Text

IPE Back-End TER Dtd Sept 1994
	ML20078E122
Person / Time
Site:	Point Beach
Issue date:	09/30/1994
From:	Amarasooriya W; SCIENTECH, INC.
To:	; NRC
Shared Package
ML20078D982	List: ML20078E047; ML20078E122; ML20078E150;
References
	CON-NRC-05-91-068-23, CON-NRC-5-91-68-23 SCIE-NRC-224-93, NUDOCS 9501310132
	Download: ML20078E122 (24)
	v • d • e

, . - - --. . ... - -

t

o :

i

1 {

i I

POINT BEACH NUCLEAR PLANT INDIVIDUAL PLANT EXAMINATION l TECHNICAL EVALUATION REPORT I (BACK-END) !

~

t I

L i

i l

)

i i

t 9501310132 940930~

PDR ADOCK 05000266 ;

P PDR l

l l

ENCIDSURE 3 l l

l'

-0 4

.e SCIE-NRC-224-93 ,

I 1

POINT BEACH i INDIVIDUAL PIANT EXAMINATION !

BACK-ENO I TECHNICAL EVALUATION REPORT i

W. H. Amarasooriya l

l l

Prepared for the U.S. Nuclear Regulatory Commission Under Contract NRC-05-91-068-23 September 1994 1

SCIENTECH, Inc.

11821 Parklawn Drive Rockville, Maryland 20852 l

y _

i O

TABLE OF CONTENTS EXECUTIVE

SUMMARY

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii I. INTRODUCDON .................. .... .. ...........1 ,

2. CONTRACTUR REVIEW FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1 Review and Identification of IPE Insights . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1.1 General Review of IPE Back-End Analytical Process . . . . .....2 2.1.1.1 Completeness . ...........................2 l 2.1.1.2 Description, Justification, and Consistency . . . . . . . . . 2 2.1.1.3 Process Used for IPE . . . . . . . . . . . . . . . . . . . . . . . . 2 '

2.1.1.4 Peer Review ofIPE . .... ............ .....2 -

2.1.2 Containment Analysis / Characterization . . . . . . . . . . . . . . . . . . . . . 3 2.1.2.1 Front-end Back-end Dependencies . . . .... . 3

.:.! 2.2 Sequence with Significant Probability . . . . . . . .. .4

.t.l.2.3 Failure Modes and Timing . . . . . . . . . ... .. .4 2.1.2.4 Containment Isolation Failure . . . . . . . . . . . . . . . . . . 5 2.1.2.5 System / Human Response . ..... ........... 6 2.1.2.6 Radionuclide Release Characterization . . . . . . . . . . . . 6 2.1.3 Accident Progression rmd Containment Performance Analysis . . . . 7 2.1.3.1 Severe Accident Progression . . . . . . . . . . . . . . . . . . . 7 2.1.3.2 Dominant Contributors: Consistency with IPE Insights . 7 2.1.3.3 Characterization of Containment Performance ......9 2.1.3.4 Impact on Equipment Behavior . . . . . . . . . . . . . . . . .9 2.1.4 Reducing Probability of Core Damage or Fission Product Release . 10 2.1.4.1 Def'mition of Vulnerability . . . . . . . . . . . . . . . . . . 10 2.1.4.2 Plant Improvements . . . . . . . .. ......... . . 10 2.1.5 Responses to CPI Program Recommendations . . . . . . . . . . . . . . . 10 2.2 IPE Strengths and Weaknesses . . . .. .. ....... ... .. .... 11 2.2.1 IPE Strengths ........... .................. .... . 11 2.2.2 IPE Weaknesses . . . . . ..... ...... ... . . .. . . . 11

3. OVERALL EVALUATION .

................................ . . 12

4. IPE INSIGHTS, IMPROVEMENTS, AND COMMITMENTS ...... .... 13

5. REFERENCES. . . . . . . . ................ ............. .........15 APPENDIX A: IPE EVALUATION AND DATA

SUMMARY

SHEET . . . . . . . . . . . . A-1 Point Beach IPE Back-End Review ii September 1994

s EXECUTIVE

SUMMARY

SCIENTECH, Inc., performed a review of the back-end portion of the Wisconsin Electric Power Corporation's (WEPCO's) Individual Plant Examination (IPE) of the Point Beach Nuclev Plant (PBNP) Units 1 and 2.

Fauske and Associates, Incorporated, led the back-end analysis. WEPCO personnel performed all of the MAAP runs and authored portions of the Level 2 Source Term Notebook. WEPCO also reviewed and approved all other portions of this analysis. A staff member of the Electric Power Research Institute (EPRI) performed independent technical review of the Point Beach Level 2 phenomenological position papers, which were used as bases for accident progression analysis of the PBNP IPE. A team, consisting of two staff members from WEPCO, a staff l member from Nuclear Utilities Services Corporation, and a staff member from Gabor, Kenton,

& Associates, Incorporated, performed independent review of the Level 2 Probabilistic Safety Analysis (PSA). The primary focus of the review was the Source Term Notebook.

