ML20034F273
| ML20034F273 | |
| Person / Time | |
|---|---|
| Site: | 05200002 |
| Issue date: | 02/24/1993 |
| From: | Mike Franovich Office of Nuclear Reactor Regulation |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 9303020531 | |
| Download: ML20034F273 (27) | |
Text
.
/
GkW g.ek
/'pm Rf C,%o 4
UNITED STATES E"'
I NUCLEAR REGULATORY COMMISSION o
f
,h WASHINGTON, D. C. 20555 t
c
'% %, /
February 24, 1993 Docket No.52-002 APPLICANT: ABB-Combustion Engineering, Inc. (ABB-CE)
PROJECT:
CE System 80+
SUBJECT:
PUBLIC MEETING OF JANUARY 4 AND 5, 1993, TO DISCUSS THE REVIEW STATUS OF THE CE SYSTEM 80+ DESIGN (PROBABILISTIC RISK ASSES On January 4 and 5, 1993, a public meeting was held at the ABB-CE facilities in Windsor, Connecticut, between representatives of ABB-CE and'the U.S.
Nuclear Regulatory Commission (NRC). provides a list of attendees. Enclosure 2 is the material presented by ABB-CE.
The staff was represented by members from the NRC's Probabilistic Safety Assessment Branch (SPSB), the Containment Systems 5 Severe Accident Branch (SCSB), and the Civil Engineering & Geosciences Branch (ECGB) as well as from the Brookhaven National Laboratory (BNL). The purpose of the meeting was to (1) discuss draft safety evaluation report (DSER) open items for the Sys-tem 80+ probabilistic risk assessment (PRA), and (2) discuss new and updated PRA-related analyses and expected completion dates.
ABB-CE opened the meeting with a status of the project.
ABB-CE indicated that all analyses requested in the DSER, except for the seismic margins analysis, will be completed by March 19, 1993.
The applicant estimates that it will take approximately 6 months to complete the PRA-based seismic margins analy-sis.
Key issues that were discussed from the System 80+ DSER and agreements that were made are identified as follows:
LEVEL 1 PRA (1)
PRA-insichts and PRA-based input to insoections. tests. analyses. and acceptance criteria flTAAC) and the reliability assurance procram (RAP):
NRC and ABB-CE staff agreed that a section is to be added in the-PRA submittal which will present in an integrated and comprehensive manner.
the results of the PRA as well as those of the uncertainty, importance and sensitivity analyses. ABB-CE staff also indicated that their PRA team will prepare a PRA-based input to ITAAC-by February 15, 1993.
(2)
Modelina of reactor coolant pump (RCP) seal LOCAs: ABB-CE did not model RCP seal LOCAs in the PRA. ABB-CE staff requested that this open item fi be closed independently of other related (deterministic) open items.
The staff disagreed and expressed the opinion that the PRA should O
010C08 9303020531 930224
.f,LW PDR-ADOCK 05200002-1
-A PDR-
V 4 February 24, 1993 reflect models and analyses reported in other parts of the CESSAR-DC. It.
was agreed that ABB-CE will evaluate the core damage frequency (CDF) when RCP seal LOCAs are taken into account, using defendable and state-of-the-art models. This estimate will be viewed either as the result of a sensitivity study or as the base case estimate of the CDF for the design depending on the resolution of the RCP seal failure' issue.
In addition, the effect of the change on CDF, due to the potential RCP-seal LOCAs, on the containment analysis, and on risk will be investigated.
(3)
Modelina of the " failure to open" reliability of the Safety Deoressuri-zation System (SDS): This DSER open item concerns the use of simple demand probabilities in estimating the unavailability of the'SDS (bleed valves). However, the staff believes that assumptions on testing and maintenance activities and intervals significantly affect the demand failure rates for motor-operated valves (MOVs).
For this reason, the staff requested the use of either time-dependent reliability techniques (e.g., hourly failure rates) in determining this probability or provide justification for the assumed probability in the PRA.
It was agreed that this issue will be treated similarly to the RCP seal LOCA issue.
If the rate for failure of M0V's to open is found in the literature (BNL suggested a recent study) then this failure rate will be used as a base case. Otherwise, a conservative failure rate will be assumed and this~
issue will be treated in a sensitivity study. Also, the effect of the.
change on CDF, on the containment analysis, and on risk will be investi-gated and subsequent results provided as inputs to ITAAC and RAP.
