ML032410278

Results of Kewaunee SDP Phase 2 Notebook Benchmarking Visit

ML032410278
Person / Time
Site:	Kewaunee
Issue date:	08/28/2003
From:	Reinhart F Division of Systems Safety and Analysis
To:	Richards S NRC/NRR/DIPM/IIPB
Franovich M, NRR/DSSA/SPSB, 415-3361
References
Download: ML032410278 (23)

Text

August 28, 2003 MEMORANDUM TO: Stuart Richards, Chief Inspection Program Branch Division of Inspection Program Management Office of Nuclear Reactor Regulation Patrick D. OReilly Operating Experience Risk Applications Branch Division of Risk Analysis and Applications Office of Nuclear Regulatory Research FROM: Mark F. Reinhart, Section Chief /RA/

Licensing Section Probabilistic Safety Assessment Branch Division of Systems Safety and Analysis Office of Nuclear Reactor Regulation

SUBJECT:

RESULTS OF THE KEWAUNEE SDP PHASE 2 NOTEBOOK BENCHMARKING VISIT During July 2003, NRC staff and contractors visited the Kewaunee Nuclear Power Plant (KNPP) in Kewaunee, Wisconsin to compare the KNPP Significance Determination Process (SDP)

Phase 2 notebook and licensees risk model results to ensure that the SDP notebook was generally conservative. The current plant probabilistic risk assessments (PRAs) internal event core damage frequency was 5.43E-5/reactor-year excluding internal flood events and inter-system loss of coolant accidents. The KNPP PRA did include an integrated PRA model with some external initiating events (i.e. fire initiators). Therefore sensitivity studies were performed to determine any impact of fire initiators on SDP color determinations. In addition, the results from analyses using the NRCs draft Revision 3i Standard Plant Analysis Risk (SPAR) model for KNPP were also compared with the licensees risk model. The results of the SPAR model benchmarking effort will be documented in the next revision of the SPAR (revision 3) model documentation.

In the review of the KNPP SDP notebook for the benchmark efforts, the team determined that some changes to the SDP notebook were needed to reflect how the KNPP is currently designed and operated. Thirty seven hypothetical inspection findings were processed through the SDP notebook and compared with the licensees related importance measures. Using the revision 0 SDP notebook, the team determined that 31 percent of the cases were less conservative, 26 percent of the cases were more conservative, and 43 percent of the cases were consistent with the licensees results. Of the conservative cases, 3 cases were two colors greater than the results obtained using the licensees model. Consequently, 24 changes were made to the SDP notebook.

CONTACT: Mike Franovich, SPSB/DSSA/NRR 415-3361

S. Richards 2 P. OReilly Using the revised SDP notebook, the team determined that 8 percent of the cases were less conservative, 32 percent of the cases were more conservative, and 60 percent of the cases were consistent with the licensees results. Of the conservative cases, all but 4 cases were one order of magnitude greater than the results obtained with the licensees model and as such were generally consistent with the expectation that the notebooks should be slightly conservative when compared with the licensees model.

Although the SDP notebook does not include external initiators, the team compared the SDP results against the licensees PRA model including internal floods and internal fire. Eight cases in the benchmark target-set would increase in importance by one order of magnitude using the licensees model. Of the eight cases, the notebook would under-estimate five cases. Those cases were unavailability of:

C motor-driven auxiliary feedwater pump A C an emergency diesel generator C an instrument air compressor C battery BRA-101 C a component cooling water heat exchanger The licensees PRA staff had substantial knowledge of both the Kewaunee PRA model and conduct of plant operations. The licensees comments greatly improved the quality and content of the SDP notebook.

Attachment A describes the process and specific results of the comparison of the KNPP SDP Phase 2 Notebook and the licensees PRA.

Attachment:

As stated

S. Richards 2 P. OReilly Using the revised SDP notebook, the team determined that 8 percent of the cases were less conservative, 32 percent of the cases were more conservative, and 60 percent of the cases were consistent with the licensees results. Of the conservative cases, all but 4 cases were one order of magnitude greater than the results obtained with the licensees model and as such were generally consistent with the expectation that the notebooks should be slightly conservative when compared with the licensees model.