Units 1 and 2 began commercial operation in December 1970 and October 1972, respectively.

Each unit was designed to produce a reactor thermal power output of 1518.5 MW and gross electric power output of 523.8 MW. The differences between the two units are minor and '

have no effect on the back-end analysis results. Each system was analyzed as it existed on September 5,1990, the PSA design and data analysis freeze date. The IPE team performed a Level II walkdown of the PBNP Unit I containment on April 17,1991, and used the information gathered to model the PBNP containment in the PBNP MAAP analysis.

From the Level I analysis, the team calculated an overall core damage frequency (CDF) of 1.04E-4 per reactor year excluding internal flooding. The largest CDF initiator was the large ;

loss of coolant accident (LOCA), followed by station blackout and transient without power j scram.

i l

The IPE team performed a Level Il PSA of PBNP using dominant Level I sequences binned j into 17 plant damage states (PDSs) (including bypass and isolation failure sequences). The

{

IPE team calculated containment response and radioactive source terms for these PDSs usmg j

MAAP 3.0B (PWR Version 19) analysis for a 48 hour5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months mission time. The team addressed '

several phenomenological issues using PBNP-specific position papers, and used a best-estimate containment failure fragility curve for the containment response analysis. The PBNP 1 IPE team used containment event trees (CETs) to characterize containment response to core melt sequences, using plant damage states quantified with Level I PSA models updated to include containment fan cooler system. Containment event tree structure is system-oriented, i.e., the containment phenomenology issues are not addressed in the CET structure. The phenomenological issues are described in Phenomenological Issue Papers," the review of which is outside the scope of this report. Also, the quantification of CET is not probabilistic.

The decision points are either 0 or 1. For a given PDS, only one path lead to an endpoint in the CET. In essence, it is the PDS that forces the direction of the path, rather than l containment phenomenological issues.

j l

Point Beach IPE Back-End Review iii September 1994

0

, The IPE team assumed that all core damage sequences continue to vessel breach and did not take any credit for the in-vessel cooling as a result of operator interventions. No credit was ,

taken for operator actions or equipment recoveries following the onset of core damage, except ;

for some station blackout (SBO) sequences. In SBO sequences where, the power was recovered within 6 hours6.944444e-5 days 0.00167 hours 9.920635e-6 weeks 2.283e-6 months , the IPE team assumed that the containment fan coolers and service water were available. The IPE team assumed that containment failure would occur for the core damage sequences where the containment pressure approached the best-estimate failure pressure or where the pressure was approaching that value at 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months .

The PBNP IPE team postulated no early containment failures other than containment isolation !

failures and bypass sequences. Severe accident phenomena including steam explosions, i molten core-concrete interactions, direct containment heating, vessel thrust forces, thermal attacks on containment penetrations, and hydrogen detonation and deflagration were found not to threaten the integrity of the PBNP containment. According to the submittal, these issues were addressed in specific position papers The IPE team performed the MAAP sensitivity '

analysis recommended in EPRI TR-100167.

The largest contributors to the fission product release frequency (FPRF) postulated were transient and SBO core damage sequences, which would cause the containment to overpressurize and fail at 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months . These initiators would contribute to 74 percent of the FPRF. The next largest contributor to FPRF would be steam generator tube rupture (SGTR) sequences, which would prevent the reactor coolant system (RCS) from cooling down and depressurizing enough to stop leakage to the ruptured steam generator.

The IPE submittal does not provide information in response to the Containment Performance Improvement (CPI) P:ogram recommendations. Related information on hydrogen detonation and deflagration is given in a position paper, which is not provided with the submittal.

WEPCO defined severe accident vulnerabilities as:

plant specific design or operation characteristics resulting in dominant core damage or ,

large fission product release frequencies significantly above the NRC's mean targets for all nuclear power plants from SECY-89-102, Implementation of Safety Goal Policy. These generic probability targets are IE-4 per year and IE-6 per year for core damage and large fission product release, respectively. The Point Beach has identified no such dominant sequences of contributors. (Section 1.0, page 10)

The PBNP has a total CDF of 1.04E-4 per year and a large fission product release frequency of 6.35E-6 per year. The IPE submittal notes that "(w)hile 6.35E-6 is significantly greater than the NRC safety goal target of IE-6, it ignores a further conservatism in the PBNP !

analysis." (Section 4.9, page 78 (of Section 4.0))

The PBNP IPE identified several plant improvements, which could reduce the CDF or large ;

release frequencies postulated. Plant improvements that would have an effect on the back-end analysis included a procedural revision to manually align attemate water sources to the suction of the auxiliary feedwater (AFW) pumps, a design modification to connect fire water Point Beach IPE Back-End Review iv September 1994

..a , -n - - - - v- , - --

I

, to condensate storage tanks, and severe accident management guidelines to mitigate the fission product release sequences. These mitigating actions were specified in the PBNP EOPs. The design modification to connect fire water to condensate storage tanks is expected to be made, the connector installed and tested, and the associated procedural revisions and training conducted by September 1994.