(4)
Additional analyses in support of assumptions (e.a.. assumptions associated with success criteria. ctc.): ABB-CE agreed to provide all analyses requested in the DSER open items.
In addition, ABB-CE agreed to submit analyses which prove that multiple steam generator tube ruptures are bounded by the accident sequences modeled in the PRA.
(5)
PRA-based seismic marains methodoloav and acolication: ABB-CE staff presented a methodology for estimating seismic margins as well as approximate sequence and plant fragility curves. ABB-CE will submit more details about the methodology by January 15, 1993. An issue that was raised during the meeting was_the proposed assumption of a rock-type site in estimating component and structure fragilities. ~This assumption introduces a spectral shape that could be inadequate for plants 'with a soil foundation (rock motion versus soil motion). Overall, the method-ology appeared to be in agreement with the staff's_ guidelines. Design-r specific fragilities will be developed for important components and
~
structures.
(6)
Fire and Flood Risk Assessments: The applicant provided an overview of the recently completed risk assessments for internal fires and internal floods.
For the internal fire analysis, a qualitative assessment was first performed, using the FIVE methodology, to demonstrate that the maximum credible fire would be the loss of one division of ell safety-related equipment. The results of the qualitative assessment were used F
~-
. February 24, 1993 L
next to perform a bounding-type quantitative assessment of risk from internal fires. A similar methodology was usedlfor estimating the risk-contribution from internal floods. Although the staff did not yet.
review in detail the fire and flood analyses, ABB-CE appeared to follow-the staff guidelines. The staff requested and ABB-CE agreed to use these analyses to draw insights about the design end to identify items that should be included in the PRA-based input tt ITAAC and RAP as well as for identifying COL items.
LEVEL 2 PRA (7)
ABB-CE's containment fragility curve is now in_ question as a result of two concerns discussed during the meeting. The concerns involve (a).the failure to account for containment penetratians which introduce 3-dimensional effects and thermal stresses _in ABB-CE's ASME code Service Level C estimates, and (b) the failure to accurately reflect the containment shell stress-strain relationship in estimating ultimate pressure capacity..The fragility curve is expected to'significantly change as'a result of these concerns, and may cause ABB-CE to revise their estimates of containment pressure loads.
Issues regarding the structural capacity of the reactor cavity were also identified. and will require further work by ABB-CE. The Level 2 analysis cannot be.com-pleted until these concerns are resolved.
(8)
ABB-CE has completed draft portions of the revised Level 2 PRA in December of 1992. This consisted principally of the _ updated containment event trees, the supporting logic models (SLMs) for each containment challenge, and preliminary / partial information regarding _ probability.
values to be used in quantifying the SLMs. However, ABB-CE has not-yet completed or requantified the Level 2 analysis.
(9)
The staff and ABB-CE teviewed the list of DSER open. items, and ABB-CE.
A described how these issues are being addressed in the updated PRA. With the exceptions noted below, ABB-CE's approach on most items appeared reasonable. However, in many cases ABB-CE _ has not completed or submit-ted the results of their reassessment, and-the staff had not reviewed the details for cases that'have been submitted.
(a) ABB-CE committed to modify ' he probabilities forL non-recovery of t
containment heat removal reported in a DSER response submittal ~to _
reflect the use of a dedicated high pressure /high capacity mobile pump to supply the containment spray header, a zero probability of recovering components inside containment, and the impact of earlier __
times to containment failure on'the recoverability of'offsite power.
(b) ABB-CE committed to assess the impact of diffusion flames insidei containment on containment shell and penetration temperature fail-ure.
.?
e
,E, L
i February 24, 1993 (c) In response to a staff concern that melt-through of the embedded-containment shell within the basemat be considered containment failure', ABB-CE agreed to do additional work to study this failure-mechanism. As part of this work, ABB-CE will include in the model-a L
failure mechanism associated with wet cavity conditions (ablation of liner with corium in the space between the-liner and the concrete and possible release from there) and will' modify the supporting logic models to identify the frequen'cy of shell melt-through while retaining.the traditional PRA definition of basemat failure as being melt-through of the entire concrete basemat.
