ML20127N351
| ML20127N351 | |
| Person / Time | |
|---|---|
| Issue date: | 01/22/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-A-45, REF-GTECI-DC, TASK-A-45, TASK-OR ACRS-2274, NUDOCS 8505230500 | |
| Download: ML20127N351 (40) | |
Text
.
spas-w e
~
~
P M of/3J3 ACRS DECAY HEAT REMOVAL SYSTEMS SUBCOMMITTEE MEETING
SUMMARY
JANUARY 22, 1985 ALBUQUERQUE,NM
Purpose:
The purpose of the meeting was for the Subcommittee to j
continue its review of the NRC Staff resolution effort for USI A-45
" Shutdown Decay Heat Removal (DHR) Requirements".
Meeting Highlights, Agreements and Requests 1.
The focus of the meeting was on the SNL Shutdown DHR Analysis Plan
- in particular, the analysis methodology for the external event contributors (seismic, fire, internal and external floods, extreme wind and tornado, sabotage).
2.
The status of the plant analysis work was overviewed. A series of visits to nine plants were scheduled, the results to be used to develop a set of plant groups for resolution of this issue on a generic basis. To date, the Point Beach and Quad Cities plant analyses are essentially complete. The rest are in earlier stages of work. The last two plant visits (San Onofre and Grand Gulf) were added to the list at the behest of the ACRS. Mr. Reed questioned the selection of iust one B&W plant for a site visit.
NRC responded that limitations of program resources and the small number of B&W plants dictated this choice.
3.
The seismic risk methodology for the SNL DHR Analysis Plan was reviewed by M. Bohn (SNL).
Seismic hazard curves are used to characterize a given site if they are available.
If no hazard curve exists, a hazard curve scaled to an exceedance probability of SE-4/ year at the SSE value would be used. The slope of the curve for higher peak ground acceleration (PGA) values would be estimated from other hazard curves for the same broad seismological province (i.e., central stable region, eastern coastal region, southern states,etc.).
Seismic response to structures and components will be performed on a best estimate bases to the extent possible. SNL DESIGNATED ORIGINAL 8505230500 850122 PDR ACRS 2274 PDR Cortified By
/8p
\\
)
I %.
DHRS Meeting Summary January 22, 1985 will'use a hierarchy of methods to estimate seismic response depending on what is feasable given Program constraints.
4.
The. fire and internal flood risk analysis methodology was discussed by T. Wheelis (SNL).
For the fire analysis, the scope includes the following:
The analysis will not replicate a fire PRA, although a similar.
(PRA) approach is used.
It will utilize existing data from published fire PRAs.
The level of detail is determined by systems analysis and readily obtainable documentation.
The analysis will evaluate residual risk due to fires.
For the internal flood analysis, the scope and analysis assumptions are similar to the fire analysis (Figs. 9&l0).
Proposed fixes include:
(1) installation of automatic flood detection systems, (2) construction of flood barriers, and (3) relocation of redundant-system components.
In response to Mr. Michelson, fir. Wheelis said the internal flood analysis does not address rupture of high energy lines out side containment. Mr. Michelson objected to this
'fter extensive Subcommittee discussion on this point, A
approach.
it was determined that SNL would look into this issue and provide additional information to the Subcommittee.
5.
Mr. McCann (JRB Associates) described the external flood and wind analyses. _These analyses take a PRA approach, but due to resource constraints have relied heavily on the results and experience of past PRA's, plus studies of the subject topics.
The final product is a core melt frequency from the subject risk contributors.
3-January 22, 1985 DHRS Meeting Summary 6.
J. Simon from the United Engineers and Constructors discussed the impact analysis methodology in closed session. The objective of this work is to' determine the impact.(cost) of proposed solutions to A-45 to support Sandia value/ impact assessments. The final UEC product is a plant. impact assessment report. The program consists-of three phases:
Phase I - initial design proposal (s); Phase II -
-on-site review plan; Phase III - final design assessment. The cost estimates are so structured in order to be applied on a generic basis for a given modification proposal.
Dr. Sullivan expressed a concern that the UEC design assessment raised the question of who is designing the plant - NRC or the licensee? Mr. Marchese said some. conceptual design work is needed as input to the value/ impact study. A cost / benefit study is now required by CRGR as part of the NRC generic backfit process.
i=
7.
In open executive session, Mr. Marchese discussed the potential schedular impact of slipping the A-45 resolution schedule. There have been a number of schedule problems which necessitated a decision on whether or rct to slip the overall A-45 program resolution plan by V 6 months. A set of progran options were considered.
NRC nanagement decided to hold the existing schedule and develop the A-45 resolution package based initially at least on two plant analyses. The remaining plant analyses would be completed on an adjusted schedule. The resolution package would be modified as necessary based on the results of the additional plant analyses. Additional discussion suggested that it may be of benefit to the NRC to proceed as noted above, at least internally, in order to help jell the Staff's A-45 resolution approach and methodology.