Although the SDP notebook does not include external initiators, the team compared the SDP results against the licensees PRA model including internal floods and internal fire. Eight cases in the benchmark target-set would increase in importance by one order of magnitude using the licensees model. Of the eight cases, the notebook would under-estimate five cases. Those cases were unavailability of:

C motor-driven auxiliary feedwater pump A C an emergency diesel generator C an instrument air compressor C battery BRA-101 C a component cooling water heat exchanger The licensees PRA staff had substantial knowledge of both the Kewaunee PRA model and conduct of plant operations. The licensees comments greatly improved the quality and content of the SDP notebook.

Attachment A describes the process and specific results of the comparison of the KNPP SDP Phase 2 Notebook and the licensees PRA.

Attachment:

As stated Distribution: SPSB: r/f

• See previous concurrence Accession #ML032410278 C:\ORPCheckout\FileNET\ML032410278.wpd NRR-096 OFFICE SPSB SPSB:SC Region III NAME *MFranovich:nxh2 *MReinhart SBurgess DATE 08/28/03 08/28/03 08/07/03 OFFICIAL RECORD COPY

SUMMARY

REPORT ON BENCHMARKING TRIP TO KEWAUNEE NUCLEAR POWER PLANT (July 15-17, 2003)

P.K. Samanta Brookhaven National Laboratory Upton, NY 11973 August 2003 ATTACHMENT A

TABLE OF CONTENTS Page

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. Summary Results from Benchmarking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

3. Additional Proposed Modifications to SDP Worksheets . . . . . . . . . . . . . . . . . . . . . . . . 10 3.1 Specific Changes to the Rev. 0 SDP Worksheets for Kewaunee . . . . . . . . . . . 10 3.2 Generic Changes in IMC 0609 for Guidance to NRC Inspectors . . . . . . . . . . . 12 3.3 Generic Change to the SDP Notebooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

4. Discussion on External Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 : List of Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 List of Tables Page Table 1: Summary of Benchmarking Results for Kewaunee Nuclear Power Plant . . . . . . . . . 6 Table 2: Comparative Summary of Benchmarking Results . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 3. Comparison of Risk Significance With and Without External Initiators at Kewaunee Nuclear Power Plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

-ii-

1. INTRODUCTION A Benchmarking of the Risk-Informed Inspection Notebook for the Kewaunee Nuclear Power Plant was conducted during a plant site visit on July 15-17, 2003. NRC staff (S. Burgess, M.

Franovich, R. Perch, and M. Tschiltz) and BNL staff (P. Samanta) participated in this Benchmarking exercise.

In preparation for the meeting, BNL staff reviewed the SDP notebook for the Kewaunee Nuclear Power Plant and evaluated a set of hypothetical inspection findings using the Rev. 0 SDP worksheets. In addition, NRC staff provided the licensee with a copy of the meeting protocol.

The major milestones achieved during this meeting were as follows:

1. Recent modifications made to the Kewaunee Nuclear Power Plant PRA were discussed for consideration in the Kewaunee SDP Rev. 1 model to be prepared following benchmarking.

2. Importance measures, including the Risk Achievement Worths (RAWs) for the basic events in the internal event model for average maintenance, were obtained from the licensee.

3. Benchmarking was conducted using the Rev. 0 SDP model and the revised SDP model considering the licensees input and other modifications that were judged necessary based on comparison of the SDP model and the licensees detailed model.

4. For cases where the color evaluated by the SDP notebook differed from the color generated by the updated licensees PRA RAW values, a judgment about the difference was made based on the detailed base case results available for the plant. Minimal cutsets evaluating the impact of the hypothetical inspection findings were reviewed to identify the reasons for the differences.

5. RAWs for the hypothetical inspection findings were also obtained using the model that included internal fire and flooding scenarios. RAWs from the internal event model and the model including fire and flooding were compared to note if the inclusion of fire and flooding resulted in an order of magnitude difference in the risk significance, i.e.,

whether the inclusion of fire and flooding would change the significance of the inspection finding.

The Rev. 1 version of the Kewaunee Nuclear Power Plant SDP notebook was prepared based on the revised plant PRA and lessons learned from the benchmarking at the site.

2.