(10)
In a breakout meeting regarding ABB-CE's use of MAAP, ABB-CE committed to provide the staff with (a) a matrix showing which of the EPRI-recommended sensitivity analyses they intend to perform, and the -
parameter values to be used, and (b) a complete set of MAAP results (plots) for one sequence. The staff agreed to review this information and to inform ABB-CE if additional information will be needed.
l STRUCTURAL ISSUES AFFECTING LEVEL 2 PRA (11) ' ASME Service level C Estimates: ABB-CE previously estimated containment pressure for Service Level C to be 145 psig at 290 *F and 135'psig at 450 *F.
Based on discussions, the staff and ABB-CE agreed that these-values would be reduced by _about 20 psi' (126 psig at 290 *F. and 113 'psig at 450 *F) if thermal stresses and 3-dimensional effects were included.
ABB-CE committed to reassess containment pressure at Servicellevel~C to account-for these effects and revise System 80+ Service Level C estimate accordingly.
(12) Ultimate Pressure Caoacity and Containment Fraaility Curve:
Based on meeting discussions, it appeared that ABB-CE did not accurately charac-terize the containment shell material stress-strain relationship in the input to their ultimate pressure capacity calculation. Specifically, at i
strain levels only slightly greater than those associated with Service..
level C (between strain levels of 0.2 percent and 0.6 percent) the containment radial displacement would increase by about 6 inches with only a slight increase in containment pressure (the stress-strain curve is essentially flat in this region). These displacements ar.e, expected to challenge containment integrity.well before failure stresses in the membrane /shell are reached.
If these~ concerns are borne out, the containment fragility curve will need to be modified to' reflect a low.
probability of containment failure near the Service. Level C pressure, but a significant probability of containment failure at pressures only slightly higher.. ABB-CE committed to (a) reassess their ultimate pressure capacity analysis, (b) further evaluate.the potential for containment failure due-to shell displacement, and (c) modify the containment fragility curve to reflect the' above effects. ABB-CE also -
indicated that if containment failure pressure is substantially reduced, i
1
e
.e February 24, 1993 ABB-CE may elect to remove certain conservatisms in their. estimates of containment pressure loads for severe accident phenomena such as direct containment heating (DCH).
(13)
Reactor Cavity Pressure Capacity: The reactor cavity design for-System 80+ is based on a design for the Duke Power Company's inactive Cherokee plant. ABB-CE believes that the existing design pressure for the cavity would provide sufficient margin to withstand loads associated with ex-vessel steam explosions; however, rebar. placement details and finite element analyses for the cavity were not available. ABB-CE agreed to develop an estimate of the ultimate pressure capacity.for the reactor cavity and provide additional design / analyses details sufficient for the staff to complete its review of this area. The NRC structural staff (ECGB) and ABB-CE agreed to dialog further on this matter. The treatment of fuel coolant interaction (FCI) in the PRA cannot proceed until this assessment is completed.
During the meeting, the NRC's structural staff also met with ABB-CE represen-tatives to discuss structural ' areas for deterministic review.
The purpose of this meeting was to identify critical areas in the System 80+ design and provide guidance on detailed design for such areas. ABB-CE presented.the general methodology for the design analysis of nuclear island and annex-structures. ABB-CE developed a 3-D finite element static model for these structures by using shell elements for the shear walls and the floor dia-phragms, solid elements for basemat, and soil stiffness r_epresented by Winkler springs. The containment seismic analysis will be based on a separate dynamic response spectrum analysis of the steel containment. vessel alone. -The' results of the 3-D model will be cross-checked with a 2-D stick model used in the dynamic SSI' analysis.
For the design of walls, floors, and connections ~, ABB-CE will apply the loads which are based on the floor accelerations that envelope all soil cases as specified in the ABB-CE application from the SSI analyses, multiplied by the appropriate mass at each elevation of each structure. and distribute the loads in accordance with various wall and floor stiffness characteristics. ABB-CE will conduct three separate static analyses corresponding to the north-south, east-west, and vertical direction, and use earthquake loading.by the.combina-tion of (100 percent,- 40 percent, 40 percent) rule. However, this rule has not been presented in CESSAR-DC and it has.to be evaluated by the NRC staff.
For embedded walls, the NRC staff requested that ABB-CE provide sample calculations to see how lateral forces are modeled from dynamic pressures to l
static loads.
For the design of the basemat, the general methodology is the same;as the design of walls, floors, and connections except activation of soil springs.
The softest. soil case will be used for basemat design.
Loading combinations will be evaluated to simulate potential uplift at each of the four corners of the basemat by the. combination of (100 percent, 40. percent, 40 percent) rule for earthquake and other sustained loads.
