ML20199A350
| ML20199A350 | |
| Person / Time | |
|---|---|
| Issue date: | 11/13/1997 |
| From: | Wang E NRC (Affiliation Not Assigned) |
| To: | Curet H SIEMENS POWER CORP. (FORMERLY SIEMENS NUCLEAR POWER |
| References | |
| TAC-M98478, NUDOCS 9801270193 | |
| Download: ML20199A350 (6) | |
Text
_
. DIE-A6 cvy 3 ,g..s
.,'j
,y t
j
}
UNITED STATES NUCLEAR REGULATORY COMMISSION ,,
wAsHiwatoN, D.C. SteeH001 R hCfy{h a NCO//af':
%, . . . . + 'f November 13, 1997 17?3 3."-9 13 lb 23
'UBLIC 00CL' MENT ROOM Mr. H. D. Curet. Manager Product Licensing Siemens Power Corporation 2101 Horn Rapids Road R P, O. Box 130 ,
Richland WA 99352-0130
SUBJECT:
REQUEST FOR ADDITIONAL INFORMAT!0N (RAI) RELATED TO THE TOPICAL REPORT, "ANF-1125(P) SUPPLEMENT 1 APPENDIX D. ANFB CRITIrAL POWER CORRELATION UNCERTAINTY FOR LIMITED DATA SETS" (TAC NO. 98478) ,
REFERENCES:
(1) Siemens Topical Report. ANFB-1125(P). Supplement 1.
Appendix D "ANFB Critical Power Correlation Uncertainty for Limited Dath fats " Siemens Power Corporation. April 1997.
(2) Attachment to Letter dated August 7, 1997. Responses to Request for Additional Informaticn for Siemens Top, cal Report, ANF-1125(P), Supplement 1. Appendix D "ANFB Critical Power Correlation Uncertainty for Limited Data Sets."
Dear Mr. Curet:
By letter dated August 7, 1997, Siemens Power Corporation (59C) transmitted Responses to Request for Additional Information for Siemens Topical Report, ANF-1125(P). Supplement 1. Appendix D "ANFB Critical Power Correlation Uncertainty for Limited Data Sets" for staff review. The U.S. Nuclear Regulatory Commission (NRC) Reactor Systems Branch (SRXB) stiff has reviewed the responses and has determined that it needs additional information in order
'fM to complete its review of the Siemens Topical Report ANF-1125(P).
Supplement 1, Appendix D.
g
- 1. Questions 3 and 9 were posed in the first round of questions to obtain clarification of the description of the analysis of variance calculations (performed in steps 4 through 7, as described on p. 3-2).
The responses provided by SPC to Outcotions 3(c), 3(d), 9(a), and 9tb) refer specifically to NUREG-CR-4604/PNL-5849, but the responses also seem to show that the actual procedures followed in the SPC analysis differ significantly from any of the methods of analysis of variance described in the referenced document.
gD # , 6%
7
- g=PJE! Tsub
.....llllllll ll]
lllll C PDR 2
- . - _ . . - - . - - . . . ~ . . - . . - -. - _
.v ,_
y ,
November 13, 1997 Mr. H. Curet .Specifically. the procedures described in the responses to Questions 3(c). 3(d); 9(a), and 9(b) do not appear to be consistent with any analysis of variance methods documented-in NUREG CR-4604/PNL-5849. To ,
resolve this apparent contradiction. SPC needs to provide a complete description of the methodology used in their analysis of variance calculations for the ATRIUM 9 data set. This description should include at a minimum, the statistical model(s) used (with clear references to specific equations taken from NUREG-CR-4604/PNL-5849. or any other appropriate reference), any modifications or simplifications-made to the basic model (with appropriate justification), and justification for performing the analysis of variance calculations in nultiple steps (i.e., steps 4. 5. and 7). since NUREG-CR-4604/PNL-5849 describes models that are capable cf performing this analysis in-essentially one step.