SUMMARY

RESULTS FROM BENCHMARKING Summary of Benchmarking Results Benchmarking of the SDP notebook for the Kewaunee Nuclear Power Plant (KNPP) was conducted comparing the order-of-magnitude results obtained using the notebook with that obtained using the plant-specific PRA. Cases for which the SDP notebook results were underestimated or overestimated were identified. Two cases of a conservative result by three orders of magnitude (i.e., the significance obtained using the notebook was three colors higher than that obtained using the plant PRA), two cases of conservative results by two orders of magnitude, and eight cases of conservative results by one order of magnitude were noted. In addition, three cases of underestimation by one order of magnitude were noted. A summary of the results of the risk characterization of hypothetical inspection findings was as follows:

8% (3 of 37 cases) underestimation of risk significance 5% (2 of 37 cases) overestimation of risk significance by three orders of magnitude 5% (2 of 37 cases) overestimation of risk significance by two orders of magnitude 22% (8 of 37 cases) overestimation by one order of magnitude 60% (22 of 37 cases) consistent risk significance.

Detailed results of Benchmarking are summarized in Table 1. Table 1 consists of eight columns. The first two columns identify the components/failed operator actions or the cases used for benchmarking. The assigned colors from the Rev. 0 SDP worksheets without incorporating any modification from the Benchmarking exercise are shown in the third column.

The fourth column gives the basic event name in the plant PRA used to obtain the risk achievement worth (RAW) for the out-of-service component or the failed operator action. The fifth and sixth columns respectively show the licensees internal RAW value and the color to be defined based on the RAW values from the latest PRA model. The seventh column presents the colors for the inspection findings based on the Rev. 1 version of the notebook. The Rev. 1 version of the notebook was prepared considering the revisions to the Rev. 0 version of the SDP notebook judged applicable following Benchmarking. The last column provides comments identifying the difference in results between the Rev. 1 SDP notebook and the plant PRA.

Table 2 presents a summary of the comparisons between the results obtained using the KNPP notebook and the plant PRA. It also shows a comparison of the results using the Rev. 0 and Rev. 1 versions of the notebook. The results show that in approximately 82% of cases the risk significance obtained using the Rev. 1 version of the notebook was either comparable or conservative within one order of magnitude compared to the plant PRA. The remaining 18%

cases include underestimation by one order of magnitude or overestimation by more than one order of magnitude. These differences result from differences in modeling assumptions and differences in the unavailability used in the plant PRA and the generic mitigation credit used in the notebooks. The reasons for the differences are discussed further below. The results also show that benchmarking significantly improved the notebook reducing the underestimations and improving the consistency (or matches) in the results. The underestimation was reduced from 31% to 8% and the matches increased from 43% to 60%.

Kewaunee PRA Modeling and Assumptions In comparing the risk significance of hypothetical inspection findings using the Rev. 1 notebook and the results of the plant PRA, some differences were noted between these models. The major factors that contributed to the differences in risk significance are summarized here.

1. Common-cause failure of Auxiliary Feedwater Pumps The Kewaunee PRA models a common-cause failure of all three of the auxiliary feedwater (AFW) pumps. Two of the pumps were motor-driven and the third pump was turbine-driven. In the notebook, a common-cause failure was assumed for the motor-driven pumps with a credit of 1 multi-train system for both motor-driven pumps. In addition, a credit of 1 ASD train was assigned for the turbine driven pump. The combined mitigation credit for the AFW function was 4. In the Kewaunee PRA, the common-cause failure probability of all three AFW pumps was approximately 3E-3. This resulted in an order of magnitude difference in credit for AFW pumps contributing to differences in risk significance between the notebook and the plant PRA.

2. Modeling Shutdown Cooling (SDC) in SLOCA The Kewaunee PRA assumed successful termination of a small LOCA following success of auxiliary feedwater (AFW), high pressure injection (HPI), and RCS cooldown and depressurization (RCSDEP). The implicit assumption was that the SDC mode will be initiated at some time. In the notebook, it was assumed that low pressure recirculation will be needed. This assumption difference also contributed to the difference in risk significance relating to operator actions and equipment involved in low pressure recirculation.

Discussion of Non-conservative Results or Underestimations by the Notebook Three cases of underestimation or non-conservative results were noted during the Benchmarking. They relate to the TDAFW pump, TSC diesel, and operator failure to initiate feed and bleed (FB). The reasons for these underestimations are discussed below:

1. A dominant contributor for the TDAFW pump failure involves the loss of offsite power (LOOP) initiator. This scenario involves station blackout (LOOP followed by the loss of the EDGs), loss of the TDAFW, and the failure to recover offsite power within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

Differences in credit for these functions between the notebook and the plant PRA results in the underestimation. The LOOP frequency was 3.39E-2 (the notebook approximation was 1E-2), the common-cause failure probability of the EDGs was 3.01E-3 (the notebook credit corresponds to 1E-3), and the failure to recover offsite power has a probability of 3.38E-1 (the notebook credit corresponds to 1E-1). The combination of these differences result in an order of magnitude underestimation.