G. Reed objected to the Staff approach, absent inclusion of the B&W plant analysis in the initial resolution package.
l
.-,e
---w
~ - - - - -
m
,.r-,.
r
- - < - - -, - ~
k.b$*d313U. fl{4
}
U. }y
} - ?(' ( ?L, !,t, a,l Date Issued: 2/4/85 g
a Oj ACRS DECAY HEAT REMOVAL SYSTEMS SUBCOMMITTEE MEETING MINUTES JANUARY 22, 1985 ALBUQUERQUE,NM
Purpose:
The purpose of the meeting was for the Subcommittee to continue its review of the NRC Staff resolution effort for USI A-45
" Shutdown Decay Heat Removal (DHR) Requirements".
/.ttendees:
Principal meeting attendees included:
ACRS NRC D. Ward, Chairman A. Marchese J. Ebersole H. Lewis C. Mark Sandia Laboratories C. Michelson D. Erickson G. Reed M. Bohr I. Catton, Consultant T. Wheelis P. Davis, Consultant W. Crammond V. Schrock, Consultant H. Sullivan, Consultant JRB Associates T. Theofanous, Censultant M. itcCann P. Boehnert, Staff UEC J. Simon Meeting Highlights, Agreements and Requests 1.
Mr. A. Marchese (NRR) introduced the topics for the meeting discussion. The focus of the meeting was on the SNL Shutdowi DHR Analysis Plan in particular, the analysis methodology for the external event contributors.
In response to Dr. Mark, Mr. Marc'1ese said that the flooding analysis only focuses on what DHR equipment may be impacted by flooding.
DHRS Meeting Minutes January 22, 1985 2.
D. Ericson provided a status report on the overall USI-A-45 effort.
He overviewed the key A-45 tasks (Fig. 1) and the overall program approach (Fig. 2). Currently, the program focus is on Item III of Figure 1.
In response to a question from Mr. Ebersole, Dr. Ericson said the SB LOCA studied is on the order of 1-1.5 inch equivalent diameter. Mr. Michelson raised the issue of breaks outside of containment and their treatment vis-a-vis A-45.
NRC indicated that they are not addressing this specific issue.
The status of the plant analysis work was overviewed. A series of visits to nine plants were scheduled te develop a set of generic plant groups for resolution of this issue. To date, the Point Beach and Quad Cities plant analyses are essentially complete. The rest ere in earlier stages of work (Figs. 3&4). The last two plant visits (San Onofre and Grand Gulf) were added to the list at the behest of the ACPS. Mr. Reed questioned the selection of just one B&W plant for a site visit.
NRC responded that limitations of program rescurces and the small number of B&W plants dictated this choice.
Drs. Mark and Lewis raised the issue of the regulatory philosophy behind the A-45 effort. After some discussion, it was noted that the program is trying to sort out what is an unacceptable risk vis-a-vis plant DHP. systems.
In repsonse to Mr. Ebersole, Mr. Marchese said he invisioned that a set of DHR criteria for future plant designs will be part of the final product of the A-45 Program.
In response to dr. Mark, SNL said all plant analyses should be complete by August 1985. Dr. Theofanous questioned the usefulness of attempting to extrapolate cora nelt damage values to the cost / benefit analyses given the limitations inherent in both analysis methodologies.
0
DHRS Meeting Minutes January 22, 1985 G. Reed suggested that NRC adopt a system similar in approach to the FAA Designated Representative in order to gain better control of quality in design and manufacture of nuclear power plants.
3.
The seismic risk methodology for the SNL DHR Analysis Plan was reviewedbyM.Bohn(SNL).
Figure 5 lists the steps taken for the seismic PRA study.
Key points noted include:
Seismic hazard curves are used to characterize a given site if they are available.
If no hazard curve exists, a hazard curve scaled to an exceedance probability of SE-4/ year at the SSE value would be used.
The slope of the curve for higher peak ground acceleration (PGA) values would be estimated from other hazard curves for the same broad seismological province (i.e.,
central stable region, eastern coastal region, southern states,etc.).
Figure 6 gives an example.
SNL has also attempted to bound the uncertainty for these hazard curves (Figure 6-A).
Failing adoption of this proposal by the NRC, SNL recocrends a fall-back of using NRC-sponsored eastern United States hazard characterization input to derive a local hazard curve.
Regarding fragilities, component fragilities will be considered but due to program constraints, structural fragil'ities will not be considered. Therefore, the resulting seismic core melt frequencies would not directly comparable to the results of other PRAs which explicitly include structural failures in their seismic risk assessments.
In response to Mr. Ward, NRC said they are not really focusing on mitigation systems for their analyses. As a result of further discus-sion, it was suggested that SNL analyze the potential for, and impact of, a cooling tower failure during an earthquake. SNL agreed this was a good suggestion.
DHRS Meeting ~ Minutes January 22, 1985 For considerations of structural and piping failure, gross structural collapse has not been found to be important and would not be included in TAP A-45 analyses.
