A Consistent One-Dimensional Multigroup Diffusion Model for Molten Salt Reactor Neutronics Calculations
Round 1
Reviewer 1 Report
This study is targeting the accurate consideration of flowing fuel in an MSR in computational predictions of neutronics parameters. While the study is limited since it demonstrates the first use of a new 1D model to represent the flowing fuel, it presents a promising method. I have only a few remarks, mainly asking for more specificity and accurateness when writing the conclusions, and for table and figure captions.
-
Abstract:
-
“generated with a high-fidelity three-dimensional (3D) Monte Carlo model developed by Serpent.” --> “with the Monte Carlo code Serpent using high-fidelity three-dimensional models”
-
Body:
-
Introduce all used symbols in the equations
-
Define “infinite spectrum approach”.
-
Figure 3 and 4: Slightly reword captions in a way that indicates that the diffusion calculation is run with XS of a specific group number, and results are compared against Serpent CE. The way it’s currently written is ambiguous and readers might think Serpent is run in multi-group mode which is not a thing.
-
Table V caption: Explicitly indicate“1D diffusion with 2 groups, 8 groups, etc.” in the table caption and/or in the table column headers.
-
Table VI, VII (double check all): Always indicate in table and figure captions if the presented data shows results of calculations (then state which code+model were used) or these are input data (then provide reference).
-
Section 5: Which flow rate was used? The information provided on MSRE is a bit thin. Provide more details so that readers can understand the flow rate (how much time does it take a slug of flow to go from the bottom of the core to the top) vs. average half life of DNP groups.
-
“These results indicate the 1D, leakage corrected diffusion model is sufficient for representing the MSRE Configuration.”, and statements in the conclusions. Please be more specific about the quantities for which the developed approach was verified and validated and avoid generalization.
Some editing of the language is recommended to remove some minor issues in the text.
Author Response
Please see the attached word file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors should improve more on the introduction by explaining more on the purpose of the 1D model aside computational efficiency and accuracy.
On line 62 - 63, the reviewer thinks the author claim to reduce the error in the 1D model by careful generation of homogenized cross sections. The reviewer thinks there will still be modeling biases not yet accounted for especially in comparison to Monte Carlo simulation. 1D versus 3D and multigroup diffusion versus transport, what amount of bias is expected? The authors should comment on these modeling biases and estimate the error.
On the use of a 1D model (which is a reduced order model) for uncertainty/sensitivity analysis, how does the authors intend to tackle the modeling bias/errors that will be propagated to uncertainty/sensitivity calculation results?
The authors should comment on the percentage error/difference in the axial flux solutions comparing the 1D diffusion and 3D Serpent calculations in Figures 3 and 4.
The authors should provide more explanation on how COMSOL was used to compute the developed 1D diffusion model.
What tool did the authors use to implement the multigroup diffusion model equations derived in Section 2? Did they write their own code or use COMSOL? How?
The authors claim the 1D model is computationally efficient, however, the authors did not comment on the comparison of computation time between the 1D and the 3D models.
Author Response
Please see attached word file.
Author Response File: Author Response.pdf
