ML20128K117

Forwards Levine & Rept,Summarizing Simplified Risk Assessment Program & Presenting Favorable Conclusions, to Support Request for Relief from Burdensome Property Damage Insurance Requirements Per 10CFR50.54(w)

ML20128K117
Person / Time
Site:	La Crosse File:Dairyland Power Cooperative icon.png
Issue date:	07/02/1985
From:	Shimshak R DAIRYLAND POWER COOPERATIVE
To:	Zwolinski J Office of Nuclear Reactor Regulation
References
NUDOCS 8507100484
Download: ML20128K117 (24)

Text

L.1,L D

COOPERATlVE

• P O BOX 817

• 2615 EAST AVE. SO.

• LA CROSSE. WISCONSIN 54602 0817 (608) 788 4 000 July 2, 1985 Mr. John Zwolinski, Chief Operating Reactor Branch No. 5 Division of Licensing U. S. Nuclear Regulatory Commission Washington, D. C.

20555

Subject:

Dairyland Power Cooperative La Crosse Boiling Water Reactor (LAC 8WR)

Provisional Operating License No. DPR-45 On-Site Property Damage Insurance

Reference:

(1) DPC Letter, Shinshak to Zwolinski dated February 7, 1986

Dear Mr. Zwolinski:

Last Thursday we met at the NBC's Bethesda office with members of your staff, the Accident Evaluation Branch and the Office of State Programs for

'the purpose of discussing in detail a report on post-accident recovery costs.

This report and Addendum I which was reviewed at the meeting, form the technical basis for our request for relief from burdenscos property damage insurance requirements as set forth by 10 CFR 50.54 (w).

Reprints of Addendum 1, in final fom, will be sent to you in a few days for record purposes.

During our meeting, we mentioned that a simplified risk assessment program had been conducted a few years ago by Mr. Saul Levine and Dr. Norman Rasmussen. The results of this limited study had been presented to Mr. Rarold Denton by them at a meeting held at his office on April 28, 1982.

The material prepared by them on our behalf and used at the meeting was never docketed. Therefore, to satisfy this objective and to accommodate Mr. J. Hulman's interest in the report, we are sending twenty (20) copies of Mr. Saul Levine's April 23, 1982 letter to Mr. James Taylor (including view graph copies) which samarised the program plan and presented conclusions favorable to LAC 9WR with respect to various major risks.

Jh AnL ldA.%f 0y gu7'M 5%'

f0

r r

-. b Mr. John Zwolinski, Chief Page 2 July 2, 1985 In the event that further information is needed about this program, please contact Dr. Norman Rasmussen at the Nuclear Engineering Department, Massachusetts Institute of Technology, Cambridge, Massachusetts 02138, phone 617-253-3802.

Very truly yours, DAIRYLAND POWER COOPERATIVE y;. 3. l } u'c re s s'+ d >

Richard E. Shinshak, Manager Special Nuclear Projects RES:daj cc:

J. Taylor /PGL-5075 J. Parkyn w/o report O. Hiestand w/ report E. Tremmel w/ report H. Devine w/ report C. Ross w/ report W. Manion w/ report C. Finnan w/ report J. May w/ report R. Mueller w/ report J. Thie w/ report N. Rasmussen w/ report i

1

/

-tNUS mg "m aaen

~

CG-SL-19-82 Project No.

3369.02 April 23, 1982 Mr. James Taylor Assistant General Manager, Power Dairyland Power Cooperative Box 817 2615 East Avenue, South La Crosse, WI 54601

Dear Jim:

Per our recent discussions, I am enclosing a summary of the program plan that was developed to assess LACBWR accident sequence probabilities.

The summary reflects, in the most positive way possible at this time, the favorable impression that Norm Rasmussen and I have of the LACBWR facility.

I believe, as you suggested, that it will be useful to Harold Denton.

If you have any questions or comments, please do not hesitate to call me.

Sincerely, ice President and Group Executive Consulting Group JC/m Enclosure (h A Han tuton Company

A.

u Summary of " Program Plan to Evaluate Accident Sequence Prob-abilities at the La Crosse Boiling Water Reactor."

1.0 INTRODUCTION

The Nuclear Regulatory Commission (NRC) in the Systematic Evaluation Program (SEP) is currently reviewing eleven early j.

generation nuclear power plants.

The purpose of these SEP reviews is to identify areas where plants may need to be up-graded to meet existing regulatory requirements.

These requirements, which have evolved over the years, were devel-oped essentially for large power

reactors, e.g.,

of the i.

