ExaSMR Benchmarks
Find a file
2018-01-16 14:14:06 -06:00
2x2-periodic Separate <filter> in tallies.xml files 2017-07-10 11:24:09 -05:00
2x2-reflector Separate <filter> in tallies.xml files 2017-07-10 11:24:09 -05:00
assembly Separate <filter> in tallies.xml files 2017-07-10 11:24:09 -05:00
pin-cell Separate <filter> in tallies.xml files 2017-07-10 11:24:09 -05:00
smr Make sure sleeve appears on fifth grid spacer 2018-01-16 14:14:06 -06:00
.gitattributes Added in support for depleted nuclides to each model 2017-02-22 16:49:53 -05:00
.gitignore Added depleted materials.xml files for each model to Git LFS 2017-02-22 16:51:02 -05:00
profile.py Reduced number of part batches and particles for more effective profiling 2017-01-04 01:45:21 +00:00
README.md Update README.md and make references to NuScale DC more explicit 2017-11-01 14:36:51 -05:00

ECP Benchmarks

This repository contains benchmarks for performance profiling of the OpenMC and Shift Monte Carlo codes for ECP. Each directory contains a build-xml.py Python script which uses the OpenMC Python API to create XML input files for OpenMC. The Python scripts and the XML files are included in this repository. Each benchmark is derived from a subset (e.g., fuel pins, assemblies) of the full-core BEAVRS model with fresh UO2 fuel at HZP conditions. Each benchmark is modeled in 2D with reflective boundary conditions along the axial dimension.

Model Descriptions

fuel-pin
A single fuel pin with reflective boundary conditions (i.e., an infinitely repeating array of fuel pins). The default configuration is the BEAVRS fuel pin with fresh 1.6% enriched UO2 fuel, but the script can be toggled to 2.4% or 3.1% enriched fuel.
assembly
A single fuel assembly with reflective boundary conditions (i.e., an infinitely repeating lattice of fuel assemblies). The default configuration is the BEAVRS assembly with fresh 1.6% enriched UO2 fuel with 24 water-filled control rod guide tubes and a central air-filled instrument tube. However, the script can be toggled to use any of the 20+ fuel assemblies in the BEAVRS model.
2x2-periodic
A 2x2 fuel assembly colorset with periodic boundary conditions (i.e., an infinitely repeating lattice of the 2x2 assembly colorset). The default configuration includes the BEAVRS assembly with fresh 1.6% enriched UO2 fuel with 24 water-filled control rod guide tubes and a central air-filled instrument tube, along with a 3.1% enriched fuel assembly with 20 burnable poisons, four control rod guide tubes and a central instrument tube. However, the script may be toggled to use any pair of the 20+ fuel assemblies in the BEAVRS model.
2x2-reflector
A 2x2 fuel assembly colorset surrounded by a water reflector. Reflective boundary conditions are used on the top and left boundaries (adjacent to the assemblies) and vacuum boundary conditions are used on the bottom and right boundaries (adjacent to the reflector). The default configuration includes the BEAVRS assembly with fresh 1.6% enriched UO2 fuel with 24 water-filled control rod guide tubes and a central air-filled instrument tube, along with a 3.1% enriched fuel assembly with 20 burnable poisons, four control rod guide tubes and a central instrument tube. However, the script may be toggled to use any pair of the 20+ fuel assemblies in the BEAVRS model.
smr
A 3D Small Modular Reactor (SMR) model that roughly mimics the design of the NuScale reactor. The reactor core has 37 fuel assemblies that alternate between 3.1% enriched and 2.4% enriched UO2 fuel. The center fuel assembly has 1.6% enriched UO2 fuel. Where possible, we have attempted to use the same parameters for the fuel assemblies and fuel rods as those specified in the NuScale design submittal to the Nuclear Regulatory Commission (NRC), for example: 264 fuel rods per assembly, 24 guide tubes per assembly, 1 instrument tube per assembly, five spacer grids per assembly, fuel rod pitch of 0.496 in, and an active fuel length of 200 cm. Many of the details of the actual NuScale fuel assembly design are redacted from the design submittal because they are export controlled/proprietary information. The purpose of our model is not to be an exact replica of the NuScale model; rather, it is intended to capture most of the physical complexities that are involved in modeling a full reactor core and to provide a suitable model for carrying out full core performance tests on the testbed architectures.

Dependencies

These scripts depend on the following Python packages:

  • mit-crpg/openmc (develop branch)
  • mit-crpg/PWR_benchmarks

Of particular note, The Python beavrs package must be installed from the mit-crpg/PWR_benchmarks repository. This package may be installed with distutils as follows:

cd PWR_benchmarks/BEAVRS/openmc/inputs
python setup.py install