The IPE took little credit for Level II recovery actions, including those where operators and i emergency response crews repaired and restored failed accident mitigation equipment; replenished depleted water tanks, compressed air supplies and batteries; and restored AC r electrical power sources. Many of these actions are covered in the PBNP emergency ,

operating procedures (EOPs), which include the Critical Safety Function Status Trees and '

Critical Safety Procedures. These EOPs are expected to help mitigate fission product release for most of the transient- and SBO initiated core damage sequences that resulted in containment late failure (around 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months ) due to overpressure (74 percent of the PBNP FPRF). Apart from these actions, the PBNP does not plan nor has made any commitments to make any plant improvements based on Level II analysir insights.

l l

l I

Point Beach IPE Back-End Review v September 1994 l

t

1. INTRODUCDON I

This technical evaluation report (TER) documents the results of the SCIENTECH review of '

the back-end portion of the PBNP Units 1 and 2 individual plant examination (IPE) submittal

[1]. This technical evaluation report complies with the requirements for reviews of the U.S.

Nuclear Regulatory Commission contractor task order, and adopts the NRC review objectives, ,

which include the following: '

To determine if the IPE submittal provides the level of detail requested in the " Submittal Guidance Document," NUREG-1335 To assess the streagths and the weaknesses of the IPE submittal To complete the IPE Evaluation Data Summary Sheet Section 2 of the TER summarizes our findings and briefly describes the PBNP IPE submittal '

as it pertains to the work requirements outlined in the contractor task order. Each portion of Section 2.1 corresponds to a specific work requirement. Section 2.2 sets out our assessment of the PBNP submittal's strengths and weaknesses. Section 3 presents our evaluation of the PBNP IPE overall, based on our review. Section 4 outlines the insights gained, plant improvements identified, and utility commi.ments made as a result of the IPE. Appendix A contains an evaluation summary sheet that SCIENTECH completed on the PBNP IPE.

I i

l l

t Point Beach IPE Back-End Review 1 September 1994 l

i

l o

~ 2. CONTRACIDR REVIEW FINDINGS e

2.1 Review and Identification of IPE Insights This section is structured in accordance with Task Order Subtask 1. :

1 2.1.1 General Review of IPE Back-End Analytical Process -

2.1.1.1 Completeness Although there is a large body of information povided in the Back-End portion of the PBNP I IPE submit'al, there are portions of the submittal that appear to be incomplete. For example, no containment structural analysis is presented or is there reference to like containments for

which structural analysis has been performed. In most creas, however, the submittal is essentially complete with respect to the level of detail requested in NUREG-1335.

i The IPE submittal appears to meet the NRC sequence selection screening criteria described in Generic Letter 88-20.

i 2.1.1.2 Description, .histification, and Consistency ,

i The IPE methodology used is described clearly. The approach frhed is consistent with the i basic tenets of Generic Letter GL 88-20, Appendix 1. I 2.1.1.3 Pmcess Used forIPE The IPE team performed a " limited-scope" Level II PSA of PBNP using dominant Level I sequences binned into 17 PDSs (including bypass and containment isolation sequences). The i IPE team calculated containment response and radioactive source terms for these PDSs using l

the MAAP 3.0B (PWR Version 19) analysis for a 48-hour mission time. The team addressed

{

several phenomenological issues using PBNP-specific position papers, and a best-estimate containment failure fragility curve for the containment response analysis. Containment event !

trees were used to characterize containment response to core melt sequences, using dant damage states quantified with Level I PSA models updated to include containment cooling systems.

2.1.1.4 Peer Review of IPE Fauske and Associates, led the back-end analysis. WEPCO personnel performed all of the MAAP runs and authored portions of the Level 2 Source Term Notebook. WEPCO also reviewed and approved all other portions of this analysis.

A staff member of the Electric Power Research Institute performed an independent technical review of the Point Beach Level 2 phenomenological position papers, which were used as Point Beach IPE Back-End Review 2 September 1994 1

o.

