reactor core is modeled to compute local power densities for more than 6 million fuel regions.
This paper presents results obtained at IRSN for this benchmark with MORET and comparisons with MCNP. The number of fuel elements is so large that source convergence as well as statistical convergence issues could cause large errors in local tallies, especially in peripheral zones. Various sampling or tracking methods have been implemented in MORET, and their operational effects on such a complex case have been studied. Beyond convergence issues, to compute local values in so many fuel regions could cause prohibitive slowing down of neutron tracking. To avoid this, energy grid unification and tallies preparation before tracking have been implemented, tested and proved to be successful. In this particular case, IRSN obtained promising results with MORET compared to MCNP, in terms of local power densities, standard deviations and computing times.
Key words: MORET / MCNP / Monte Carlo / performance / local power / PWR reactor
© Owned by the authors, published by EDP Sciences, 2014