- 2. The response to Question 3(d) states that "An analysis of the ATRIUM-9 data using a two-level design from statistical design of experiments affirms that the pressure has a negligible effect and no importance in combination with the flow."
a) describe the two-level design method used in this analysis, with specific reference (including page number (s) and any equation numbers) to the relevant model(s) in NUREG-CR-4604/PNL-5349.
b) shs the major steps in.the calculation (s) involved in this analysis, and explain why the results support the conclusion that the pressure has a negligible effect in combination with the flow.
- 3. The final value of the additive constant uncertainty (ACU) is calculated by dividing the ECPR uncertainty by the factor 1.99. This factor is the average value of the ratio of the ECPR uncertainty and the FEFF uncertainty. as calculated for the data from each bundle in the ATRIUM-9 data set. -The response to Ouestion 4 dn the first round of questions shows that the value 1.99 is a simple linear average of these ratios for
- the 11.. test bundles in the ATRIUM 9 data . set. - The value of this ratio.
varies from 1.80 to 2.39 for the 11 test bundles, so that the standard deviation about the average value of 1.99 is approximately 0.03. The
-_value obtained for the_ACV uncertainty would vary by more than *10% over the range of variation in the ratio. If a typical criterion such as the 2-sigma limit is applied to obtain a bounding value for the ratio. it
.would be approximately 1.93, rather than 1.99. If this value is used 7
J
, - - , n- , ,, . . - ,
a .
November 13, 1997 Mr. H. Curet nstead of 1.99, the ACV uncertainty is calculated as approximately L.0201. an increase of about 3.1%.
a) Justify the use of the arithmetic average of the ratios of ECPR uncertainty and FEFF uncertainty for the 11 ATRIUM-9 test bundles as the appropriate factor for the calculation of the ACU uncertainty from the ECPR uncertainty.
- 4. The response to Question 7 in the first rounJ of questions is incomplete. The procedure for obtaining the value of ECPR = 1.037 for -
the distribution free tolerance limit has been verified by consulting the reference Exrerfmental Statistics, H.G. Natrella NBS Handbook 91.
However, the SPC response contains no explanation of the reported method of estimating the uncertainty of this ECPR value as g ,
(EcPR-ECPR) data point number
,N9 5 where ECPR = experimental CPR calculated for the data point that satisfies the distribution free tolerance limit such that there is a 95% probability that ,
95% of the total population has a smaller ECPR %
value (in this case, the 18th largest ECPR value in the data base, which is 1.037 -- data point 361.1 from test bundle STS9.3)
ECPA
= mean value of ECPR for the full data set ses = number of standard deviations to bound 95%
of the data points of an equivalent normal distribution a) Provide a complete description of the statistical model underlying this method of estimating the uncertainty of a correlation's prediction for a given point in a data set, and justify th application of this model to predictions with the ANFB correlation for the ATRIUM-9 data set.
i I
Mr. H. Curet 4 November 13, 1997
- 5. The effect of different non-uniform. axial power profiles is accounted for by calculating the variance nf the mean ECPR values for the 7x/
ATLAS data sets grouped by axial power profile (as shown in Table 4.6).
a) What is the statistical analysis methodology or model that is '
being employed in this analysis? (Provide complete details on assumptions and equations taken from any references cited.)
b) Justify the application of this methodology or model, explaining why it can be assumed to provide an appropriate means of determining the effect of axial power distribution on the ECPR uncertainty. Alternatively describe a different model that is '
applicable to this problem, and use it to determine an estimate of the effect of axial power distribution on the ECPR uncertainty, for use in the estimate of overall ECPR uncertainty for ATRIUM-9 fuel.