5 i

bases for accident progression analysis of the PBNP IPE. One of the er ajor comments to come out of the review was the asser % that the technical basis was v eak for concluding i that molten core-concrete interaction el&t) was not a credible mech mism for containment. The IPE team addressen i4 eg nt by performing zn in-depth review of the issue and providing a stronger technical bass $ g. basic position stated i

A 15-member team, including 10 staff members from WEPCO, performed an initial review of the PBNP PSA documentation. His review team made severd hundred commen '

editorial and clarifying nature, but made no major technical comments. Rese comments are documented on Form QP 5 3.1, " Nuclear Power Department Document Review." The submittal does not provide sufficient information to judge what fraction of this initial i

document review effort was related to the back4nd analysis.

t A four-member team, consisting of two staff members from WEPCO, a staff member from Nuclear Utilities Services Corporation, and a staff member from Gabor, Kenton, ;

and Associates, Incorporated, performed an independent review of the Level 2 Safety Ana PSA. The primary focus of the review was the Source Term Notebook. !

2.1.2 Centainment Analysis /Onaractedzation l 2.1.2.1 Front-end Back-end Dependencies The coupling between front-end and back-end analysis is based on PDSs. De definition of PDSs and the binning of accident sequences into PDSs are discussed in Section 3.1.5 of the i submittal (pages 298 through 320 of Section 3.1). He IPE team developed PDS trees to collapse core damage sequences based on their similarity of primary and containment .

conditions. The PDS trees are identical to core damage event trees with the addition of a j final node representing the availability of containment fan coolers. The success criteria for this node is at least one containment fan cooler unit running. Table 3.1.5-1 (pages 304 and 305 of Section 3.1) of the submittal shows the results of binning of accident sequences into 16 PDSs (the PDS with containment isolation failure, AFALI is not included).

The IPE team identified the following parameters to characterize PDSs.

Initiating event type RHR system operability i t

Containment fan coolers Estimated time of core melt The team did not address the state of the containment with respect to isolation hs in the definition of PDSs. They estimated a frequency for containment isolation failures oy multiplying the frequency of the dominant PDS (AFAL with a frequency of 3.7E-5 per by 3.2E-4 which was characterized as a generic containment isolation probability. This yielded a containment isolation frequency 1.2E-8 per year.

Point Beach IPE Back-End Review 3 September 1994 i

v ' ~

r- - - -- _ . - . - --

, Containment bypass events (e.g.,ISLOCA) were not addressed in the definition of PDSs. t The IPE 1eam did not perform source term analysis for ISLOCA events. They analyzed these events as part of SGTR sequences.

Operability of containment spray pumps was not included in the PDS definitions. The IPE team authors argue that the containment fans are adequate to control containment pressure.

They further argue that, because no early containment failure is likely, except for bypass or isolation failure sequences, the source term releases are not affected by the operation of the spray system.

2.1.2.2 Sequence with Significant Probability For the back-end analysis, the PBNP IPE team selected 17 PDSs,14 of which were capable of causing CDFs at levels greater than IE-8, and 3 that could cause CDFs at levels less than IE-8. (These 17 PDSs are listed in the C-Matrix in Appendix A to this report.)

Section 4.7.3, page 46 (of Section 4.0) of the submittal, notes that the IPE team screened Level I results using the criteria suggested in NUREG 1335, which are summarized as follows-All systemic sequences greater than IE-7 per year CDF All sequences within the upper 95 percent of the CDF All systemic sequences in the upper 95 percent of containment failure frequency All containment bypass sequences with frequencies greater than IE-8 per year.

It appears that the PBNP IPE met the sequence selection criteria, as outlined in Appendix 2 to the Generic Letter 88-20.

2.1.2.3 Failure Modes and Timing Section 4.1.1 page 1 (of SNtion 4.0) of the submittal, lists the following important containment data:

Design pressure (psig) 60.0 Design temperature ('F) 286 Inner diameter (ft) 105 Interior height (ft) 150.25 Cylinder shell thickness (ft) 3.5 Dome thickness (ft) 3.0 Internal free volume (ft') 1,070.000 Cavity floor thickness (ft) 8.0 Cavity floor area (ft') 362

)

Point Beach IPE Back-End Review 4 September 1994

e

The IPE team performed a plant-specific structural analysis using the MONTEC code to calculate the internal pressure capacity and the likely failure locations associated with this pressure. Figure 4.4-1, page 17 (of Section 4.0) of the submittal, shows the calculated PBNP containment fragility curve. The rnedian failure pressure is 177 psia. The dominant failure locations are the cylindrical sction of the shell wall in the tendon material and the basemat/shell junction. In performing the source term analysis, members of the IPE team conservatively used the failure location at the cylindrical section. They used the MAAP computer code to calculate the time at which the containment would fail under predicted loads.

The team considered the following causes of containment failure unlikely to occur within the Level 1148-hour mission time: steam explosion, molten core-concrete interaction, direct containment heating, vessel thrust force, thermal attack on containment penetrations, and hydrogen detonation and deflagration. Containment failure mechanisms that the team did consider likely were: containment overpressurization, containment bypass, and failure to isolate containment.