400 MWe to 1300 MWe range.

l For some reactors, particularly those with a low power level, and an associated low radionuclide inventory, the imposition of backfits, developed to meet requirements for large power reactors as a means of achieving low public risk, may be of questionable validity.

This appears to be especially true for the La Crosse Boiling Water Reactor (LACBWR), for which a con-l

. sequence analysis has been performed by Sandia National l

-Laboratories.

That analysis predicts that, even for the most l

severe type of accidents postulated for large power reactors, no early fatalities and only 125 latent cancer fatalities may occur.

Conversely, even though the public consequences of a severe accident at LACBWR have been predicted to be extremely small, it is important to know whether undesireable events such as core melt accidents and failure of containment, are of fairly low probability.

To investigate these questions, Dairyland Power Cooperative invited S. Levine, Vice President and Group Executive, Consulting

Group, NU9 Corporation, and N. C. Rasmussen, Professor of Nuclear Engineering, 1

Massachusetts Institute of Technology to visit and review the LACBWR facility.

This meeting occurred on March 22, 1982 at which time Dairyland personnel conducted a tour of LACBWR and cnswered questions posed by Messrs. Levine and Rasmussen.

Based on this meeting and additional interaction between NUS Cnd Dairyland staff, Messrs. Levine and Rasmussen felt that the probability of a core melt with a radionuclide release to the environment, resulting only from internal plant failures at LACBWR, is likely to be quite low.

This belief is based on the extensive system redundancies in the LACEWR design and the relatively modest amount of energy generated by the core, both of which allow a great deal of flexibility in mitigating the effects of accidents.

The development of quantitative estimates of core melt and containment failure probabilities will, however, require some further analysis.

Because of the extremely low consequences that could result fr.om even the most serious potential acci-dent at LACBWR, the level of detailed technical analysis required _in probabilistic risk assessments performed for lar-ger reactors would not be appropriate for LACBWR; rather a cimpler methodology, relying more on judgment and experience, would be more appropriate.

If this analysis proceeds, the ovent trees that will be constructed will illustrate the strength of safety system designs in the LACBWR plant.

2.0 TECHNICAL CONSIDERATIONS a

Although the integrated plant response to a severe accident cannot be determined without a more complete analysis, a pre-liminary review of the plant has revealed that significant cystem redundancy exists to perform various post-accident aitigating functions.

The ability to perform these functions is further enhanced because the very low power level of the

~

2 l

1

r 4

core and the correspondingly small decay heat allows both time and flexibility to use the many alternate methods provided in the plant to mitigate accident effects.

Post-accident func-tions and the avaia.aSle means of performing them have been catalogued in Tables 1 and 2 for various initiating events, based on a brief review of the LACBWR facility.

A number of unusual characteristics of the LACBWR design which lead to increased flexibility of plant response in some situations can be seen in Tables 1 and 2.

For example, for LOCAs, (i) the two train high pressure core spray system has adequate flow capacity to provide core cooling over the entire range of pipe break sizes, rather than just for smaller breaks as in larger reactors, (ii) the alternate core spray system, a low pressure ECC system, has redundant trains, each of whose pumps is powered by a dedicated diesel; this system, combined with the manual depressurization system, can also provide core cooling over the entire pipe break spectrum, (iii) the low pressure core spray system is a gravity fed system supplied by water from the overhead storage tank; this system, when combined with the manual depres'surization system, can provide core cooling, for some. period, over the entire break spectrum and (iv) portable emergency service water supply system pumps can also be used with the manual depressurization system to pro-vide core cooling over the entire break spectrum.

Thus, it can be seen that significantly more system capabil'ity in pro-viding emergency core cooling exists at LACBWR than at most other reactors.

Most modes of containment failure at LACBWR are expected to be less likely than at large power reactors because of the fol-lowing considerations (1) the ratio (P V /P), which is a CD p

1 Where PCD = Containment design pressure, Vy = containment free volume and P = reactor power level.

3 wueusrcririfor:stBIM

?

i measure of the energy absorbtive capacity of the containment, is comparable to or greater than even large dry PWR contain-ments, (ii) the ratio of containment basemat thickness to core power is much greater for LACBWR than large power reactors, (iii) the amount of energy potentially released to the con-tainment after an accident is so small that heat transfer methods like radiation from the steel shell or convection through the component cooling water piping can transfer decay heat to the environment and (iv) the containment is designed to be flooded to the level of the core mid-plane.

