Generation of the Neutron XS ENDF/B-VIII.0 library for the MOC STREAM code and the Monte Carlo MCS code. In this document, ENDF/B-VIII.0 XS data verification and validation uses STREAM code and MCS code. In the case of the unique nuclear data library for STREAM, the neutron transport analysis code, ENDF files for each nuclide were produced in group XS using the NJOY code system, and in a second step the STREAM XS library was produced using the STREAM production system library, NTOS.
Using this multi-group nuclear data production system, STREAM XS library was produced for all nuclides based on ENDF/B-VII.1 nuclear data library and ENDF/B-VIII.0 nuclear data library. To assess the accuracy of the library, STREAM code and MCS code were compared simultaneously for each ENDF version. STREAM code nuclear source used both ENDF/B-VIII.0 and ENDF/B-VII.1, and their results compared to each of the MCS code results.
When STREAM's results were compared for each version of ENDF, the NCA benchmark found that the difference between the effective multiplication factor was within 100 pcm for the problem, confirming that the ENDF/B-VIII.0 STREAM XS Library was correctly produced. Also in the case of VERA benchmark and ICSBEP benchmark, the difference between the effective multiplication factor was within 300 pcm for the problem of ENDF/B-VII.1 XS Library results and ENDF/B-VIII.0 XS Library results.
INTRODUCTION
DESCRIPTION OF ENDF/B-VI0
CODE DESCRIPTION
STREAM
A neutron transport analysis MOC code, STREAM (Steady and Transient Reactor Analysis Code by Method of Characteristics), was developed at UNIST. MOC mesh physics codes are used to generate cross-sectional data for nodal codes, where nodal codes are used to model the coupled neutronics and thermo-hydraulic behavior of the entire reactor core during steady and transient operation. Once the flux distribution is known, the cross sections can be condensed and homogenized to the structure required by the nodal code.
The nodal code then joins the different grids together to build different groups of fuel in the reactor core. STREAM is developed to perform an entire LWR core calculation with the direct transport analysis method and the two-step method. STREAM has the capability to analyze the entire LWR core through the two-step method (with PARCS or RAST-K 2.0) and the direct transport (2-D) method.
It has several functions: Multi-group cross-section generation (Pin-based pointwise energy lowering method, equivalence theory for structural material, resonance scattering correction, improved neutron current method, inflow transport correction), transport solver (method of characteristics, T-Y optimal quadrature sets, Collection modular ray tracing method, method of direct neutron path binding, P0~P5 scattering source processing, coarse-mesh finite difference acceleration), Depletion (Matrix exponential method, Chebyshev rational approximation method, Chain with ~1400 isotopes, Predictor/ corrector), and generation of few-group constants (discontinuity factor, two-group cross section, Critical spectrum with fundamental mode calculation). In STREAM, Pin-based Slowing-down Method (PSM) is a newly developed method to solve braking problem, and it is more accurate method than other methods. 1) MOC code 2-dimensional quarter core design, (2) Ray tracing design for quarter core (3) Ray tracing for a pin.
MCS
For verification of the MCS code, Monte Carlo codes such as McCARD, OpenMC, Serpent, MCNP, KENO are used and the accuracy of the MCS whole-core analysis is interpreted.
STREAM ENDF/B-VIII.0 LIBRARY GENERATION PROCESSING WITH NJOY2016
UNRESR produces efficient self-shielded cross sections for resonance reactions in the unresolved energy range. The THERMR module generates pointwise neutron scattering cross sections in the thermal energy range and adds them to an existing PENDF file. GROUPR generates self-shielded multi-group cross sections, group-to-group scattering matrices, photon production matrices, and charged particle multi-group cross sections from pointwise input.
The NTOS reproduces reprocessed nuclear cross-section data from NJOY in a library format for STREAM. To reduce the need for internal STREAM calculations, NTOS itself collects and processes nuclear cross-nuclear data from NJOY nuclear data to produce a library. The nuclear cross-section processing within the NTOS is performed based on temperature and nuclide from the NJOY output file, and the XSs are:.
This removes the nuclear cross section data of the parts that are not required and pre-calculates the required nuclear cross section from the nuclear transport analysis. In addition, because ENDF/B-VIII.0 is only calculated as NJOY2016, ENDF7.1 is also calculated as NJOY2016 to prevent errors from NJOY versions.
ENDF/B-VIII.0 AND ENDF/B-VII.1 URANIUM XS DATA COMPARISON
- U-238 Total XS
- U-238 Absorption Reaction Rates Comparison with ENDF/B-VIII.0 and ENDF/B-VII.1
- STREAM – U-238 Absorption RR & Reactivity difference
- MCS – U-238 Absorption RR & Reactivity difference
- U-235 Total XS
- U-235 Absorption Reaction Rates Comparison with ENDF/B-VIII.0 and ENDF/B-VII.1
- STREAM – U-235 Absorption RR & Reactivity difference
- MCS – U-235 Absorption RR & Reactivity difference
Compared to U-238 absorption reaction rates and reactivity difference between ENDF/B-VIII.0 and ENDF/B-VII.1, 4.9% w/o NCA stick is used. The reactivity difference of each group between ENDF/B-VIII.0 and ENDF/B-VII.1 is within the allowed range, such as -61 pcm, and the total reactivity difference is approx. -420 pcm in STREAM. U-238 Absorption Reaction Rates Comparison of 4.9w/o Fuel Pin with ENDF/B-VIII.0 and ENDF/B-VII.1 in STREAM.