2.1.2.4 Containment isolation Failure The PBNP IPE team concluded isolation would cause the containment to fail under any of the following circumstances (in addition to the condition where all check valves in fluid lines must fail):

A fluid line of mechanical penetration, which is to close manually during power operation, is left unisolated.

A fluid line, with isolation valves that are to close automatically following generation of an isolation signal, but fail to close.

A fluid line, which is a part of a safety system required to remain open following generation of an isolation signal, and is not closed by the operators even if the system is " failed" or if system operation has terminated.

(Section 4.4.2, page 18)

The: PBNP IPE team used containment isolation failure as a top event in the PBNP CET, PDS AFALI represent the sequer es with containment isolation failure. As noted in Section 3.1.5.1, page 301 (of Section 3.1) of the submittal "[t]he frequency of this plant damage state is only 1.2E-8/ year, [ conditional probability of 0.00031](JAL frequency of 3.7E-5 times )

generic containment isolation failure rate of 3.2E-4 = 1.2E-8/ year)." )

l Point Beach IPE Back-End Review 5 September 1994 l

l 1

" f . .. j

.a j

,, . 2.1.2.5 System / Human Response Section 4.2, page 11 (of Section 4.0) of the submittal, states that f'

[d}ue to the conservative assumptions utilized in the human reliability analysis of the l PBNP PSA, namely that almost no credit will be taken for operator actions or equipment recovery following the onset of core damage.

'Ihe IPE team members argue that EOPs are designed to prevent core damage events and are preventive measures by nature. They further state that, if core damage has occurred, by definition, the EOPs were not used properly or they were not effective in mitigating the accidents.

This viewpoint seems simplistic. Human actions often are key contributors to accident progressions. Using the above argument, one can also say that, if core damage has occurred, the safety systems are not effective in mitigating the accident and theiefore no need exists to fix them! The IPE team should investigate system / human responses and phenomenological aspects in a probabilistically integrated manner in order to realistically identify the

- vulnerabilities following an accident.

2.1.2.6 Radionuclide Release Gianceerization Section 4.7, pages 37 through 60 (of Section 4.0) of the submittal, describes how the PBNP radionuclide release was characterized. Table 4.7-3, on pages 52 and 53, defines 19 release categories. However, the IPE team calculated that the PBNP releases would occur only under the conditions described in the following four categories:

A No containment failure within the 48-hour mission time, but eventually failure could occur without accident management action; noble gases and less than 1/10-percent 3 volatiles released G Containment failure prior to vessel failure with noble gases and up to 10-percent volatiles released (containment isolation impaired)

S No containment failure (leakage only, successful maintenance of containment integrity; containment not bypassed; isolation successful) i T Containment bypassed with noble gases and no more than 10-percent volatiles released Conditional probaibility of releases in the above categories are 0.174,0.00031,0.766, and 0.061, respectively, for A, G, S, and T (Table 4.7-5, page 60). Table 4.7-4, on pages 58 and 59 of the submittal, lists MAAP run summary results of the PBNP source term analysis.

7 I

PcN Reach WE Back-End Review 6 September 1994

__w .-.

, Generic Letter 88-20 states that the following should be reported:

any functional sequence that has a core damage frequency greater than lx104 per reactor year and that leads to containment failure which can result in a radioactive release magnitude greater than or equal to BWR-3 or PWR-4 release categories of WASH-1400.

The PBNP IPE appears to have met this reporting requirement.

2.1.3 Accident Progression and Containment Perfonnance Analysis 2.1.3.1 Severe Accident Progression The PBNP IPE team found that the following severe accidents are not likely to threaten the PBNP containment: steam explosions, molten core-concrete interactions, direct containment heating, vessel thrust forces, thermal attacks on containment penetrations, and hydrogen detonation and deflagration. The basis for this finding is given in phenomenological position papers, which are not provided in the submittal. Summaries of these position papers are given in Section 4.4.3, pages 19 through 25 (of Section 4.0) of the submittal. These summaries are not sufficient to judge how well the PBNP understood the threat to the containment from severe accidents.

2.1.3.2 Dominant Contributon: Consistency with IPE Insights Table 1 in this report shows the results of SCIENTECH's comparison of the dominant contributors to the PBNP containment failure with those contributors identified during IPEs performed at the Diablo Canyon, Maine Yankee, Palo Verde, Kewaunee, Zion, and Haddam Neck plants, and with the NUREG-1150 PRA results obtained at Zion and Surry.

The PBNP IPE team found that no early containment fatic-s would result from severe accident paenomena, including steam explosions, direct contamment heating, vessel thrust forces, thermal attack on containment penetrations, and hydrogen detonation and deflagration.

The calculated late containment failure conditional probability, given core damage, was 17.4 percent, which is significantly lower than that for HNP (54 percent). The major reason for this difference is that HNP has a thinner basemat (5 feet thick as compared to 8 feet at PBNP), and a dry cavity during all sequences, which resulted in basemat penetration; PBNP had a flooded cavity during all sequences, which mitigated core-concrete interactions. ;

l l

l i

l l

1 Point Beach IPE Back End Review 7 September 1994

-a ,.