Thus, it is expected that containment failures due to basemat melt-through or overpressure due to non-consensables generation or hydrogen burn are unlikely.

Recent work in the area of vessel steam explosion indicates that such a failure mode would also be unlikely.

However, because the containment is ventilated dur-ing normal operation, it is expected that the possibility of a containment that cannot be isolated at the time of an accident would be more likely than for other reactors; even so, this failure mode would _ probably not be greater than 10-2 per demand.

'The preceding discussion indicates the basis for the feeling that severe accidents due to internal plant failures are rela-tively low likelihood events.

The question is then raised about whether common mode failures due to earthquakes and other " external" events contribute significantly to the like-lihood of severe accidents.

However, the potential for cer-tain external events to provide a core melt probability of greater than about 10-4 per reactor year can be immediately dismissed, principally on the basis of the frequencies of these events.

Their contribution to core melt would be even less when the potential for plant systems to mitigate the 4

mu=uara=r:raE6d@N

9 event is considered.

The external events which have been con-sidered and concluded to be unimportant contributors to a high core melt probability at Lacrosse are:

t (1)

External floods (ii)

Tornadoes (iii)

Aircraft related hazards (iv)

Truck transportation related hazards It is also likely that barge and rail transportation hazards i

are negligible contributors to a high core melt probability.

i seismic occurrences and internal fires require further con-j sideration to determine their potential contribution to the core melt probability.

I 3.0 TECHNICAL APPROACH TO A SIMPLIFIED PRA s

i Even though.the public consequences associated with potential accidents at LACBWR are predicted to be small, it is.important I

to know the likelihood of severe events such as core melt and

).

containment failure, given core melt.

More exactly, for

(

LACBWR, it is important only to ascertain whether the prob-abilities of these severe events are likely to be fairly l

small.

To acquire this information, a

simplifled i

probabilistic analysis could be conducted.

This analysir

{

could be done on a limited basis only to demonstrate that the core melt probability is not likely to exceed some value and l

to make an evaluation of the likely containment failure modes and probabilities.

Levine and Rasmussen suggest that the demonstration of a core melt probability in the range of 10~3 l

to 10~4 per reactor year. p uid be sppropriate for LACBWR. The l

simplified analysis would use plant-specific event trees to l

identify potential core melt accident sequences. To determine the containment failure modes associated with these core melt 5

NUS CORPORATION

sequences, an event tree modeling containment failure would also be developed.

The failure probabilities of accident-mitigating systems would not be determined by fault trees, but rather'by a combination of simpler techniques:

expert judg-ment, the construction of Boolean expressions which model system failure modes or.the use of plant operating experience data.

The simplified approach suggested was developed from the insights and techniques used in risk assessments previously performed by the NRC and others; specifically, the Reactor Safety Study (WASH-1400), the Reactor Safety Gtudy Methodology Applications Program (RSSMAP) and the Integrated Reliability Evaluation Program (IREP).

Expert judgment would be used by this approach both as the primary means for performing various tasks or as verification of the results of analyses.

The expert judgment that will be used is that of recognized authorities in the risk assessment field and other areas as needed for the study.

The expert judgment used for the proba-bilistic analysis will be provided by S. Levine and N. Rasmussen.

It is also believed that expert judgment would be particularly important to evaluate the effects of potential seismic events on plant components.

This judgment will be provided by Dr. Robert Kennedy, President, Structural Mechan-ics Associates and Dr. Spencer H. Bush, Senior Staff Consul-tant, Battelle Northwest Laboratories.

4.0 CONCLUSION

The judgment of Levine and Rasmussen is that the probability of a core melt and containment failure at LACBWR is likely to be relatively low.

These probabilities cannot be quantified without further analysis.

However, if it is desired only to ascertain whether the probability of a core melt is not greater than some value, say 10~3 to 10-4 per reactor year, 6

NUS CORPORATON

then a simplified probabilistic analysis can be used.

Con-tainment f ailure can also be assessed in a similar manner.

Such a simplified probabilistic analysis would use insights and techniques from previously documented risk assessments; it would also rely on expert judgment in the area of risk and reliability and in other technical areas, as indicated.

As stated earlier, this less rigorous approach is believed to be appropriate for the LACBWR plant in view of the fact that there would be negligible consequences if a severe accident were to occur.

