ML20035C689
| ML20035C689 | |
| Person / Time | |
|---|---|
| Site: | Wolf Creek  |
| Issue date: | 03/26/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20035C683 | List: |
| References | |
| NUDOCS 9304080320 | |
| Download: ML20035C689 (4) | |
Text
-
E
\\o,i
[
UNITED STATES NUCLEAR REGULATORY COMMISSION j
.Y WASHINGTON, D, C. 20555 o,,,
\\..... *#,c i
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO WOLF CREEK NUCLEAR OPERATING CORPORATION'S REPORT "0VALIFICATION OF STEADY STATE CORE PHYSICS METHODOLOGY FOR WOLF CREEK DESIGN AND ANALSYIS" WOLF CREEK NUCLEAR OPERATING CORPORATION WOLF CREEK NUCLEAR GENERATING STATION DOCKET NO. 50-482
1.0 INTRODUCTION
By letter dated January 15, 1992, Wolf Creek Nuclear Operating Corporation (WCNOC) submitted the report " Qualification of Steady State Core Physics j
Methodology for Wolf Creek Design and Analysis" (Ref.1). This report describes the methodology that WCNOC uses to perform core design and analysis for its Wolf Creek reactor. The report also presents the benchmarking analyses to demonstrate the qualification to perform steady-state core physics calculations for reload design and licensing analysis. Additional information was submitted by letter dated March 12, 1993.
WCNOC is using the standard Ba5cox & Wilcox (B&W) code sets and modeling techniques without modification. The physics code set consisted of NULIF, PEEL, PDQ07, N0ODLE, and FLAME, as well as the interface codes. B&W provided training and technical guidance as well as formal reviews of the Wolf Creek calculations to ensure that WCNOC was using the codes and applying the models accurately and correctly.
To establish the accuracy of the steady-state methods and to determine the reliability factors for safety analyses, WCNOC has performed benchmarking against operating data from the first four cycles of Wolf Creek.
Predictions include:
rod worth, boron worths, the boron letdown curve, reactivity coefficients, fuel pin powers and nodal power distributions. Based on the benchmarks, reliability factors were established for the physics parameters which are input to the reload safety evaluation. Additional data for cycles 5 and 6 has been provided in the submittal dated March 12, 1993 (Ref. 2) and in the Wolf Creek Rod Swap Methodology Report (Ref 3).
~
2.0 EVALUATION The WCNOC's topical report TR-91-0018 WOI documents the methodology, the benchmarking performed to establish the calculational accuracy, and the determination of the reliability factors for important safety parameters.
WCNOC is using standard Babcox & Wilcox code sets and modeling techniques without modification. This decision was made in order to eliminate any uncertainties in the process that might be introduced by WCNOC-specific modifications. B&W provided training and technical guidance as well as formal-9304080320 930326 PDR ADOCK 0500 2
P
~.
reviews of the WCNOC calculations and draft topical reports to ensure that WCNOC was using the codes and applying the models accurately and correctly.
Throughout the benchmarking phases, WCNOC has maintained close contact with the B&W design staff.
B&W design personnel provided several mid-phase reviews of WCNOC calculations as well as further training at the WCNOC site. Since WCNOC has obtained the approved B&W methodology and uses it without modification, this evaluation focused only on the benchmarking analyses and the statistical methods used in determining the reliability factors.
Furthermore, because of the intense involvement of B&W in the WCNOC calculational process, a side-by-side comparison of WCNOC and B&W calculations was not performed.
The Wolf Creek benchmarks were performed against 4 cycles of actual plant data for the topical report.
Since preparation of the repnrt, data has been obtained for cycles 5 and 6.
Measured data has been compared to the calculated data for these cycles as well. The benchmarks to measured data include differential and integral rod worth, soluble poison boron worth, boron letdown with cycle burnup, reactivity coefficients, and power distribution statistical analysis.
2.1 Rod Worth Rod worth measurements are made at the beginning of the cycle at hot zero power (HZP) to verify the predicted shutdown margin. Wolf Creek has used both the boron dilution and the rod exchange methods for rod worth measurement.
The worth of the control banks was measured in cycles 1-4 with boron dilution; in cycle 1, the worth of the shutdown banks was also measured.
Rod exchange was performed in cycles 1, 3, 5 and 6.
The agreement between the measured and predicted values is excellent - less than 10% difference in all but a few low worth cases.