For each load case, the-analysis will be repeated until all soil springs are in compression because the soil t
M
.- February 24, 1993:
j springs cannot take tension in case of uplift. This procedure.is to be l
repeated for each of the four corners._ The applied loads are computed by enveloping basemat stresses from each simulation of design. This procedure is to be repeated for a second foundation condition to confirm the assumption that the chosen soil cases control the design.
For the accurate prediction of uplift and slide of basemat, the NRC staff requested that the flexibility of. '
je considered by the finite element approach with the model boundaries determined by.the St.. Venant principle.
ABB-CE committed to perform sensitivity analyses for all soil cases to-demonstrate-the acceptability of the results from models using Winkler soil springs. The staff pointed out that the design of the nuclear island basemat should take into account the construction sequence under soft soil conditions in addition to the evaluation for governing load combinations involving dead load, live load, earthquake, and design basis accident. loading that causes internal pressures in containment and other subcompartments.
for the detailed design of the nuclear island structures, CE provided the general methodology such as:
(1)
Determination of critical structural members from analysis results.
(2)
Determination of critical local loadings that may govern design of -
i members.
(3)
Development of typical reinforcing schemes for walls, floors, basemat, and connections for the existing member sizes.
(4)
Determination of maximum shears, forces and moments-in members by applying the enveloped forces, moments, and any additional loads to members from analysis results.
(5)
Check adequacy of members and connections per Codes and Standards.in CESSAR-DC Table 3.8-4.
(6)
Redesign memvers that do not meet code and standard nquirements.
ABB-CE committed to develop detailed design for:
(1)
Nuclear island basemat.
(2)
Shear walls including connections to the basemat and end walls.
(3)
Floor slab connections to shield building.
i (4)
Free standing portion of the shield building, particularly in the dome region.
(5)
The free standing steel containment structure.
i
?
' February _24, 1993 (6)
Reactor cavity. walls, including static and dynamic pressure capacity for severe accident loading conditions.
The NRC staff requested that reinforcement requirement of the reactor cavity' for the severe accident impulse load be specified. Also, the NRC staff identified two additional areas for detailed design:
(1) steel containment embedment area for the circumferential buckling due to both internal as well as external pressures and seismic loads and (2) the steam generator cavity area.
For additional confidence in the detailed design calculations, ABB-CE will have a peer review performed by Dr. Robert Kennedy in addition to the quality assurance program.
The NRC staff also noted that Table 3.2-1 of CESSAR-DC lists other seismic Category I structures and requested that for safety-related structures and non-safety-related structures that house safety-related systems and equipment such as radwaste building, emergency diesel generator building, and station service water pump structure, design calculations should be completed consis-tent with the CESSAR-DC commitments, and demonstrate the overall integrity of these structures under governing loads and load combinations. Also, the structural dynamic responses (including seismic floor response spectra of these structures at the locations of safety-related systems and equipment) and other structural loadings that are to be applied to safety-related systems and equipment should be completed.
The NRC staff also pointed out inconsistencies in: Figure 3.8-1 in CESSAR-DC for personnel air lock plan view and Figure 3.1-1 in the System 80+ Severe Accident Phenomenology and Containment Performance report for. containment internal structure arrangement.
(Original signed h '
Michael X. Franovich, Project Manager Standardization Project Directorate-Associate Directorate for Advanced Reactors and License Renewal Office of Nuclear Reactor Regulation.
Enclosures:
1.
List of Attendees 2.
Material Presented by ABB-CE cc w/o enclosures:
See next page DISTRIBUTION w/ enclosures:
Docket File-PDST R/F TMurley/FMiraglia,12G18 PDR.
DCrutchfield WTravers RBorchardt-PShea TWambach JMoore,'15GB18 RPerch, 8H7 GGrant, 17G21 DISTRIBj) TION w/o enclosures:
ACRS (11)
NSaltos, 10E4 SJLee, 7H15 RPalla,10E4 GBagchi, 7H15 JKudrick,.8Dl-ADrozd, 8E2 MFranovich
_. w 0FC:
LA:PDST:ADAR PM:PQS AR-..
SC g4 NAME: PShea MFranc h:sg
. RBorcha DATE: 02/pf 02/y7b
.7 02/2Jf93 0FFICIAL: RECORD COPY:
DOCUMENT 4AME:MSUM010.MXF
.ABB-Combustion Engineering, Inc.
