HTR-10 Griffin Neutronics Results

Contact: Javier Ortensi, [email protected]

This section is a short summary of the results section of (Ortensi et al., 2018) and looks at the various results carried out by Griffin for various HTR-10 calculations. The two main input files are the htr-10-critical.i and the htr-10-full.i which are documented in detail in the Griffin Neutronics Model Section.

Running an Input File

If the user has access to Griffin built upon a working Moose Framework the following command from the shell prompt will execute the input file. The mpirun command specified will make Griffin use four cores/processes to perform the calculation.



mpirun -n 4 /projects/griffin/griffin-opt -i htr-10-critical.i

Alternatively, if using an HPC and running in parallel a submission script or interactive session can be submitted via PBS as seen here Binary Access with INL-HPC.



mpirun -n 4 griffin-opt -i htr-10-critical.i

Initial Critical Core Results

The first set of results for this model deal with the htr-10-critical.i input file which is the initial criticality benchmark. Table 1 summarizes the results of Griffin using tensor diffusion coefficients (TDC) and super homogenization (SPH) compared against MCNP and Serpent using ENDFB-VI and Serpent using ENDFB-VII.r1. As can be seen in the table below, Griffin matches with the accepted Serpent solution very well.

Table 1: Eigenvalues computed with different codes for the initial critical configuration. (MCNP results obtained from (IRPEEP, 2006))

Code	$var element = document.getElementById("moose-equation-d4c0bb60-86fb-40d9-93b1-6d1ed6adc43a");katex.render("k_eff", element, {displayMode:false,throwOnError:false});$	uncertainty rel. error (pcm)
MCNP (ENDFB-VI)	1.01190	+/- 21
Serpent (ENDFB-VI)	1.01025	+/- 5.1
Serpent (ENDFB-VII.r1)	1.00023	+/- 2.3
Griffin TDC-SPH-Diffusion	1.00089	67.3

Additionally, the flux distribution in the critical core is plotted in Figure 1 below. The effect of the upper cavity is clearly visible as the flux is almost flat within this region. The thermal flux peaks in the bottom of the core are also visible.

Figure 1: Flux distribution in the critical core (axial centerline) (Ortensi et al., 2018).

Full Core Results

For the Full Core case, there are a multitude of different cross section and equivalence libraries that can be switched out in the htr-10-full.i in order to model the full htr-10 reactor at different temperatures and rod positions. These results are compared against Serpent and results listed in literature.

Effect of Temperature on Reactivity

The first Full Core benchmark problem is to compute the eigenvalues for the full core at three different uniform core temperatures: 27, 120, and 250 degrees C (300, 393, and 523 degrees K respectively). The three different equivalence and cross section libraries that these results require come from the libraries 'htr-10-full-ARO','htr-10-393K', and 'htr-10-523K'. These can be changed in the [Globalparams] block by changing the 'library_name' to the appropriate library for the different temperatures. The results for this study are shown in Table 2 below.

Table 2: Eigenvalues at three temperatures computed with various codes for the full core. The Griffin computed eigenvalues are obtained with SPH corrected diffusion. (Non-INL calculations cited at (IRPEEP, 2003))

Code	$var element = document.getElementById("moose-equation-e6d435b7-ee93-4b33-9aa8-89439d819716");katex.render("k_eff", element, {displayMode:false,throwOnError:false});$ (27 C)	$var element = document.getElementById("moose-equation-d0af1a60-0ec5-4563-a9ea-76568376e330");katex.render("k_eff", element, {displayMode:false,throwOnError:false});$ (120 C)	$var element = document.getElementById("moose-equation-b098b7e4-dc34-4ecc-923c-72ef6118defc");katex.render("k_eff", element, {displayMode:false,throwOnError:false});$ (250 C)
Serpent (INL)	1.12242 +/- 13	1.11068 +/- 20	1.09298 +/- 14
Griffin (INL)	1.12242	1.11061	1.09249
VSOP (China)	1.1358	1.1262	1.1111
MCNP4 (China)	1.1381	-	-
VSOP 2-D (Germany)	1.1468	1.1334	1.1160
VSOP 3-D (Germany)	1.1368	1.1232	1.1054
TRIPOLI4 (France)	1.1474	-	-
VSOP_PBMSR (SA)	1.1286	1.1196	1.1047

Effect of Control Rod Configuration on Reactivity

The last full core problem looks at various control rod configurations and compares the results from Griffin against Serpent. The all rods out (ARO), all rods in (ARI) and one rod in (1RI) results for Diffusion and SPH corrected Diffusion are compared against the Serpent results in Table 3 below. In order to obtain these results the user needs to update the 'library_name' in the [GlobalParams] block to the appropriate libraries 'htr-10-full-ARO','htr-10-full-ARI', and 'htr-10-full-1RI' respectively.