Therefore, the reactivity difference of MCS may be different from the reactivity difference of STREAM. The reactivity difference of each group between ENDF/B-VIII.0 and ENDF/B-VII.1 is within the allowable range, such as 100 pcm, and the total reactivity difference is about 296 pct in MCS. U-238 Absorption Reaction Rates Comparison of 4.9w/o fuel pin with ENDF/B-VIII.0 and ENDF/B-VII.1 in MCS.
In comparison of U-235 absorption reaction rates and the difference in reactivity between ENDF/B- VIII.0 and ENDF/B-VII.1, 4.9% w/o pin NCA is used. The reactivity difference of each group between ENDF/B-VIII.0 and ENDF/B-VII.1 is within the allowed range as 100 pcm and the total reactivity difference is about 279 pcm in STREAM. The shape of the difference in reactivity of the U-235 MCS with ENDF/B-VIII.0 and ENDF/B- VII.1 is very similar to the shape of the change in reactivity of the U-235 Streams.
XS COMPARISON OF MAJOR ISOTOPES
Helium's XS has no difference between ENDF/B-VIII.0 and ENDF/B-VII.1 library versions.
NCA BENCHMARK RESULTS
In the experiments, the axial length of the tungsten rods is 110 cm (Covers the entire range of critical water height). These pins in the test area are surrounded by a block of polystyrene containing 1000 ppm boron. The radius of the polystyrene blocks is 0.6124 cm and a total of 4 stainless steel rods as support for the polystyrene block sheet.
This core is designed to simulate the start-of-cycle (BOC) conditions of a PWR reactor. This core was designed to simulate end-of-cycle conditions (EOC) of a PWR reactor. As shown, Table 3 uses the ENDF/B-VIII.0 XS library and Table 4 uses the ENDF/B-VII.1 XS library.
MCS and STREAM have small differences in k values within 84 pcm with both XS libraries. Figure 90~94 is a comparison of pin power distribution with STREAM results with each version of the ENDF library and NCA benchmark measurement data. NCA Core 1 Pin power distribution comparison with STREAM and measurement data with ENDF/B-VIII.0, ENDF/B-VII.1 and ENDF/B-VII.0.
NCA Core 2 Pin current distribution comparison with STREAM and measurement data at ENDF/B-VIII.0, ENDF/B-VII.1 and ENDF/B-VII.0. NCA Core 3 Pin current distribution comparison with STREAM and measurement data at ENDF/B-VIII.0, ENDF/B-VII.1 and ENDF/B-VII.0.
ICSBEP BENCHMARK RESULTS
Under the iron-dominant condition, STREAM ENDF/B-VIII.0 results are 184 pcm less than STREAM ENDF/B-VII.1 results. XS change of Fe atoms with ENDF/B-VIII.0 is significant so it may affect the results. In the case of HEU-MET-FAST-055 and HEU-MET-FAST-060, the difference is whether it is tungsten or not.
For ENDF/B-VIII.0, the difference in the multiplication factor between STREAM and MCS in these two cases is within -73. It seems that the tungsten XS difference effect between ENDF/B-VII.0 and ENDF/B-VII.1 is not too large to have a significant effect on the multiplication factor. As shown in Table 15, the STREAM ENDF/B-VIII.0 library in ICSBEP benchmark cases using metallic fuel agrees well with the MCS ENDF/B-VIII.0 results.
When the ENDF/B-VIII.0 STREAM results are compared as a whole with the ENDF/B-VII.1 STREAM results, it can be considered to fit within a maximum of 264 pcm.
VERA BENCHMARK RESULTS
For the validation of the STREAM ENDF/B-VIII.0 XS library, the STREAM ENDF/B-VIII.0 results and STREAM ENDF/B-VII.1 results are compared. In Table 8, the difference in the multiplication factor between the STREAM ENDF/B-VIII.0 results and the reference value is within 179pcm in most cases, except for Problem 2G and Problem 2H. Furthermore, except for Problem 2G and Problem 2H, the ENDF/B-VIII.0 results tend to agree better with the reference than the ENDF/B-VII.1 results.
In case of problem, 2G and 2H contain 24 joysticks; are each rod AIC and rod B4C. This tendency was also observed in the Monte Carlo code, the MCS results in Table 9. As shown in Table 10, the STREAM ENDF/B-VIII.0 XS library results are up to -40 pcm by comparing the MCS results.
CONCLUSIONS