1. TSC diesel was underestimated because of reasons similar to that for the TDAFW pump, as discussed above. The dominant contributor involves the LOOP initiator followed by the common-cause failure of the EDGs, failure of the TSC diesel, and failure to recover offsite power within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. The difference in LOOP frequency, EDG common-cause failure probability, and the probability to recover offsite power results in an order of magnitude underestimation.

2. The risk significance of the operator failing to conduct feed and bleed was underestimated because of the difference in modeling common-cause failure of AFW pumps. As discussed earlier, because of the difference in modeling common-cause failure of AFW pumps, there was an order of magnitude difference in credit for the AFW pumps.

Operators would perform primary feed and bleed following failure of the AFW pumps, and the order of magnitude difference in the credit resulted in the underestimation.

Discussion of Conservative Results by the Notebook Twelve cases of overestimation (two cases by three colors, two cases by two colors, and eight cases by one color) were noted during Benchmarking. The reasons for these overestimations primarily relate to the differences in assumptions between the notebook and the plant PRA, differences in the PRA input data and the notebook generic credits, and the screening approach used in SDP evaluations.

An overestimation by three colors was noted for a battery charger and the operator failure to conduct low pressure recirculation (LPR). The reasons for these three orders of magnitude overestimations are discussed below.

1. In evaluating the risk significance of a battery charger using the notebook, it was assumed that the battery charger failure will lead to failure of the corresponding DC bus. At the Kewaunee plant, there was a swing charger which can be aligned but was not credited in the PRA. However, there was a procedure for aligning the charger and it was expected to be aligned within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. The failure of the DC bus due to failure of the battery charger was also expected to be identified within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and, accordingly, in the PRA model the failure of the battery charger has a low likelihood of failing the DC bus. The assumption in the SDP screening evaluation that the battery charger failure will remain undetected and will lead to failure of the corresponding DC bus resulted in the three orders of magnitude overestimation.

2. Operator failure to conduct LPR was overestimated by three orders of magnitude because of the difference in assumptions between the notebook and the plant PRA in modeling SLOCA, as discussed earlier. Requiring LPR in a SLOCA following successful depressurization when AFW and HPI functions have been successful significantly increases the significance of the LPR function.

Overestimation by two orders of magnitude was noted for two cases: 1 boric acid transfer (BAT) pump and 1 SG PORV. The reasons for these overestimations are discussed below.

1. The BAT pump was used in conducting emergency boration in an ATWS and in carrying out RWST refill. ATWS frequency in the plant PRA was an order of magnitude lower and the probability of operator failure to carry out RWST refill dominates the failure of the RWST refill function. In the notebook, a failure of one of the BAT pumps was evaluated considering the baseline credit of the sequences involving the RWST refill function which was based on the operator failure because of its dominant error probability. This, along with the difference in the ATWS frequency, resulted in the two orders of magnitude overestimation.

2. Overestimation by two orders results for one SG PORV failure because of the lack of recovery credit application in the benchmarking exercise and the SDP process of counting base case credits even when multiple redundancies were left. The plant has a procedure for manually opening the SG PORV which was not credited in the SDP evaluation. Also, 5 safety valves were available to carry out the steam relief function in case of failure of the SG PORV. Credits for the additional redundancies were limited resulting in the overestimation.

In addition, eight cases of overestimation by one order of magnitude were noted. Some of the overestimations resulted from the difference in initiating event frequency between the notebook and the plant PRA. These cases are not discussed further here.

Changes Incorporated Following Benchmarking Resulting in Updating of Benchmarking Results No significant change was made to the notebook following benchmarking that resulted in changes to the benchmarking results.