Docket No.52-002 cc:
Mr.'C. B. Brinkman, Acting Director Nuclear Systems Licensing Combustion Engineering, Inc.
1000 Prospect Hill Road Windsor, Connecticut 06095-0500 Mr. C. B. Brinkman, Manager Washington Nuclear Operctions Combustion Engineering, Inc.
12300 Twinbrook Parkway, Suite 330
-Rockville, Maryland 20852 Mr. Stan Ritterbusch Nuclear Systems Licensing Combustion Engineering, Inc.
1000 Prospect Hill Road Post Office Box 500 Windsor, Connecticut- 06095-0500 Mr. Sterling Franks U. S. Department of Energy NE-42 Washington, D.C.
20585 Mr. Steve Goldberg Budget Examiner' r
725 17th Street, N.W.
Washington, D.C.
20503 Mr. Raymond Ng 1776 Eye Street, N.W.
Suite 300 Washington, D.C.
20006 Joseph R. Egan, Esquire Shaw, Pittman, Potts & Trowbridge 2300 N-Street, N.W.
Washington, D.C.
20037-1128 L
L
4 9
NRC AND ABB-CE PRA MEETING JANUARY 4 AND 5, 1993 WINDSOR, CONNECTICUT N_AMAJ
.A_EFJIll ATIOJ MIKE FRANOVICH NRC B0B YOUNGBLOOD BNL NICK SALTOS NRC RUPERT WESTON ABB-CE SUE CERALDI ABB-CE DAVE FINNICUM ABB-CE STAN RITTERBUSCH ABB-CE S0TRAB ESTANDIARI ABB-IMPELL MATT JACOB ABB-CE TODD OSWALD DUKE ENGR. & SVCS.
LYLE GERDES ABB-CE SEUNG J. LEE NRC G0VTAM BAGCHI NRC JACK KUDRICK NRC BOB PALLA NRC ROGER WAGSTAFF DUKE ENGR. & SVCS, PETER M. LANG DOE CHUEN-CHING LIN BNL ANDRE DR0ZO NRC RAY SCHNEIDER ABB-CE-DAVID FINNICbM-ABB-CE 1
CE/NRC MEETING ON SYSTEM 80 + PRA
-JANUARY 4 AND 5,1992 AGENDA JANUARY 4,1992 Morning Session (8:30 a.m. - 12:00 p.m.)
y 3
1.
OVERVIEW OF DSER LEVEL 1 PRA OPEN ITEMS:
- Review of ABB-CE Responses
- Cross-Reference with Updated PRA
- Discussion II.
UPDATED LEVEL 1 ANALYSES
- Discussion
- Schedule 111.
FIRE AND FLOOD ASSESSMENTS
- ABB-CE Assessment Approach
- Discussion IV.
PRA-BASED INSIGHTS / INPUT-TO ITAAC/DAC, RAP AND RELATED DSER OPEN ITEMS b
- {
Enclosure '2-i
h 9.
CE/NRC MEETING ON SYSTEM.80 + PRA.
JANUARY 4 AND 5,1992 9
AGENDA-JANUARY 4,1992.
Afternoon Session (1:00 p.m. to 5:00 p.m.)-
V.
OVERVIEW OF DSER SEISMIC OPEN ITEMS AND PRA-BASED MARGINS ASSESSMENT
- ABB-CE Proposed Approach
- Discussion
- Schedule VI.
CONTAINMENT FRAGILITY CURVE Vla. SEVERE ACCIDENT ISSUES RELATED TO STRUCTURAL DESIGN
- Containment Penetrations
[
- Reactor Cavity Design / Capacity.
JANUARY 5,1992 Morning Session (8:00 a.m. to 12:00 p.m.)
Vll.
OVERVIEW OF OSER LEVEL 2 AND LEVEL 3 I
OPEN ITEMS AND RELATED ISSUES ~
- Discussion
- Schedule l
Vll.
SUMMARY
OF ACTION ITEMS AND COMMITTMENTS i
l l
I 1
.\\
UPDATE AND REVISION OF SYSTEM 80 + PRA OUTLINEi 1.
INTRODUCTION 2.
METHODOLOGY 3.
DETERMINATION OF INOTIATING EVENTS 4.
EVENT SEQUENCE ANALYSIS 5.
DATA ANALYSIS 6.
SYSTEMS ANALYSIS 7.
