RMC/CTF multiphysics solutions to VERA core physics benchmark problem #8. (August 2020)

• Record Type: Journal Article
• Authors: Luo, Zhen; Li, Hao; Yu, Ganglin; Ma, Yugao; Li, Kaiwen; Guo, Xiaoyu; Li, Yaodong; Wang, Kan
• Is Part Of: Annals of nuclear energy. Volume 143(2020)
• Publication Date: 2020-08
• Subjects: VERA benchmark -- Neutronics and thermal hydraulics coupling -- RMC/CTF code system -- Critical boron concentrations; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2020.107466
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: The first 48h startup of VERA Problem #8 calculation was simulated. Burnup calculation with 2.5 million burnup regions was completed using a hybrid parallel mode. The effects of axial-layering on multiphysics solutions were studied in the coupling work. RMC/CTF is capable of multi-physics coupling for simulations of large-scale nuclear reactors. Abstract: The Virtual Environment for Reactor Applications (VERA) Core Physics Benchmark Progression problems #1 to #10 based on the 17 × 17 Pressurized Water Reactor (PWR) Watts Bar Unit 1 designed by Westinghouse have been proposed by the Consortium for Advanced Simulations of Light Water Reactors (CASL) as core physics benchmarks. These problems were selected to assist nuclear software and methods developers to develop the capabilities needed to accurately model nuclear power reactors. These problems have 10 different conditions ranging from a simple 2D pin cell to the full core time-dependent depletion. VERA Problem #8, Physical Reactor Flux Maps, describes the first 48 h of a reactor startup with the power gradually increasing from 0% to 100% including the control rod movements which must be considered. This problem was solved in this study using the continuous energy Monte Carlo code RMC coupled with the subchannel thermal-hydraulics code CTF in a hybrid computational strategy. The effects of axial-layering on the solutions were studied with the model divided into 10 layers in the axial direction with the analyses considering both the cost and the accuracy with VERA Problem #8 simulated on the Tianhe II supercomputer. A large number of particles was simulated to get a reliable solution with the mean uncertainty in the power distribution reduced to about 2.6% for all the 526, 640 burnup regions for each iteration between RMC and CTF. The eigenvalue trajectory of the critical boron concentration during the startup was provided with the predicted k eff which differs from the common continuous decline. The RMC/CTF code also gives a 3D view of the relative power density distribution and the relative uncertainty distribution at three typical time points and the subchannel outlet temperatures whose peak locations are consistent with the radial power distribution. The RMC/CTF solution is the first published coupled Monte Carlo neutronics/subchannel thermal hydraulics solutions for Problem #8.