Validation of the SCALE/Polaris–PARCS Code Procedure With the ENDF/B-VII.1 AMPX 56-Group Library: Boiling Water Reactor

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript entitled “Validation of the SCALE/Polaris−PARCS Code Procedure with 2 the ENDF/B-VII.1 AMPX 56-Group Library: Boiling Water 3 Reactor” presents the SCALE/Polaris–PARCS code procedure for the validation (comparison calculations(C) and measured (E) data) in Peach Bottom-2 (cycles 1−3), Hatch-1(cycles 1−3) and Quad Cities 1 (cycles 1-3 ) BWRs.

General Comments:

A very comprehensive manuscript presenting the validation of ENDF/B-VII.1 AMPX 56-Group Library using SCALE/Polaris−PARCS Code. Uncertainties and biases for reactivity and power distributions have been analysed by comparing calculations and results with the plant-measured data. These bias (C-E) and uncertainties for these three first cycles of operation for Peach Bottom-2, Hatch-1 and Quad Cities 1 are very interesting for licensing purposes to be applied in the BWRs. However, as it is concluded in the manuscript more efforts are needed in Polaris–PARCS system to provide a more reliable and robust tool for users.

Major comments

I have two major comments/questions that I suggest to the Authors additional discussion in the manuscript.

First, in the section 2.2.4 the description of “BWR benchmark calculations” is described. A more detailed description (giving values) of the branches and depletion (history conditions) cases is needed.

Do the Authors use the NUREG – CR 7164 for the description of the cases Tmod, Tfuel, %Void, etc …? Do the Authors check the Branches and historical nodal conditions in the three BWRs . Do they cover the full set of conditions? How many extrapolations were needed? Do they analyse the impact of refinement branches and depletion cases?

If the historical %void distribution goes from a typical value of 70% to 80% or 90% will the bias show any improvement?

In my opinion, the number of points outside of cross-sections domain is a very important question (avoiding potential issues in the extrapolation) that may explain the differences in axial and radial power distributions. I would recommend to the Authors to do this extra work.

Second, do the BWRs have any required physics characteristics to be used for validation of calculations? For PWRs, the “ANSI/ANS-19.6.1-2011 American National Standard Reload Startup Physics Tests for PWRs” is an important way to confirm the validation of calculations in PWRs. I would recommend to the Authors to give some BWR Target values in order to asses the performance of this validation. This may help readers about the impact of the bias and uncertainties.

Some minor comment/typo:

-          In Section 3.1.1., It says:”… The calculated HFP reactivities for PB2, Hatch1 and QC1 are illustrated in Figures 4 and 5 for all cycles …”

 It should say:”… The calculated HFP reactivities for PB2, Hatch1 and QC1 are illustrated in Figures 3 and 4 for all cycles …”

Last comment :

Why do the Authors use ENDF/B-VII.1? Do the Authors plan to extent this work to other recent nuclear data evaluations such as ENDF/B-VIII.0 or VIII.1? Any comment about it is also appreciated.

Author Response

Please find attached file including reviewer's comments and authors' responses.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for submitting your manuscript titled "Validation of the SCALE/Polaris−PARCS Code Procedure with the ENDF/B-VII.1 AMPX 56-Group Library: Boiling Water Reactor" to JNE.

I found your paper to be interesting and informative. However, I recommend a major revision to enhance clarity in certain sections. I will provide general and specific comments following this message.

I am eager to review the revised manuscript.

Kind regards,

Reviewer.

 

---

General comments:

To improve readability, adding a list of all abbreviations at the end would be beneficial. This would enhance the paper's flow and provide a central reference for all abbreviations used. It's important to note that the same abbreviations can have different meanings in various scientific fields. Including this list would assist readers in comprehending the content more easily and enhance the manuscript's overall quality.

Please review the unit dimensions in the manuscript to ensure the use of SI units. Imperial, US customary and other units should be converted to SI units whenever possible.

 

Specific comments:

Line 45.

Please provide a summary of what is included in the two-step procedure.

Line 54.

Could you explain why the 56-group library was chosen? How many groups did the previous library contain? Why was the decision made to limit the selection to only 56 groups and not more? Please provide more details about this choice.

Line 55.

A traversing in-core probe (TIP) is an instrument used in nuclear reactors to measure variables such as neutron flux distributions within the reactor core. It would be helpful to specify what type of data is being discussed here, as the current information is unclear.

Figure 1.

We need more details for this figure. For instance, what do the different color schemes for the fuel pins represent? Is it a variation in material composition, temperature, or something else? Additionally, why were those two particular inserts scaled up?

Figure 2.