G 7

TABLE 1 PRELIMINARY SUCCESS CRITERIA POR SYSTI!MS PERFORMING POST-LOCA FUNCTIONS Emergency Containment Emergency Core Containment Post Accident IACA Core OP Protection Cooling (Long OP Protection Radioactivity SIZE Suberiticality Cooling (ECI)

(4 126 Hours)

Term)

( > 126 hours0.00146 days <br />0.035 hours <br />2.083333e-4 weeks <br />4.7943e-5 months <br />)

Removal

• STEAM 1 of 2 HPCS Trains 1 of 2 HPCS and CSS and OHST Makeup BREAK RPS or Containment OBST Makeup or, CSS and OBST r

or 1 of 2 ACS trains Design o_r Component Cooling Makeup

^

BIS or 1 of 2 ACS trains Water System LPCS and MDS or or og MDS and LPCS and External Water MDS and 3 of 4 ESWSS OHST Makeup on Containment E

3 of 4 ESWSS SMALL 1 of 2 HPCS Trains 1 of 2 HPCS and CSS and OBST Makeup LIQUID RPS o_r_

Containment OHST Makeup ol CSS and OHST BREAK or MDS and 1 of 2 ACS Design or Component Cooling Makeup r

BIS Trains 1 of 2 ACS trains Water System or or or SC and 1 of 2 ACS LPCS and MDS and External Water Trains OHST Makeup on Containment o

or MDS and LPCS 3 of 4 ESNSS E

MDS and 3 of 4 ESWSS LARGE 1 of 2 HPCS Trains 1 of 2 HPCS and CSS and OHST Makeup LIQUID RPS ol Contaitynent OHST Makeup or CSS and OHST BREAK or LPCS Design or Component Cooling Makeup BIS of 1 of 2 ACS trains Water System 1 of 2 ACS trains or or of LPCS and OBST External Water 3 of 4 ESWSS Makeup on Contairunent or j

3 of 4 ESWSS 1

1 K2ys ACS:

Alternate Core Spray System LPCS: Low Pressure Core Spray System l

BIS:

Boron Injection System (uses HPCS pumps)

MDS:

Manual Depressurization System I

CSS:

Contairunent Spray System (manual)

OHST: Overhead Storage Tank I

ESWSS: Emergency Severe Water Supply System RPS:

Reactor Protection System i

HPCS:

High Pressure Core Spray System (harsh environment considered in design)

Operation of the CSS is not necessary to meet 10CFR100 dose requirements.

TABLE 2 PRELIMINARY SUCCESS CRITERIA POR SYSTEMS PERFORMING POST TRANSIENT FUNCTIONS RCS Overpressure Emergency Transient Suberiticality Protection RCS Integrity Core Cooling (Short)

Residual Heat Removal General RPS PCS Closure of all PCS PCS (T )

or or Open Relief Valves or or 3

Recirculation SC and/or Safety SC and SC Makeup SC and SC Makeup Pusp Trip and Relief Valves or or Boron Injection pr 1 of 2 HPCS trains Decay Heat Removal MDS-p1 System and MDS and 1 of 2 ACS component cooling trains water systems 01 3 of 4 ESWSS trains Loss of Power RPS SC and/or Safety /

Closure of all SC and SC Makeup SC and SC Makeup conversion or Relief Valves Open Relief Valves or or Sy: stem Recirculation og 1 of 2 HPCS trains Decay Heat Removal (T )

Pump Trip and MDS 2

g System and Boron Injection MDS and 1 of 2 ACS component cooling trains water systems E

3 of 4 ESNSS trains Lors of RPS SC and/or Safety /

Closure of all SC and SC Makeup SC and SC Makeup Offsite or Relief Valves Open Relief Valves or or Power Recirculation or 1 of 2 HPCS trains Decay Heat Removal (T )

Pump Trip and MDS and Emergency System, component 3

Boron Injection Diesel Generator cooling water systems

o_r, and Emergency r

MDS and 1 of 2 ACS Diesel Generator

[

trains gj or n

3 of 4 ESNSS trains O

$K;ys ACS:

Alternate Core Spray System O

HPCs High Pressure Core Spray System j

MMB:

Manual Depressurization System d

PCS:

Power Conversion System (Main Steam and Peedwater)

O SCs Shutdown Condenser Z

h BACKGROUND o

BECAUSE OF ITS LOW POWER, THE CONSEQUENCES OF A SEVERE ACCIDENT AT LACBUR ARE PREDICTED TO BE EXCEEDINGLY SMALL o

HOWEVER, IT IS IMPORTANT-TO KNOW WHETHER THE LIKELIHOOD OF THE UNDESIREABLE EVENTS (CORE HELT AND CONTAINMENT FAILURE) ARE OF FAIRLY LOW PROBABILITY o