EXTERNAL EVENTS ANALYSIS 8.
PROBABILISTIC SHUTDOWN RISK ANALYSIS
. 9.
LEVEL 1 RESULTS
- 10. SENSITIVITY ANALYSES
- 11. SEVERE ACCIDENT PHENOMENOLOGY
- 12. CONTAINMENT PERFORMANCE ANALYSIS M 1 "9
- 13. RISK ANALYSIS
- 14. CONTAINMENT PERFORMANCE AND RISK ANALYES SENSITIVITY ANALYSIS
- 15. CONCLUSIONS
- 16. REFENENCES
?
s, 5
9 Table 9.1-1 CORE DAMAGE FREQUENCY CONTPJBUTION BY INITIATING EVENT.
MEAN CORE '
DAMAGE ERROR PERCENT INITIATING EVENT FREQUENCY FACTOR
- OF TOTAL 12rge less-Of-Coolant-Accident (LLOCA) 1.6E-07 5.7 7.4 Medium Loss-Of-Coolant-Accident (MLOCA1) 2.1E-07 5.2 9.9 Medium less-Of-Coolant-Accident (M LOCA2) 2.4E-07 5.3 11.4 Small less-Of-Coolant-Accident (SLOCA) 1.6E47 13.7 7.8 -
Iarge Secondary Side Break (LSSB) 2.0E-09 14.8 0.1 Steam Generator Tube Rupture (SGTR) 3.0E-07 12.6 14.3 Loss of Feedwater Flow (LOFW) 3.7E-07 5.8 17.6 Other Transients (TOTH) 3.4E-07 7.1 16.4 Imss Of Offsite Power (LOOP) 1.9E-08 8.0 0.9 Station Blackout with Battery Depletion 2.1E-08 9.2 1.0 less of Component Cooling Water (CCW) Div 2 3.1E-10 10.1 =
0.0 less of 4.16 Kv Bus 2.3E-10 7.3 0.0 loss of 125 VDC Vital Bus 6.9E-11 '
6.5 0.0 -
l' Anticipated Transient Without Scram (ATWS) 1.7E-07 8.2
' 8.1' Interfacing System LOCA 5.2E-10 234.0-0.0 Loss of HVAC 5.2E-09 19.3 0.3 Vessel Rupture 1.0E-07
-10.0 4.8.
Internal Events - Total 2.1E-06
- 2.6 100.0 Tornado Strike Events 2.4E-07 6.6 ~
90.2 Fire (scoping estimate) 1.9E-08 **
7.2 -
Flood (scoping estimate) 6.9E-09**
2.6 -
External Events - Total 2.6E.07**
100.0 Internal Events 2.1E-06 2.6 66.8 External Events 2.6E-07" 8.5 4
Shutdown Risk 7.7E-07**
.2.41 TOTAL 3.1E-06" 100.0 This value represents the mean of the combined accident sequences for intemal initiators, and not the sum of the mean for each internal initiator.
I Best Estimate.
9-3
l SYSTEM 80 + FIRE RISK ASSESSMENT 3
QUALITATIVE ASSESSMENT
- Use FIVE Qualitative Assessment Methodology
- Evaluate Fire Barriers / Separation on Room-by-Room and.
Level-by-Level Basis
- Identify Equipment in Each Area
- Demonstrate That Most Severe Possible Impact of The Maximum Credible Fire Would Be Loss'of One Division.
of All Safety Equipment.
QUANTITATIVE ASSESSMENT
- Assume Maximum Credible Fire f
- Estimate Frequency for Fire Using Generic Industry l
Data
- FIVE Report
+
- Estimate Failure Probability for Detection and Suppression Systems Using Generic Industry Data
- Calculate Frequency for Unsuppressed Fire'.
- Substitute Fire Frequency Estimate for " Loss of Component Cooling Water" Frequency in the LCCW Event Sequences.
- Requentify LCCW Sequences and Sum results To Obtain Quantitative Estimate of Fire Risk.
ESTIMATED FIRE RISK CONTRIBUTION IS 1.9E-08/YR i
i SYSTEM 80+ FLOOD RISK ASSESSMENT i
QUALITATIVE ASSESSMENT l
- Use Methodology'Similiar to FIVE Methodology
- Evaluate Flood Separation on Room-by-Room and Level-by-Level Basis
-Identify Equipment in Each Area
-Identify Potential Flood Sources
- Demonstrate That Most Severe Possible impact of The Maximum Credible Flood Would Be Loss'of One Division.
of All Safety Equipment.
i QUANTITATIVE ASSESSMENT 3
- Assume Maximum Credible Flood 1
- Estimate Flood Frequency Based on identified Flood '
Sources
- Substitute Flood Frequency Estimate for " Loss of Component Cooling Water" Frequency in the LCCW -
t Event Sequences.