Table 3: Relative difference of Griffin computed eigenvalues (pcm) and reaction rates (%) with the Serpent solution for the full core with different rod configurations (ARO -all rods out, ARI - all rods in, and 1RI - one rod in) (Ortensi et al., 2018)).

Configuration	Eigenvalue - $var element = document.getElementById("moose-equation-2cff02f1-c1f0-41dd-9019-29ac3994b053");katex.render("k_eff", element, {displayMode:false,throwOnError:false});$	Eigenvalue - $var element = document.getElementById("moose-equation-1d185dd9-65a2-45e5-b9c0-758bf87859fb");katex.render("\\Delta", element, {displayMode:false,throwOnError:false});$ (pcm)	Absorption - RMS	Absorption - Max	Generation - RMS	Generation - Max
Diffusion
ARO	1.15096	2542.7	20.6	136.19	6.6	13.3
ARI	0.97094	1061.3	24.6	158.07	10.4	21.6
1RI	1.13118	2378.5	21.3	161.9	6.97	14.6
SPH Corrected Diffusion
ARO	1.12347	93.9	0.2	0.5	0.1	0.19
ARI	0.96145	74.1	0.2	0.51	0.09	0.16
1RI	1.10622	119.4	0.2	0.54	0.1	0.22

From Table 3 above, it's clear that the SPH Correct Diffusion method makes all the difference when it comes to accuracy against Serpent. With the SPH Correction, Griffin can match the eigenvalue and the Absorption and Generation rates calculated by Serpent with maximum errors down to half a percent.

A three-dimensional rendering of the scalar flux in group 10 for the one rod in case is provided in Figure 2 below. The depression of the flux around the inserted control rod positioned just outside the core region is clearly visible.

Figure 2: Rendering of the scalar flux in group 10 (most thermal flux) for the one-rod in configuration. The depression of the flux around the inserted control rod positioned just outside the core is clearly visible (Ortensi et al., 2018)).

Lastly, all three rod configurations can be examined in Figure 3 which is a slice at z = 270 cm (core region) for 1RI, ARI, and ARO. The line plots represent the flux along the light green line shown on the 1RI 2D plot for each of the configurations. The ARO and ARI have symmetric solutions whereas the 1RI has the asymmetric solution due to only one region where the flux is depreciated due to the insertion of only one control rod. The thermal peak at the reflector interface is present for the ARO configuration and still slightly visible for the 1RI along the line out plot as well.

Figure 3: Rendering of the scalar fluxes in group 10 at z = 270 cm (lower core level) for all three rod configurations: one rod in (1RI), all rods in (ARI), and all rods out (ARO) along with traces of the scalar fluxes along a line intersecting the inserted control rod in the 1RI configuration (Ortensi et al., 2018)).

Acknowledgments

This results description section is summarized from (Ortensi et al., 2018) and prepared for the VTB by Samuel Walker.

References

IRPEEP. Evaluation of high temperature gas cooled reactor performance:benchmark analysis related to initial testing of the httr and htr-10. Technical Report IAEA-TECDOC-1382, International Reactor Physics Experiment Evaluation Project, 2003.

@techreport{TECDOC,
    author = "IRPEEP",
    Date-Added = "2016-06-22 18:46:25 +0000",
    Date-Modified = "2016-06-22 18:48:35 +0000",
    Institution = "International Reactor Physics Experiment Evaluation Project",
    Number = "IAEA-TECDOC-1382",
    Title = "Evaluation of high temperature gas cooled reactor performance:Benchmark analysis related to initial testing of the HTTR and HTR-10.",
    Year = "2003"
}

IRPEEP. Evaluation of the initial critical configuration of the HTR-10 pebble-bed reactor. Technical Report HTR10-GCR-RESR-001, International Reactor Physics Experiment Evaluation Project, 2006.

@techreport{IRPhEP,
    author = "IRPEEP",
    Date-Added = "2016-06-22 18:46:25 +0000",
    Date-Modified = "2016-06-22 18:48:35 +0000",
    Institution = "International Reactor Physics Experiment Evaluation Project",
    Number = "HTR10-GCR-RESR-001",
    Title = "Evaluation of the Initial Critical Configuration of the {HTR}-10 Pebble-bed reactor.",
    Year = "2006"
}

J. Ortensi, S. Schunert, Y. Wang, V. Laboure, F. Gleicher, and R. Martineau. Benchmark Analysis of the HTR-10 with the MAMMOTH Reactor Physics Application. Technical Report INL/EXT-18-45453, Idaho National Laboratory, 2018.

@techreport{HTR-10Benchmark,
    author = "Ortensi, J. and Schunert, S. and Wang, Y. and Laboure, V. and Gleicher, F. and Martineau, R.",
    Title = "{Benchmark Analysis of the HTR-10 with the MAMMOTH Reactor Physics Application.}",
    Institution = "Idaho National Laboratory",
    Number = "INL/EXT-18-45453",
    Year = "2018"
}

Running an Input File
Core Results Initial Critical
Results Full Core
Acknowledgments
References