To help readers who are not familiar with each part of the flowchart of the Polaris-PARCS procedure for BWR, it would be useful to provide a one-sentence summary for each element shown in the figure. This can be included either in the figure caption or as a brief summary in the text.

Figure 3.

Countries often use different units to represent small variations in the effective neutron multiplication factor (keff). In Canada, for instance, the National Research Universal (NRU) nuclear reactor uses the unit "mk" to indicate changes in keff (Nguyen, S., B. Wilkin, and T. Leung; Current developments and future challenges in physics analyses of the NRU heavy water research reactor; AECL-CW-152600-CONF-001; Atomic Energy of Canada Limited, 2011). To foster better understanding and consistency across various contexts, clarifying the relationship between the unit "pcm" and keff would be advantageous. 

Table 1.

Please include the SI units (in brackets) for the parameters in the table, as this is a requirement from JNE. SI Units (International System of Units) should be used. Imperial, US customary and other units should be converted to SI units whenever possible. Please use SI units for "Core inlet enthalpy (Btu/lb)" and "Core flow rate (Mlb/h)".

Line 126.

Please explain in more detail how the conversions from the AMPX 56-group library to two-group cross sections were carried out. What criteria were used for this process and why were they chosen? One of the key points of this manuscript was the use of the AMPX 56-group library for boiling water reactors. Please provide additional details on why the two-group cross sections were tabulated.

Line 142.

Could you please explain the specific temperature conditions for HFP used in the calculations? Additionally, does HFP pertain to the specifications of the BWR plants listed in Table 1, such as core power?

Figure 5.

Comparing the HFP TIP data for PB2, Hatch1, and QC is challenging when the x and y axes have different scales on the plots. Could you please remake the graphs so that the x and y scales are the same across all 9 plots? This will enhance the readability of the manuscript and make it easier for readers to compare the results.

Line 185.

I disagree with the following statement: "Because there is no measurement uncertainty for reactivity." This should be stated as an assumption for this specific case. Furthermore, even if the authors assume no measurement uncertainty for reactivity in this particular scenario, a conservative approach can still be applied. One could use a typical value for measured reactivity uncertainty based on summary of uncertainties tables, depending on the level of reactivity (see for example: "Dos Santos, Adimir, "Guide to the Expression of Uncertainty," International Reactor Physics Experiments Evaluation Project (IRPhEP), Paris, OECD Report, NEA/NSC/DOC (2017)"). It is possible that the maximum value of measured reactivity uncertainty could contribute to the differences discussed in this work.

Lines 190,191.

The text mentions that a normality test was not conducted for the HFP reactivities and TIP data because there are sufficient data points to assume a normal distribution. However, if there are enough data points, why not confirm that the data are normally distributed? In the manuscript, a 95x95 one-sided tolerance limit is employed to analyze uncertainties for the BWR reactivity and TIP data. Nevertheless, one-sided tolerance limits are suitable for normally distributed data. Therefore, the authors should validate the normality of the data.

Table 2.

The meaning of some abbreviations in the table is unclear. In the field of statistics, "KS" usually refers to the Kolmogorov-Smirnov test.

Line 183 mentions that Sd represents the 'total' standard deviations. In the column labeled 'Method', parametric statistics are presented for 'measurement', 'calculation', and 'difference'. Does this imply that 'total' is synonymous with 'difference'?

Lines 208, 209.

Please check if the data points follow a normal distribution. It is crucial to confirm that the data points follow a normal distribution. If this is not verified and the data points do not follow a normal distribution, one cannot use a one-sided tolerance limit.

Figure 6.

When showing measured values on a comparison graph, it's expected to also display the measurement uncertainties for each data point. This additional information helps in comparing the measured data with the calculated data. Please give more details about the measurement uncertainties for the data shown in the figure.

Lines 234 to 245.

In the model statement discussed in line 185, the uncertainty for the calculated reactivity equals the total uncertainty. In other words, the measurement uncertainty for reactivity was not considered. The significant reactivity differences observed in cycle 1 could potentially be explained by this parameter that was not considered in the model. This needs to be researched further for clarity or explicitly stated as an assumption in this discussion.

Line 247.

Please ensure the use of SI units.

Table 3.

In the first column, are the uncertainties listed for the 'Difference' row the same as the uncertainties for the 'Total'? Does it imply that in the 'Difference' row, 'KSm' should be 'KSd'? Additionally, does 'K9595' mean the same thing as 'K95x95'?

Table 4.

In the first column, are the uncertainties listed for the 'Difference' row the same as the uncertainties for the 'Total'?

Do you have uncertainties for 'Calculation' for this case as they are presented in Figure 7?

Line 275.

It is mentioned here that a sensitivity analysis was conducted for the thermal expansion of PB2. This contradicts the statement in line 144: "However, thermal expansion was not considered for simplicity."