- Requantify LCCW Sequences and Sum results To 1
Obtain Quantitative Estimate of Flood Risk.
i ESTIMATED FLOOD RISK CONTRIBUTION IS 6.9E-09/YR.
1 i
P
~
.m.
i l
t SYSTEM 80 + PRA INPUT TO ITAAC/DAC AND RAP i
l DESIGN CONFIGURATION ASSUMPTIONS
- Review Chapter 6 of PRA on System By System Basis
- Summarize Design Configuration Assumed in PRA Analyses For Each System P
E EQUIPMENT CAPABILITY ASSUMPTIONS
.h
- Review Chapters 4 and 6 of PRA
-Identify System Success Criteria and Equipment Performance Assumptions Used in PRA.
- Summarize Assumptions For Each System i
EQUIPMENT IMPORTANCE
- Component importances WRT Core Damage Frequency Is Presented in Chapter 9 of PRA.
- For Each System, List Most important Conponents From The Component importance Analysis List in Chapter 9. of PRA Report
-Identify How Each Listed Component impacts Risk DESIGN RELIABILITY ASSURRANCE PROGRAM
- A Design Reliability Assurance Program Plan Has Been Developed For System 80 +.
- The DRAP Plan Has Been Submitted to the NRC for Review
L-r 3
SYSTEM 80+ ALWR A PRA-BASED SEISMIC MARGIN ANALYSIS GENERAL APPROACH l
l PRESENTED.TO:
U.S. NUCLEAR REGULATORY COMMISSION l
~
l
- j I
PRESENTED BY:
j ABB COMBUSTION ENGINEERING i
(.
JANUARY 4,1993 A EDED
(
7%..>E7 59.,
l av
,. m
.1
,f, J
m a
a a
4 2
P
+
b g:
_4 y
i METHODOLOGY.-
a j
Develop Seismic. Event and: Fault Trees:
Prunning of Trees y
=
' Develop Seismic Fragilities for Selected l
Structures, Systems and Components
.r Use Both the Min / Max and the: Convolution Methods to Analyze Seismic 3 Sequences L Leading
~
to Core Damage a
,ln Evalu'ation 'of Containment Isolation and Bypass 7
i-;
Identification of Plant Seismic Vulnerabilities' Adjustment of Design Committments to Achieve Desired HCLPFs 1
l
.i i
k SI BI'
')
=(
1 4..55.59.
g a
Hj
~
~~
(
3 SEISMIC EVENT AND FAULT TREES I
r OBJECTIVE Develop a math. model that describes the combinations of components failures / successes that may lead to plant damage during a seismic event Obtain an understanding of seismically induced accidents at the plant Dominant contributors to the Ilkelihood of j
~!
core damage and releasa Insights into the plant design (e.g. reduntant safety systems)
E A BD ED v
t ORE
/
y.
m SEISMIC EVENT AND FAULT TREES ELEMENTS OF THE SEISMIC SYSTEMS ' ANALYSIS-Develop. a Seismic Event Tree
- Initiating Event:
Earthquake Ground Motion at the Plant-Site
- Top Events:
Plant components and safety systems required to prevent-and' mitigate core damage &-
release
. Develop fault. trees to. model the.fsilure of Top Events in the event tree
- include Seismic & Random Component Failure.