BECAUSE OF THE LOW POWER AND THE EXTENSIVE SYSTEM REDUN-DANCIES IN LACBWR, IT APPEARS THAT THE LIKELIHOOD OF THESE EVENTS (DUE TO INTERNAL PLANT FAILURES) WOULD BE QUITE LOW 4

J i

!' S1

EMERGENCY CORE COOLING MORE ECCS CAPABILITY THAN MOST LARGE PLANTS o

FOUR SYSTEMS CAN COOL CORE OVER ENTIRE SPECTRUM 0F PIPE BREAK SIZES:

o 1 0F 2 TRAINS OF HPCS l

0 1 0F 2 TRAINS OF LOW PRESSURE ACS AND MDS o

GRAVITY FED LPCS AND MDS o

3 0F 4 PORTABLE ESWSS AND MDS o

fiANUAL DEPRESSURIZATION NOT REQUIRED FOR LARGE BREAKS AND CERTAIN SMALL BREAKS o

SOME COMMON PIPING T0 CONSIDER l

f\\.

1

~

my.

v.

.s

, %.7,

~,

• (.. i l

.f l.

,, j.y s

\\

's

,A

.CONTAINRENT CHARACTERISTICS t

e s (;, -

o LOW POWER LEVEL I!AKjS CCtf; TAI:4 MENT FAILURE UNLIKELY (SE, BCM)

't o_

EllERGY ABSORPTIVE CAPABILITY (P P) COMPARABLE CD F

TO LARGE-DRY %!R CONTAINNENTS f,f, d{..

, s, f

DECAY hey S0 LOW THAT IT CAN BE REMOVED THROUGH 0

tI STEEL /SHELL OR., COMPONENT [ COOLING WATER SYSTEM x

y 9

t

.ig o [e' CONTAINf1ENT BASEMAT THICKNESS TO CORE POWER

.x GR$ATER FOR LACBWR THAN FOR LARGE REACTORS r

4 a,

y,

,o CONTAINMENT CAN BE FLOODED TO CORE MIDPLANE TO PROVIDE i

s LARGEHEAT4SIflK

,)

( - ( "I x

~

of < GRAVIT8'FEI'j C0'NTlyNMENT SPRAY, SYSTEM AVAILABLE FOR POTENTIALsPRESSURE REDUCTION AMO RADI0 ACTIVITY REf10 VAL l'

t o

-SINCE: CONTAINMENT IS VENTILATED ItANORMAL OPERATION, l

4 CLOSURE SYSTEM PR0f'ADILITY DE FAILURE'HAS TO BE

~.e,.

3' EXAMINED Sh

. s:

n, s

-.s

.\\, r '

• f

?

a

)

4, 93 7,

"4 5

,s i

i 5.

$g 4

5 f

a '

,i g

~

,1 l'

s

(

l(

)

~

• j 8

4 k

l },

T, S3 M:x.

4 s

i x

e m

REACTOR SHUTDOWN o

WELL DESIG!1ED SCRAM SYSTEM o

MAtlUAL BORON INJECTION SYSTEM CAN QUICKLY SHUTD01lN REACTOR o

RECIRCULATION PUMP TRIP LIMITS POWER S4

EXTERNAL EVENTS o

THESE EVENTS PMY CAUSE CONN 0N MODE FAILURES OF SYSTEMS USING FREQUENCY OF 10-4 PER RY SCREENING ALLOWS MANY o

EXTERNAL EVENTS TO BE ELIMINATED FROM FURTHER CONSIDER-ATION:

EXTERNAL FLOODS, TORNAD0ES, AIRCRAFT CRASHES, TRUCK ACCIDENTS, AND PROBABLY BARGE AND RAIL ACCIDENTS o

SEISMIC EVENTS AND INTERNAL FIRES ARE fl0RE DIFFICULT TO ESTIMATE AND WOULD REQUIRE FURTHER ANALYSIS S5

i

SUMMARY

o FROM A BRIEF REVIEW 0F THE LACBWR PLAf!T, IT IS OPINION OF 0F LEVINE AND RASMUSSEN:

o THAT THE CORE MELT PROBABILITY AT LACBWR, DUE TO INTERNAL INITIATIflG EVENTS, IS NOT LIKELY TO BE GREATER THAN 10-3 PER RY o

THAT MANY EXTERNAL EVENTS ARE LIKELY TO CONTRIBUTE A FREGUENCY OF LESS THAN 10-4 PER RY TO THE CORE MELT PROBABILITY.