V Table 1. Centainment Failure as a Percentage of Total CDF: PBNP Results Cesupseed with die Results of the Diable Canyon, Maine Yankee, Pale Verde, Kewaunee, and Zion IPEs -

and with the Zion and Sony NUREG-1150 PRA Results Containment Diablo Maine Palo Verde Kewaunee Zion Haddam Zion / Surry/ Point Failure Canyon Yankee IPE IPE' IPE Neck NUREG-. NUREG- Beach IPE' IPE 2 IPE 1150 1150 IPE CDF 8.8E-5 7.4E-5 9.0E-5 6.6E-5 4.0E-6 1.8 E-4 6.2E 4.lE-5 1.04E-4 (per rx year)

Early 4.6 8 10 0 0 0.18 1.5 1 0 Late 66.6 48 14 49 5 54 25 6

. 17.4 Bypass 1.8 2.1 4 8 30 6.5 0.5 12 6.1 t Isolation 7

O' O.023 2 0.5 NA **

0.031 Intact 20 43 72 43 63 39 73 81 76.6 Bypass and isolation combined Included in early failure l

Reflects the IPE results without taking credit for recovery of containment heat removal 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months after event initiation 2

Values do not add to "100" Probability is less than 0.001, conditional on core melt

~

Point Beach IPE Back-End Review 8 September 1994

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ , - . _ _ . _ _ . ~ __ -

, i i

.o ,

t 2,1.3.3 Characterization of Containment Pesfonnance !

. The PBNP IPE team characterized the containment performance using CETs that consisted of j the following top events: '

Top Event 1 - Accident initiator I Top Event 2 - Containment not bypassed Top Event Containment isolation intact

'j' Top Event 4 - High-pressure melt ejection i Top Event 5 - No early containment failure Top Event 6 - RHR pumps j

Top Event 7 - Containment fan cooler units !

De charreterizations of Top Events 1,2,3,6, and 7 were taken from PDSs. Only two top ,

events, " Top Event 4 - High pressure melt ejection" and " Top Event 5 - No early containment ,

failure" addrersed phenomenological events. The submittal notes that occurrence of severe -

i accident phenomena and associated sensitivities do not threaten the containment integrity, nor would they result in early containment failure. Therefore, " Top Event 5 - No early

}

containment failure" has a probability for success of I for CET quantification and best- !

esnmate source term analysis. The IPE team reported that it took Top Event 5 into account ;

primarily for completeness and to indicate that phenomenological uncertainties have been j considered through the phenomenological evaluation summaries." (Section 4.5.1, page 31 (of i Section 4.0)) '

Containment event tree structure is system-oriented, i.e., the containment phenomenology issues are not addressed in the CET structure. The phenomenological issues are described in "Phenomenological Issue Papers," the review of which is outside the scope of this report. j Also, the quantification of CET is not probabilistic. The decision points are either 0 or 1. i For a given PDS, only one path can lead to an endpoint in a CET. In essence, it is the '

definition of the PDS, rather than containment phenomenological issues, that forces the direction of the path.

2.1.3.4 Impact on Equipment Behavior '

The only equipment that is specifically credited following the onset of severe accident conditions is containment fan coolers. The IPE team has reviewed the functionality of fan !

coolers, including the potential for aerosol blockage, damage from corium'for HPME, and flooding, and found that the fan coolers would remain operational in a post core damage ,

environment. The completeness of this assessment is satisfactory. '

-l l

1 l

1 Point Beach IPE Back-End Review 9 September 1994 i b

i

o 2.1.4 Reducing Pmbability of Core Damage or Fission Paduct Release b

2.1.4.1 Definition of Vulnerability ne WEPCO defined severe accident vulnerabilities as:

plant specific design or operation characteristics resulting in dominant core damage or large fission product release frequencies significantly above the NRC's mean targets for i all nuclear power plants from SECY-89-102, Implementation of Safety Goal Policy.

These generic probability targets are IE-4 per year and IE-6 per year for core damage ;

. and large fission product release, respectively. The Point Beach has identified no such dominant sequences of contributors. (Section 1.0, page 10)

The PBNP has a total CDF of 1.04E-4 per year and a large fission product release frequency of 6.35E-6 per year. The IPE submittal notes that although this is "significantly greater than the NRC safety goal target of IE-6, it ignores a furJ er conservatism in the PBNP analysis."