2.2 Boron Worth Comparison of predicted to measured boron endpoints ranges from 0 to -27 PPM with an average value of effectively zero. The differential boron worth comparison between measurement and prediction for HZP shows excellent agreement. The calculation of the boron letdown curve (the critical concentration of soluble boron at 100% rated power, AR0, as core burnup increases during the cycle) is a measure of the ability of the core model to predict the hot critical multiplication factor vs. burnup. The letdown curves for measurements and prediction were examined for cycles 2-4.
The agreements are excellent.
There is essentially no difference between the two curves.
2.3 lemperature Coefficients The is0 thermal temperature coefficient (ITC) is measured with the reactivity computer at HZP. Measurements were compared with predictions for cycles 1-4.
The differences were comparable with those seen throughout the industry for all but one case. The inferred moderator temperature coefficient is obtained by subtracting the predicted Doppler temperature coefficient (DTC) from the ITC. This value was compared with the predicted HTC and the results were
presented in the topical report. The agreement is satisfactory. Doppler only power coefficient measurements were made at the beginning of cycle 1.
In general, the power coefficient measurement is not extremely accurate. The WCNOC measurements showed an average of 5.3% difference, which is considered good.
2.4 Power Distributions The power distribution measurements which consist of in-core flux maps are used to determine the core peaking factors. A total of 48 flux maps were used in the WCNOC analysis (38 from cycles 1-4 which appeared in the original topical report plus three from cycle 5 and seven from cycle 6).
In the Wolf Creek reactor there are 58 instrumented locations.
Except for a few flux maps, measurements were taken at all 58 detector locations. The axial signals are collapsed to 49 axial levels. The three bottom and the six top levels are omitted because the power level is very low at these locations and the measurement to prediction relative differences are large due to the steep gradients. The sum total of FQ comparisons was 108,280 and the total of FAH comparisons was 2707.
Based on this database the reliability factor was 1.063 for FQ and 1.042 for F6H. Values of 1.075 for FQ and 1.050 for FAH will be used for safety and margin calculations.
2.5 Other Physics Parameters Other physics parameters such as Beta-effective and the prompt neutron lifetime are not measurable. These parameters were calculated and compared to values calculated for similar cores by other vendors. The Beta-effective values match-up very well. There is quite a difference in the calculated value for prompt neutron lifetime. The licensee has stated that, "the WCNOC safety analysis staff have performed sensitivity studies on transients to determine the effect of varying the input value of the prompt neutron lifetime.
In all cases the effect on total system performance was negligible".
2.6 Continuina Model Validation and Verification Thus far the WCNOC database is the sum of the comparisons of physics measurements to calculated parameters of the first six cycles of operation at Wolf Creek. The database ill be updated as an integral part of the core follow program from cycle 6 forward. The results of the physics tests at the beginning of cycle as well as the measurements that are part of the core follow program will be evaluated and added to the database to update and verify the reliability factors and biases used in the safety and margin calculations.
2.7 Model Aeolications to Reactor Operations Using the methodology described, WCNOC will perform all the required calculations to determine the depletion characteristics of the final core
design for each cycle and to provide the necessary input to Plant Operations for initial startup,- support of cycles operations, core follow, and Technical Specification surveillance.
3.0 CONCLUSION
The staff has reviewed the Wolf Creek Nuclear Operating Corporation topical report " Qualification of Steady State Core Physics Methodology for Wolf Creek Design and Analysis" and the supplemental information provided by letter of March 12, 1993. The methodology is the B&W methodology which has previously been approved. The WCNOC benchmarking results compare well with the measured data. WCNOC has used these benchmarks to establish reliability factors for the physics parameters for which input is provided to the reload safety evaluation. Based on the staff evaluation in' Section 2.0 above, the staff concludes that the methodology presented in the Topical Report 91-0018 W01,
" Qualification of Steady State Core Physics Methodology for Wolf Creek Design and Analysis", as supplemented by the letter dated March 12, 1993, is i
acceptable for use by WCNOC for future cycles of the Wolf Creek Station.
4.0 REFERENCES
t 1.
F. T. Rhodes (WCNOC) letter to NRC dated January 15, 1992, submitting Topical Report TR-91-0018 WOI.
2.
B. D. Withers (WCNOC) letter to NRC dated March 12, 1993.
3.
F. T. Rhodes (WCNOC) letter to NRC dated August 21, 1990, submitting.
Topical Report TR-90-0024 WOI.
Principle Contributor:
M. Chatterton Date:
March 26, 1993 l
2 l
l
-