Table 1. Summary of Benchmarking Results for Kewaunee Nuclear Power Plant Internal Events CDF = 5.43E-5/reactor-yr, excluding internal flooding, at Truncation Level of 1E-10 RAW Thresholds are: W = 1.02, Y = 1.18, R = 2.84, RR=19.42(1)

No. Component Out SDP Basic Event Name RAW Plant CDF SDP Comments of Service or Before Color After Failed Operator Action Component

1. 1 Accumulator Y 33-TK-ACC1A-RP 1.02 W Y over by one order of magnitude

2. 1 MDAFW pump Y 05BPM-AFW1B-PR 1.42 Y Y (B)

3. 1 TDAFW pump Y 05BPT-AFW1C-TM 2.92 R Y under by one order of magnitude

4. 1 CCW pump Y 31-PM-CCW1B-PR 2.52 Y Y

5. 1 CCW HX Y 31-HX-CCW1B-LK 1.42 Y Y

6. 1 Charging pump W 35-PM-CHGP1B-PS 1.41 Y Y

7. 1 Boric Acid W 35-PM-BATPIA-PS 1.0 G Y over by two orders of Transfer Pump magnitude

8. 1 HPSI pump (B) Y 33-PM-SI1B-PR 1.41 Y Y

9. 1 Condensate G 03-PM-CDP1A-PR 1.0 G G pump (Truncated)

10. 1 MFW pump G 05APM-FWP1A-PR 1.01 G G (Truncated)

11. 1 EDG Y 10-GE-DG1B-PR 2.94 R R

12. 1 IA compressor G 01-CM-SIAC1F-PR 1.01 G G

13. 1 SG PORV W 1.0 G Y over by two orders of magnitude

14. 1 PORV FTO W 36-AV-PR2B-FO 1.03 W W

15. 1 PORV FTC R IE-SLO 1.56 Y Y RAW for SLOCA was used; thats how PRA models SORV.

No. Component Out SDP Basic Event Name RAW Plant CDF SDP Comments of Service or Before Color After Failed Operator Action

16. 1 Primary SRV W 36-SV-33113-FO 1.03 W W FTO

17. 1 RHR pump (A) Y 34-PM-RHR1A-PS 1.28 Y Y

18. 1 RHR HX (A) Y 34-HX-RHR1B-LK 1.08 W Y over by one order of magnitude

19. 1 SW pump W 02-PM-SW1A2-PR 5.52 R R

20. 1 AC Bus (Bus 5) R 39-BS-BUS5-SG 19.17 R RR over by one order of (Case Run) magnitude

21. 1 DC Bus (BRA- R 38-BS-BRA 102-SG 40.56 RR RR 102) (Case Run) 22 1 DC Bus (BRB- R 38-B5-BRB 104-SG 7.25 R RR over by one order of 104) (Case Run) magnitude

23. 1 battery (BRA- R 38-BY-BRA 101-OP 16.53 R R 101)

24. 1 Battery Charger R 38-BC-BRA 108-OP 1.03 W RR over by three orders of (BRA-108) (Case Run) magnitude

25. RHR Pump pit W 17-FNPMPPITA-PS 1.28 Y Y Fan coil Unit A

26. TSC Diesel NA 10-GE-TSC-DG-PR 1.65 Y W under by one order of magnitude

27. 1 MSIV FTC Y 06-AV-MS1A-FC 1.38 Y R over by one order of magnitude

28. 1 PORV Block Y - NA NA W Comparable RAW not valve FTC available 29 Piggyback valve NA - NA NA G Comparable RAW not RHR-299 available Operator Actions

30. Operator fails to W 05A-MF2- - -HE 2.84 Y Y recover MFW No. Component Out SDP Basic Event Name RAW Plant CDF SDP Comments of Service or Before Color After Failed Operator Action

31. Operator fails to Y 36-OBF- - - HE 10.58 R Y under by one order of initiate FB magnitude

32. Fails HPR R 36-2TRN-REC-HE 3.62 R R

33. Operator fails to W 35-CH2- - - HE 1.60 Y R over by one order of start a charging magnitude pump

34. Fails LPR Y 34-LR1- - - HE 1.10 W RR over by three orders of magnitude

35. Fails RCSDEP in W 06-OC1- - - HE 1.01 G W over by one order of SLOCA magnitude

36. Isolation of Y 06-IS2- - - HE 3.37 R R ruptured SG

37. RCS cooldown Y 06-OC3- - - HE 3.33 R R and depressurization in SGTR

38. Refill RWST in NA Case Run 1.45 Y Y SGTR

39. Emergency W Case Run 1.0 G W over by one order of boration in ATWS magnitude Note:

1. RR signifies a risk impact between 1E-3 and 1E-2.

Table 2: Comparative Summary of Benchmarking Results Comparisons Rev. 0 SDP Notebook Rev. 1 SDP Notebook (Following Benchmarking)