Author Response

Please find attached file including reviewer's comments and authors' responses.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Excellent work. Good response to my comments/questions. No further comments and/or questions.

Author Response

Dear reviewer,

We really appreciate your review and acceptance.

Your valuable review comments for round 1 were very helpful in improving the manuscript. 

Best regards,

Kang Seog

 

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

I appreciate the effort you have put into revising your manuscript titled "Validation of the SCALE/Polaris−PARCS Code Procedure with the ENDF/B-VII.1 AMPX 56-Group Library: Boiling Water Reactor". The revised version explains the comparison of the simulated code results with the measurements for BWR much better.

However, I believe the manuscript still assumes a significant level of prior knowledge from its readers, particularly regarding the specific abbreviations used in the paper. While this is a very good and much-needed research, I feel it would greatly benefit from being more readable. For example, it would be wonderful if a first-year student in nuclear engineering could understand the majority of the paper without needing to look up additional resources/references.

Thank you for your hard work on the research/manuscript and contribution to the field.

Best regards,
Reviewer

Author Response

Comment #1

However, I believe the manuscript still assumes a significant level of prior knowledge from its readers, particularly regarding the specific abbreviations used in the paper. While this is a very good and much-needed research, I feel it would greatly benefit from being more readable. For example, it would be wonderful if a first-year student in nuclear engineering could understand the majority of the paper without needing to look up additional resources/references.

 

 

Response #1

We significant detailed the Section 2.1 as follows:

 

Detailed reactor physics analysis for PWR and BWR has been challenging because it requires significant computing capacity due to complex geometry and large size of reactor core. The most reliable reactor physics analysis for LWR can be performed using continuous energy (~100,000 energy points) Monte Carlo neutron transport calculations with multi-physics analysis, including neutron transport, thermal-hydraulic feedback, and depletion. Because this reactor physics analysis requires high-performance computing using super-computers, it is not practical at all.

A more practical and reliable analysis can be achieved by performing direct 3D deterministic transport calculations with multi-physics using a code package such as the Virtual Environment for Reactor Application (VERA) [17] with coarse-group (~50 groups) neutron cross-section library. The number of energy groups is significantly reduced compared to continuous energy Monte Carlo calculation, but there is almost no approximation in geometry, which requires much less computing capability, making it more practical. However, because this approach still requires several thousand processors to simulate PWR or BWR cores using large-size cluster, it may not be practical, and a more practical approach is necessary.

The most practical analysis procedure is the two-step procedure, for which only serial calculations are performed. Nuclear vendors have developed two-step code packages for LWR analysis. The two-step procedure includes: (a) 2D neutron transport calculations for single assemblies and reflectors adjacent to a single fuel assembly to obtain multiplication factors and neutron flux distributions. Then, assembly- or reflector-homogenized few-group (2 or more) neutron cross sections are obtained by flux-volume weighting and energy group collapsing. (b) 3D few-group nodal diffusion calculations for the whole core are performed using the assembly- and reflector-homogenized few-group cross sections obtained by the first step (a).

Assembly- and reflector-homogenized few-group cross sections and power form factors to be used in pin power reconstruction are obtained by performing 2D lattice transport calculations using Polaris with the ENDF/B-VII.1 AMPX 56-group library and then flux-volume weighting with energy group collapsing from 56-group to 2-group. Figure 1 demonstrates the representative Polaris single assembly model with reflecting boundary condition and the Polaris radial reflector model adjacent to a single assembly to obtain assembly- and reflector-homogenized 2-group cross sections. Various reactor states are considered in depletion and branch calculations. Then, few-group cross sections are tabulated using GenPMAXS as a function of various reactor states such as burnups, moderator densities, moderator and fuel temperatures, and control rod insertions. Once cross section table-sets are generated for fuel assemblies with axial heterogeneities and radial and axial reflectors, 3D whole-core diffusion calculations are performed using PARCS. Figure 2 is a flowchart of the Polaris−PARCS procedure for BWR analysis in which two-phase thermal-hydraulic feed-backs are considered using PATHS. Table 1 provides the six reference cases for each type of fuel assembly and each reference case includes 24 branch states as provided in Table 2.

Typically, the two-step procedure for BWR analysis introduces much larger errors or biases compared to the two-step procedure for PWR analysis. While PWR reactivity is controlled mostly by soluble boron during normal operation, BWR reactivity is controlled only by control rods. Because few-group cross sections cannot be generated that account for all the operation history of control rod movements in the BWR, more errors are introduced in the simulation results. In addition, two-phase thermal-hydraulic models and self-shielded cross sections for high voids would introduce more errors and biases.