(Section 4.9, page 78 (of Section 4.0))

2.1.4.2 Plant Impmvements As described in Section 6 of the submittal, many design improvements have been . 1

)

implemented at PBNP. Most of these improvements affected the front-end analysis of the ;

plant. Improvements that did affect the back-end analysis were a procedural revision to l

manually align alternate water sources to the suction of the AFW pumps, a design i modification to connect fire water to condensate storage tanks, and severe accident management guidelines to mitigate the fission product release sequences. These mitigating ;

actions are specified in the PBNP EOPs.

2.1.5 Responses to CPI Pmgram Recommendations One CPI program recommendation on PWRs with large, dry containments was that utilities evaluate their containment and equipment vulnerabilities to hydrogen combustion (local and global) as a part of their IPEs and that they identify need for improvements in PWR procedures and equipment.

I Information on hydrogen detonation and deflagration is given in a position paper, which was not provided with the submittal. Section 4.4.3, pages 19 and 20 of Section 4.0, notes that the PBNP IPE team performed plant-specific analysis of a station blackout core damage sequence using worst-case assumptions for hydrogen production. This analysis demonstrated that not enough hydrogen would accumulate to cause deflagration and thus would not challenge the containment integrity.

During a containment walkdown, the IPE team has reviewed the areas for possible localized hydrogen pocketing which could ignite and causing failures of equipment.

Point Beach IPE Back-End Review 10 September 1994

1

, 2.2 - FE Strengths and Weaknesses i 2.2.1 IPE Strengths i

1.

De submittal describes the source term characterization in detail: the initial inventory -

of the source term, fission product transport and removal mechanisms, and MAAP run j summary of source term analysis (timing and fission product distribution).

]

l

2. The WEPCO. staff appears to have been involved significantly in the back-end analysis. l

3. The IPE team performed a MAAP sensitivity analysis that is recommended in -

EPRI TR-100167.

(

4. The IPE submittal is well written and well organized.

2.2.2 IPE Weaknesses

1. The CET split fraction values used in the back-end submittal were extremely l conservative (e g., O or 1). Use of these values might have produced unrealistic results and masked some insights.

2. It is difficult for us to understand how the submittal authors justify the quantitative -i results of the back-end portion of the IPE. The key issues that drive the important l

quantitative results are not addressed in the CETs. (The phenomenological uncertainties l of early containment failure resulting from steam explosions, direct containment heating, j vessel thrust forces, and direct contact of shell with fuel debris are addressed in I Phenomenological Issue Papers.) Although the CETs include a top event, "early ;

containment failure," this node is always a success with a value of 1. 1 1

It is not imperative that all the issues be addressed in the context of the CETs.

However, CETs (in PRAs and other IPEs) do provide a traceable and understandable map of the quantification process and product. Absent this, the relationship between the quantitative results in the CETs and the detcrministic analysis that supports the results must be delineated and understood.

His relationship appears to be absent in the IPE submittal. For example, with regard to i the phenomenological uncertainties mentioned above, no deterministic analysis appears evident in the IPE submittal; only abstracts of "phenomenological evaluation summaries."

l

3. The submittal does not show an understanding of the relationship between PDSs and containment performance.

Point Beach IPE Back-End Review 11 September 1994 1

l i

l

3. i OVERALL EVALUATION ,

l 1

As discussed in Section 2, this IPE submittal contains a large amount of back-end j

information, which contributes to the resolution of severe accident vulnerability issues at PBNP. The questions raised in Appendix B address some areas and issues that do not appear to be addressed completely in the IPE submittal.

The fundamental conclusion stated in this IPE submittal is that the PBNP containment will I not fail from severe accident loadings (pressure, temperature) during the first 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months after a severe accident occurs. (There is an 6-percent probability of a containment bypass and a very small " containment isolation" failure probability.)

We also note the following: l The IPE did not appear to consider early containment failures (steam explosions, direct i containment heating, vessel thrust forces, and direct contact of shell with fuel debris) in CET top events, but did consider them in phenomenological evaluation summaries, which were described only qualitatively in the IPE submittal.

The CET split fraction values used in the back-end submittal were extremely conservative (e.g.,0 or 1). Use of these values might have produced unrealistic results and masked some insights.

Point Beach IPE Back-End Review 12 September 1994 i

3 6

i 4 4. IPE DiSIGHTS, IMPROVEMENTS AND COMMfrMENTS Although the team did not explicitly state the insights gained by performing the IPE, SCIENTECH was able to identify the following within the submittal:

- i No early containment failures were expected to result from severe accident phenomena including steam explosions, direct containment heating, vessel thrust forces, thermal attack on containment penetrations, and hydrogen detonation and deflagration.

l

PBNP is expected to have a flooded cavity at the time of vessel failure during nearly all accident scenarios analyzed for the PSA. This is due to the fan cooler condensate drains, refueling cavity drains, and general area drains in containment discharging to this 'l region. Having a wet cavity with floor area sufficient to limit the depth of any molten l corium ensures that significant steam generation will continue to occur, while molten ;

core-concrete interaction will be limited. Extended steam generation due to maintaining :

a water pool above the molten corium in the cavity, assuming no containment safeguards ~

heat removal mechanisms present and no operator actions to establish a heat removal