Total Number of Cases Comparable = 37 Number of Cases Percentage Number of Cases Percentage SDP: Less Conservative 11(1) 31 3 8 SDP: More one order 8 23 8 22 Conservative two orders 1 3 2 5 three orders 2 5 SDP: Matched 15 43 22 60 Comparable RAW not available or 4 2 not modeled in the Notebook Note:

1. 10 cases by one order of magnitude and 1 case by two orders of magnitude.

3. ADDITIONAL PROPOSED MODIFICATIONS TO SDP WORKSHEETS 3.1 Specific Changes to the Rev. 0 SDP Worksheets for Kewaunee The changes made to the Kewaunee Nuclear Power Plant notebook to develop the Rev. 1 version during and after the plant onsite benchmarking visit are summarized here and are also included in the updated notebook.

Changes made to the Kewaunee Nuclear Power Plant Rev. 0 Notebook to complete the Rev. 1 Notebook

1. Changes to Table 1 1.1 Loss of Component Cooling Water (LCCW) was moved from Row III to Row II based on the revised plant-specific frequency.

1.2 Loss of one 125 VDC bus was replaced with Loss of 125 VDC BRA 102 (LDCA) and loss of 125 VDC Bus BRB 104 (LDCB) initiators. Both LDCA and LDCB were placed in Row II.

2. Changes to Table 2 2.1 AFW dependency on SW was removed. However, it was noted that HVAC was needed for AFW pump A which in turn depends on SW.

2.2 A footnote was added stating that HVAC requirement was for CCW Pump B only.

2.3 A footnote was added for charging pumps to note that one AC bus supports two of the charging pumps.

2.4 It was noted that CCW dependency for HPSI and RHR pumps was for recirculation phase only.

2.5 TSC diesel was added as a separate row. A footnote was added to discuss the operation of the TSC diesel.

2.6 A footnote was added to explain the alignment of the air compressors.

2.7 A footnote was added to note that SG PORVs have backup nitrogen accumulators.

2.8 Separate rows were added for RWST and Reactor Water Makeup Pumps.

3. Changes to the Worksheets (Table 3) and Event trees 3.1 RWST refill following failure of recirculation (HPR and/or LPR) was credited for many initiators. The applicable initiators were TRANS, TPCS, LCCW, LDCA, LDCB, SLOCA, SORV, and MSLB. In case of LCCW, HPR was lost due to the initiator. Applicable worksheets and event trees were modified.

3.2 Credit for starting the MFW in the PCS and MFW function was changed from operator action = 3 to operator action = 2 based on plant-specific HEP. In the LOIA worksheet, it was changed to operator action = 1.

3.3 LOSW worksheet and event tree were modified to include operator failure to trip the RCPs.

3.4 LCCW worksheet and event tree were modified to include high pressure injection and RWST refill to mitigation RCP seal LOCA. Revised plant analyses showed that High pressure injection pumps do not need CCW for the injection mode.

3.5 Loss of 125 VDC Bus BRA 102 (LDCA) and Loss of 125 VDC Bus BRB 104 (LDCB) worksheets were included replacing the single loss of DC Bus (LDC) worksheet.

3.6 Loss of Instrument Air (LOIA) worksheet was modified to include the bleed and feed capability. Pressurizer PORVs have backup accumulators.

3.7 SLOCA worksheet and event tree were modified to require low pressure recirculation (LPR) following success of HPI, AFW/MFW, and RCS depressurization. Plant PRA does not require LPR; it assumed success. Operator action credit for LPR was changed from 2 to 3. Also, RCSDEP mitigation capability description was clarified.

3.8 SORV worksheet and event tree were modified similar to SLOCA. In addition, credit for the block function was changed from operator action = 1 to 1 train.

3.9 MLOCA worksheet and event tree were modified to eliminate RCS depressurization.

Core damage was assumed following failure of high pressure injection function in a MLOCA.