MORE ANALYSES WOULD BE NEEDED TO CONSIDER THE EFFECTS OF EARTHCUAKES AND FIRES o

THAT MANY FAILURE MODES OF THE LACBWR CONTAINMENT ARE LESS LIKELY THAN FOR OTHER REACTORS o

IF ADDITIONAL BASES FOR THESE C0flCLUSIONS IS NECESSARY, A MORE THOROUGH ANALYSIS CAN BE CONDUCTED BY A SIMPLIFIED PROBABILISTIC ANALYSIS

. m J

1 APPROACH TO A SIMPLIFIED PROBABILISTIC ANALYSIS o

IF NRC FEELS IT TO BE NECESSARY, A SIMPLIFIED PROBABILISTIC ANALYSIS CAN BE PERFORMED o

IT WOULD PRODUCE A ROUGH ESTIMATE OF CORE ftELT AND.

CONTAINMENT FAILURE PROBABILITIES; LESS PRECISE APPROACH THAii TYPICAL PROBABILITY ANALYSES WOULD BE ADEQUATE BECAUSE OF LOW CONSEQUENCES o

IT WOULD BE CONDUCTED ONLY TO THE EXTENT NECESSARY TO ASCERTAIN WHETHER THESE PROBABILITIES ARE LIKELY

. TO BE FAIRLY LOW o

CORE MELT PROBABILITY IN THE RANGE 10-3 TO 10-4 PER REACTOR YEAR IS SUGGESTED AS AN ACCEPTABLE RAllGE FOR LACBWR 9

9 i

4 k

S7

Y MAJOR CHARACTERISTICS OF ANALYSIS o

INTERNAL AND EXTERNAL INITIATING EVENTS WOULD BE CONSIDERED o

f1ETHODOLOGY ADAPTED FROM PREVIOUSLY CONDUCTED RISK ASSESSMENTS I.E. WASH-1400, RSSMAP, IREP o

EXPERT JUDGMENT WOULD BE USED EXTENSIVELY 4

8

..-A

INTERNAL INITIATING EVENTS o

PLANT SPECIFIC EVENT TREES WOULD BE CONSTRUCTED TO COVER SPECTRUM OF LOCAS AND TRANSIENTS o

SYSTEM FAULT TREES NOT USED; SIMPLER TECHNIQUES, INCLUDING EXPERT JUDGMENT, WOULD BE USED o

GENERIC DATA, MODIFIED BY LER AND LACBWR EXPERIENCES AND EXPERT JUDGMENT, WOULD BE USED o

UNCERTAINTIES WOULD NOT BE ESTIMATED ACCIDENT SEQUENCES WITH PROBABILITIES LESS THAN 10-o SCREENED FROM FURTHER ANALYSIS

'S9

CONTAINMENT FAILURE o

SIMPLE ANALYSES AND EXPERT JUDGMENT WOULD BE USED TO IDENTIFY LIKELY CONTAINMENT FAILURE MODES o

PROBABILITIES OF FAILURE MODES WOULD BE ESTIMATED S10

SEISMIC OCCURRENCE o

SIGNIFICANT DETERMINISTIC ANALYSIS PERFORf;ED ON SYSTEM /

STRUCTURES STRESS RESPONSE o

SITE SPECIFIC SEISMICITY CURVE AVAILABLE

,o COMPONENT FRAGILITIES WOULD BE DEVELOPED FROM EXISTING ANALYSES AND EXPERT JUDGMENT o

C0f1BINE SEISMICITY AND FRAGILITY INFORMATION FOR " MINIMUM PATH SET" TO B0UND SEISf11C CONTRIBUTION TO CORE MELT PRO-BABILITY A

S11

o 4

FIRES o

IDENTIFY AREAS WHERE POTENTIAL COMMON MODE FAILURES COULD OCCUR o

USE GENERIC DATA FOR FIRE FREQUENCIES Ill AREAS OF CONCERN o

IDENTIFY SYSTEMS AFFECTED BY FIRE o

MAKE CONSERVATIVE ASSUMPTIONS REGARDING SYSTEf1 DAMAGE TO DETERMINE IF SPECIFIC SEQUENCES ARE SIGNIFICANT CONTRIBUTORS TO CORE MELT PROBABILITY

. S12

E, RESULTS o

POINT ESTIMATES OF CORE MELT AND CONTAINMENT FAILURE PROBABILITIES o

SIGNIFICANT ACCIDENT SEQUENCES AND CONTAINfiENT FAILURE MODES DEFINED

S13