' means, is predicted to lead to containment failure on overpressurization approximately l 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months after the accident initiation." (Section 4.6, pages 35 and 36 (of Section 4.0))

)

- l

[T]he PBNP buildings housing components modeled in the Points Beach PSA generally ,

incorporate a large, wide open layout, with few doors, high ceilings, and a large bare concrete surfaces (e.g., primary auxiliary building, turbine hall, and water intake facility .

pump house). Therefore, natural circulation cooling with the large heat sinks available j typically provide adequate cooling of spaces for both habitability and equipment cooling ;

for long periods of time without forced ventilation and cooling. In addition, the ambient outside temperatures at PBNP are very moderate." (Section 6.0, page 2 (of Section 6.0)) >

PBNP containment heat removal can be achieved through the use of any one of the following systems: one containment fan cooler unit, one RHR pump with its associated RHR heat exchanger, or one containment spray pump taking suction from the associated RHR heat exchanger discharge. (Section 4.5.1, page 34 (of Section 4.0))

The largest contributors to FPRF were transient and SBO core damage sequences, which would cause the containment to overpressurize and fail at 48 hours5.555556e-4 days 0.0133 hours 7.936508e-5 weeks 1.8264e-5 months . These initiators contributed to 74 percent of the FPRF. The next largest contributor to the FPRF was SGTR sequences, which would prevent the RCS from cooling down and depressurizing enough to stop leakage to the ruptured steam generator. (See Section 1.3, page 9 (of Section 1.0))

The PBNP IPE team identified several plant improvements, which could reduce the CDF or large release frequencies postulated. Plant improvements that would affect the back-end analysis are a procedural revision to manually align alternate water sources to the suction of the AFW pumps, a design modification to connect fire water to condensate storage tanks, and Point Beach IPE Back-End Review 13 September 1994

e

severe accident management guidelines to mitigate the fission product release sequences.

These mitigating actions are specified in the PBNP EOPs.

The design modification to connect fire water to condensate storage tanks is expected to be ,

made, the connector installed and tested, and the associated procedural revisions and training '

conducted by September 1994. (See Section 6.2, page 4 (of Section 6.0))

f i

t I

t i

Point Beach IPE Back-End Review 14 September 1994

w l

1 a l

5. REFERENCES *

1. Wisconsin Electric Power Company," Point Beach Nuclear Plant Individual Plant Units I and 2 Examination Report," June 1993.

l l

i t

l l

1 1

1 l

1 i

N

.i I

l 1

l l

l Point Beach IPE Back-End Review 15 September 1994

)

m . -

p. I APPENDIX A i IPE EVALUATION AND DATA

SUMMARY

SHEET PWR Back-end Facts Plant Name Point Beach Containment Type PWR, large, dry Unique Containment Features None found Unique Vessel Features <

{

None found '

r t

Number of Plant Damage States 17 Ultimate Containment Failuir Pressure 162 psig i

Additional Radionuclide Transport And Retention Structures None Conditional Probability *Ihat *Ihe Containment is Not Isolated

.00031 l

l 1

i

)

Point Beach IPE Back-End Review A-1 September 1994

APPENDIX A (continued)

IPE EVALUATION AND DATA

SUMMARY

SHEET Important insights, Including Unique Safety Features PBNP containment heat removal can be achieved through the use of any one of the following systems: one containment fan cooler unit, one RHR purnp with its associated RHR heat exchanger, or one containment spray pump taking suction from the associated RHR heat exchanger discharge. (Section 4.5.1, page 34 (of Section 4.0)) '

Implemented Plant Improvements None l

l Point Beach IPE Back-End Review A2 September 1994

1 1

1

j l

J APPENDIX A (continued) l IPE EVALUATION AND DATA

SUMMARY

SHEET l l

C-Manis !

l l

{

PDS Frequency Early Late Intact i Per Year Failure Failure ,

AFAL 3.7E-5 1 '

TAAE 1.8E-5 1 i

TAFE 1.4E-5 1 SFAL 1.2E-5 1 i

TFAE 1.2E-5 i RF-L 4.3E-6 1 TFFE 2.2E-6 1 I i

RA-E 2.0E-6 1 SFFL 1.9E-6 1 I 4

AFAE 7.9E-7 1 !

SAAE 4.lE-7 1 SFAE 3.2E-7 1 SAFE 3.lE-8 1 AFALI 1.2E-8 1 AFFE <lE-8 1 i AFFL <1E B 1 SFFE <!E-8 1 Point Beach IPE Back-End Review A-3 September 1994 i

!

ML20078E122

Contents

Text

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

Navigation menu

ML20078E122

Text

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

Navigation menu

Search