3.10 In the LLOCA worksheet, credit for LPR was changed from operator action = 2 to operator action = 3.

3.11 LOOP worksheet and event tree were modified to include the use of the TSC diesel which can support two of the charging pumps following a station blackout. The team did not credit the TSC diesel with charging pumps for reactor coolant pump (RCP) seal LOCA prevention. TSC diesel operation which charging pumps was credited for RCP seal LOCA mitigation in the station blackout (SBO) portion of the LOOP event tree. NRC position (2003) on the WOG 2000 RCP seal LOCA model was that there would be a 20 percent chance of a seal LOCA if seal cooling is not restored in approximately 15 minutes of a SBO initiator (thermal-hydraulic instability phenomena). At KNPP, the TSC diesel would automatically start; however, the charging pumps would be manually loaded in approximately one hour and therefore would not have been available to prevent a seal LOCA.

For long term loss of seal cooling, the probability of seal failure was negligible for seals that are qualified for high temperature exposure which KNPP had installed. In the SBO case, the TSC diesel and charging pump(s) extend the time to core uncovery for seal LOCA scenarios. Combined with the 9 hour1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> estimated time to battery depletion, KNPP could withstand SBOs of longer duration (assuming successful TDAFW pump operation).

In addition, charging and CCW pumps for seal cooling were explicitly modeled following success of EDGs (SEAL function, seal LOCA prevention).

3.12 LEAC worksheet and event tree were modified to delete sequences involving successful closure of the PORVs. These sequences duplicate sequences in the LOOP worksheet.

RCSDEP function mitigation capability was also revised.

3.13 SGTR worksheet and event tree were modified to require RWST refill following successful high pressure injection but failure to isolate the faulted generator or to equalize pressure. SGI function mitigation capability was revised to explicitly include the MSIV on the ruptured SG. EQ function mitigation capability was revised to include the condenser dump valves.

3.14 In the ATWS worksheet, the mitigation capability for long term shutdown through emergency boration was changed to include the RWST suction path. The AFW function mitigation capability was revised to 2/3 AFW pumps. The steam relief path through 1 of 2 SGPORVs or 1/5 safety valves for each SG was added.

3.2 Generic Changes in IMC 0609 for Guidance to NRC Inspectors None.

3.3 Generic Change to the SDP Notebooks None.

4. DISCUSSION ON EXTERNAL EVENTS The Kewaunee Nuclear Power Plant integrated PRA model includes internal floods and internal fire.

The CDF in the integrated model including fire and flooding was 2.87E-4/reactor-yr. The integrated model was used to assess whether the inclusion of the external initiators will result in increased risk significance for components or operator actions. The assessment was carried out by evaluating the RAWs for a set of components and operator actions for the model that included the fire and flood initiators and then, comparing them with the RAWs calculated previously for internal initiators.

Table 3 presents the comparisons for the same set of components and operator actions that were used for benchmarking. Obtaining RAW for some items required separate computer runs which were not conducted for the integrated model. RAW for these items were not available and are noted as NA.

To obtain the color for the component being out of service or the failed operator action, new thresholds were obtained. A comparison of the RAWs for the internal initiators with those obtained using fire and flood initiators showed that in eight cases the color or the risk significance would have increased by an order of magnitude if the risk contributions of external initiators were included.

These items are noted in the table.

Although the KNPP SDP notebook does not include external initiators, the team compared the SDP after results of Table 1 against the licensees PRA model including internal floods and internal fire.

In the eight cases noted above, the notebook would under-estimate five cases. Those cases were unavailability of:

C motor-driven AFW pump A C an emergency diesel generator C an instrument air compressor C battery BRA-101 C a component cooling water heat exchanger Table 3. Comparison of Risk Significance With and Without External Initiators at Kewaunee Nuclear Power Plant CDF including internal flooding and fire = 2.87E-4/reactor-yr at truncation level of 1E-10 RAW Thresholds are: W=1.003, Y=1.035, R =1.35, and RR = 4.47(1)

No. Component Out Basic Event Name SDP After Plant CDF RAW Plant CDF color Comments of Service or (from Color for including including Failed Operator Table 1) Internal external external Action Event initiators initiators Component

1. 1 Accumulator 33-TK-ACC1A-RP Y W NA

2. 1 MDAFW pump 05BPM-AFW1B-PR Y Y 1.26 Y Risk significance (B) (3.23 for (R for Pump A) increases by one order Pump A) for pump A