- 6. The conclusion that there is no additional uncertainty in ECPR due to the effects of local pin neaking (see p. 4-5) is not supported by the results presented in Tables 4.6 and 4.7. The differences in the variance between the different groupings of the data are on the order of 10-20%. and show clear differences between groups with nominal peaking (approximately 1.25) and high peaking (approximately 1.4). even when obtained in test bundles with the same axial power distribution. It is not correct to assert that any additional uncertainty due to radial pin peaking "is already present in the variance between the means of the axial shape data from Table 4.6."
a) Using an appropriate statistical model, such as analysis of variance, determine the effect of local pin peaking on the ECPR uncertainty for the 7x7 ATLAS data. Apply this uncertainty to the determination of the overall uncertainty in the ECPR for the ATRIUM-9 data.
',,,c . .
November 13, 1997-Mr. H. Curet You have requested that. portions of the ir. formation, "ANF-1125(P). Supplement 1, Appendix 0, ANFB Critical Power Correlation Uncertair.ty for limited Data Sets." be exempt from mandatory public disclosure. While the staff has not completed its review of your request in accordance_with the requirements of 10 CFR 2.790, that portion of the submitted information is being withheld from public disclosure pending the staff's final determination. -The staff c-oncludes that these questions and comments do not contain those portions of the information for which exemption is sought. However, the staff will withhold this letter from public disclosure'for 30 calendar days from the date -i of 'this letter to allow Siemens the opportunity to verity the staff's conclusions. .If, after that time, you do not request that all or portions of '
- the information be withheld from public disclosure in accordance with 10 CFR
'2.790, this letter will be placed in the NRC Public Document Room.
These questions. affect nine or fewer respondents, and therefore are not subject to review by the Office of Management and Budget under P.L.96-511.
. It will be very helpful if you could respond this RAI within 30 days. If you have any questions regarding this matter, you may contact me at (301) 415-1076.
Sincerely, b
Egan Y. Wang, Reactor Engineer Generic Issue and Environmental' Projects Branch Division of Reactor Program Management-Office of Nuclear Reactor Regulation
- s. *
" ' jd.'H. Curet Novwmber 13, 1997 4
You h6ve requested that portions of the information. "ANF-1125(P). Supplement 1.
Appendix 0. ANFB Critical Power Correlation Uncertainty for limited Data Sets." be exempt from mandatory public disclosure. While the staff has not campleted its review of your request in accordance with the requirements of 10 CFR 2.790, that portion of the submitted information-is being withheld from public disclosure pending the staff's final determination. The staff concludes that these questions and comments do not contain those portions of the information for which exemption is sought. However, the staff will withhold this letter from public disclosure for 30 calendar days from the date of this letter to allow Siemens the opportunity to verify the staff's conclusinns. If, after that time, you do not request that all or portions of the information be withheld from public disclosure in accordance with 10 CFR 2.790, this letter will be placed in the NRC Public Document Room.
These questions affect nine or fewer respondents, and therefore are not subject to review by the Office of Management and Budget under P.L.96-511.
It will be very helpful if you could respond this RAI within 30 days. If you have any questions regarding this matter, you may contact me at (301) 415-1076.
Sincerely.
Original Signed By:
Egan Y. Wang Reactor Engineer Generic IssJe and Environmental Projects Branch Division of Reactor Program Management Office of Nuclear Reactor Regulation
- "************ 30 - day d i s t r i but i on Hol d ***************************
DISTRIBUTION:
PGEB R/F FAkstulewicz TEssig *PDR EWang THuang Bsheron GHolahan TCollins LPhillips DOCUMENT NAME: 971016.RAI r - copy -in.itunm.nu.ncio - . r - noecoy
_ ,. *.goco ni. m .i. ma. w c - oginoot .ituwm.nv.nciom.
0FFICE PGEB g g SC:PGEB/'7' l SC:SRXPf((f AC:PGEB/) D:MSA /
NAME EWang:ss FAk,stitiMWicz LPhiWVfpI/ TEssig '/IQ GHolaM DATE 11/12/97 11/)d/37
~
11//3/97 11/13/97 11/ //97\
OFFICIAL RECORD COPY 4 Q
l
.