- include Operator Failure-1-
- Model Possible Dependencies in Component Failure in addition to equipment damage / failure, seismic 4
events
- may, Fall / damage structures P
- Damage' Underground Piping A ED ED e
M.. ED ID o,
r 3
SEISMIC EVENT AND FAULT TREES ELEMENTS OF THE SElSMIC SYSTEMS ANALYSIS Fault Tree Analysis includes, Component Seismic Failures Random Component Failures Elements of the Fault Tree Top Event Gates (and/or)
ADD
(
tEEEB 2".A BRCrwh BW[fu A$f j
r 3
PRUNNING OF EVENT / FAULT TREES h
The-Outcome from 1st task will 'be a detailed list of Structures,. Systems, and Components' (SSCs) for which' Seismic Fragilities (or HCLPFs) is-required t
Prunning the -Trees involves detailed Interaction with-the fragility analysts to identify critical components 1
Screen out strong / rugged components Provide fragility analysts with success criteria i
Significant communication between. Systems analyst-and the Fragility analyst is expected. at this stage' For Screened out SSCs, a surrogate fragility based on generic fragilities will be used Other than above, use of Generic Fragilities will be minimized The outcome will be a reduced list of SSCs for which fragilities /HCLPFs is required a
Y A ED ID e
M EDID AstA BRoww BWERI -
r DEVELOP PLANT SPECIFIC SEISMIC FRAGILITIES Develop Approximate Fragilities Based on an Approach as Suggested by Dr. Robert P. Kennedy Use this approach for 1st pass. evaluation in determining plant seismic vulnerabilities For Dominant contributors only, evaluate SSC fragilities using rigorous techniques 1
-j The approximate approach for 1st pass evaluation. Is-outlined below:
Determine SSC HCLPF84 using EPRI CDFM approach as per EPRI-NP-6041-SL, Rev.1-For RLE, use CMS 3 spectral--shape and use Rock j
Site conditions l
1 Report HCLPFs based on PGA L
Compute HCLPFso= CDFM/1.2 L
Assume Sc=0.4 l
Median Acc. = HCLPFso e(**2.3pc)
L
= 2.1 CDFM
\\
Based on Median Acceleration, and pc, Fragility-Curve can be -constructed f
A ED ED -
t ED 7%. EF.n,
.4 i
a 2
,c
+
3 r
DEVELOP PLANT SPECIFIC SEISMIC FRAGILITIES Using This Approach, Can Obtain Risk Numbers Which are Within o Factor of 2 of the actual Risk numbers in a Full SPRA.
The prediction is usually on the conservative side Using the CDFM Approach, it -is expected that the most contribution to Seismic Margin (HCLPF) will come from the response side:
Use of single site category _ (rock) versus 3
a suite of 13 site conditions Use of one control motion (CMS 3) versus envelop of 3 motions Various other factors on the response side as outlined in EPRI NP-6041-SL, Rev.1, such as frequency ' shifting and spectral peak clipping i
if detailed design is available,. then margin on the capacity side may also be computed l
if SSC specific HCLPF falls below desired level (0.6g-for Containment isolation and bypass, and 0.5g for.
components contributing to core damage),
committments in design criteria will be made accordingly
't A ED ED t
M EDER-
' I m-.
+
4.
3.00 2.50 Design iasis r
CMS 3 Rock Site (0.59)
CMS 3 Rock Site (0.3g)
-.2 l
3 1.50 if 1e.
k k,,
.i
~
m.
,.x j
^m
%g i _..!
~~"W Y
" 'o
-J x
/
d f
O.00
. 0.10 IM Frequency (liz)
Comparison of Example Design' Basis Spectra Lwith SMA CMS 315% Damped --Raw-Spectra.
(Elev. :._+50* Top of Basemat _'E-W Direction)
~
.r y
3 1
f DETERMINE SEISMIC SEQUENCES LEADING TO CORE DAMAGE Based on the suggested approach,. for each SSC-on the comp'onent list, both HCLPF84 using CDFM approach and approximate fragliity data will be available Thus, the option of using both the Min / Max and the Convolution Approach is open to analyze.
seismic sequences which could lead to core damage Once Plant ' seismic vulnerabilities for beyond Design Basis Earthquake are identified, committments will be made in design to'
[
ensure a desired HCLPF value I
v j
MENED ASEA BR>h BWERI
[,
n
(*_
y 3..
1
~
-CONTAINMENT ISOLATION & BYPASS Identify. each cutset :whose HCLPF is less 'than 0.6g considering random failures and :using1the.
l Min / Max. Approach j
Evaluate'HCLPF for Active and Passive SSCs Important to containment Isolation
.l q
Examine the ruggedness of potential ~ containment S
Bypass paths to identify any :HCLPF below. 0.6g j
!q
~{
Report any SSCs identified as a resultiof' thef above 1
procedure and discuss the effects of combination
~
of sequences-identified in 1st-step abovefwith i
potential containment bypass-and Lisolation :. failures j
y 1
l f
I A-ED ED i
.t 1.%. E9,Ep.
s _,
.. =
-