3. 1 TDAFW pump 05BPT-AFW1C-TM Y R 1.41 R

4. 1 CCW pump 31-PM-CCW1B-PR Y Y 2.74 R Risk significance increases by one order

5. 1 CCW HX 31-HX-CCW1B-LK Y Y 2.61 R Risk significance increases by one order

6. 1 Charging pump 35-PM-CHGP1B-PS Y Y 1.06 Y

7. 1 Boric Acid 35-PM-BATPIA-PS Y G NA Transfer Pump

8. 1 HPSI pump (B) 33-PM-SI1B-PR Y Y 1.23 Y

9. 1 Condensate 03-PM-CDP1A-PR G G 1.0 G pump (Truncated)

10. 1 MFW pump 05APM-FWP1A-PR G G 1.0 G (Truncated)

11. 1 EDG 10-GE-DG1B-PR R R 4.19 RR Risk significance increases by one order

12. 1 IA compressor 01-CM-SIAC1F-PR G G 1.02 W Risk significance increases by one order BNL #04334 No. Component Out Basic Event Name SDP After Plant CDF RAW Plant CDF color Comments of Service or (from Color for including including Failed Operator Table 1) Internal external external Action Event initiators initiators

13. 1 SG PORV Y G 1.0 G

14. 1 PORV FTO 36-AV-PR2B-FO W W 1.01 W

15. 1 PORV FTC IE-SLO Y Y 1.11 Y

16. 1 Primary SRV 36-SV-33113-FO W W 1.01 W FTO

17. 1 RHR pump (A) 34-PM-RHR1A-PS Y Y 1.18 Y

18. 1 RHR HX (A) 34-HX-RHR1B-LK Y W 1.14 Y Risk significance increases by one order

19. 1 SW pump 02-PM-SW1A2-PR R R 1.93 R

20. 1 AC Bus (Bus 5) 39-BS-BUS5-SG RR R NA (Case Run)

21. 1 DC Bus (BRA- 38-BS-BRA 102-SG RR RR NA 102) (Case Run) 22 1 DC Bus (BRB- 38-B5-BRB 104-SG RR R NA 104) (Case Run)

23. 1 battery (BRA- 38-BY-BRA 101-OP R R 6.96 RR Risk significance 101) increases by one order

24. 1 Battery Charger 38-BC-BRA 108-OP RR W NA (BRA-108) (Case Run)

25. RHR Pump pit 17-FNPMPPITA-PS Y Y 1.06 Y Fan coil Unit A

26. TSC Diesel 10-GE-TSC-DG-PR W Y 1.13 Y

27. 1 MSIV FTC 06-AV-MS1A-FC R Y NA

28. 1 PORV Block - W NA Aug. 14, 2003 valve FTC 29 Piggyback valve - G NA RHR-299 Operator Actions BNL #04334 No. Component Out Basic Event Name SDP After Plant CDF RAW Plant CDF color Comments of Service or (from Color for including including Failed Operator Table 1) Internal external external Action Event initiators initiators

30. Operator fails to 05A-MF2- - -HE Y Y 1.35 Y recover MFW

31. Operator fails to 36-OBF- - - HE Y R 2.74 R initiate FB

32. Fails HPR 36-2TRN-REC-HE R R 1.5 R

33. Operator fails to 35-CH2- - - HE R Y 1.11 Y start a charging pump

34. Fails LPR 34-LR1- - - HE RR W 1.02 W

35. Fails RCSDEP in 06-OC1- - - HE W G 1.01 W Risk significance SLOCA increases by one order

36. Isolation of 06-IS2- - - HE R R 1.45 R ruptured SG

37. RCS cooldown 06-OC3- - - HE R R 1.44 R and depressurization in SGTR

38. Refill RWST in Case Run Y Y 1.09 Y SGTR

39. Emergency Case Run W G NA boration in ATWS Note:

Aug. 14, 2003

1. RR signifies a risk impact between 1E-3 and 1E-3.

ATTACHMENT 1 LIST OF PARTICIPANTS Sonia Burgess NRC/Region III Mike Franovich NRC/NRR Robert Perch NRC/NRR Mike Tschiltz NRC/NRR Pranab Samanta BNL John Poloski INEEL Edward Coen Kewaunee Nuclear Plant/NMC John Helfenberger Kewaunee Nuclear Plant/NMC