Local structural failures (masonry walls, crane, etc.) would be included as determined from a plant visit.
Inertia-induced piping failures would not be included.
Interbuilding pipe failure due to soil failures would be included, if necessary.
Seismic response to structures and components will be performed on a best estimate bases to the extent possible.
SNL will use a hierarchy of methods to estimate seismic response depending on what is feasable given Program constraints (Fig.6-C).
.The steps taken to calculate the probabilistic failure calculations are noted in Figure 7.
4 The fire and internel flood risk analysis methodology was discussed by T. Wheelis (SNL).
For the fire analysis, the scope includes the following:
The analysis will not replicate a fire PRA, althcugh a similar (PRA) approach is used.
It will utilize existing data from published fire PRAs.
The level of detail is determined by systems analysis and readily obtainable documentation.
i l
The analysis will evaluate residual risk due to fires.
Figure 8 details the analysis assumptions.
Failure criteria used are:
for cables-auto ignition temperature or time at temperatur?;
- m. -
DHRS Meeting Minutes January 22, 1985 circurity - temperature at or above 150'F; motors - 20% above maximum operating temperature limits. Mr. Michelson raised the issue of.inadvertant fire supppression actuation. Mr. Wheelis said they are not addressing this issue.
He did note that SNL has some work underway in this area though as part of another project.
Dr. Theofanous expressed a concern that some of the analysis assumptions are based on "that's the best we can do" or its the
" state of the art".
He wasn't sure how to interpret the analysis results, given this situation.
The approach used by SNL for the plant analysis is four stepped:
(1) identify potentially significant fire areas, (2) visit the plant, (3) quantify the results, and (4) recomend proposed fixes.
Mr. Reed asked if the proposed fixes in Item 4 would be reviewed by the NRC Staff vis-a-vis Appendix R requirements. Mr. Marchese and SNL indicated these fixes would be integrated with the Appendix R effort.
There was discussion of addressing manual fire suppressier. and the possible damage to equipment by the fire teans.
SNL said they perform a separate analysis of this possibility to show the core melt contribution from this event.
For the internal flood analysis, the scope and analysis assumptions are similar to the fire analysis (Figs. 9&l0).
Proposed fixes include:
(1) installation of automatic flood detection systems, (2) construction of flood barriers, and (3) relocation of redundant system components. The failure criteria used are:
for electrical cabinets - six inches of standing water; for motors - dependent on specific design.
In response to Mr. Michelson, Mr. Wheelis said the internal flood analysis does not address rupture of high energy systems out of J
1 DHRS Meeting Minutes January 22, 1985~
containment.
Mr. Michelson objected to this approach. Mr.
Marchese said he would reexamine this point.
.Mr. Reed noted that' high energy line systems have been analyzed in other contexts.
After extensive Subcommittee discussion on this point, it.was detennined that SNL would:look into this issue and provide additional information.to the Subcomittee.
5..
Mr. McCann -(JRB Associates) described the external flood and wind analyses. These analyses take a PRA approach, but.due to resource constraints have relied heavily on the results and experience of past PRA's, plus studies of the subject topics.
Figure 11 is_a schematic of the study approach used. The final product is a core melt' frequency from the subject risk contibutors.
The flood risk analysis for each plant is a 3-man week effort. A
. set of hazard curves is developed (Fig. 12) and a fragility analysis is conducted for plant components and structures. These two studies are used as input to the risk analysis (Fig. 13) to develop a fragility curve as a function of systen failure probability (Fig. 14). The final result is the core melt probability (Fig. 15).
The wind hazard and fragility analysis is similar in approach to thefloodanlaysis(Fig.16). NRC has provided the mean hazard j
curve for tornadoes and straight winds. The fragility analysis j
involves. identification of vulnerable components via a plant walkdown and document review (Figs. 17&l8). JRB Associates conducted a tornado uncertainty analysis (Fig. 19). The product is a set (5) of tornado hazard curves with-the mean curve equal to the
.NRC curve. The above information, along with a missile impact and systems failure analysis, are input to the risk analysis to develop the' core melt probability (Fig. 20).
~
DHRS Meeting Minutes January 22, 1985 6.
J. Simon from the' United Engineers and Constructors discussed the impact analysis methodology in closed session. The objective of this work is to determine the impact (cost) of proposed solutions to A-45 to support Sandia value/ impact assessments. The final UEC product is a plant impact assessment report. The program consists of three phases:
Phase I - initial design proposal (s); Phase II -
.on-site review plan; Phase III - final design assessment. The cost
~
estimates are so structured in order to be applied on a generic basis for a given modification proposal.
Dr. Sullivan expressed a concern that the UEC design assessment.
raised the question of who is designing the plant - NRC or the licensee? Mr. Marchese said some conceptual design work is needed as input to the value/ impact study. A cost / benefit study is now required by CRGR as part of the NRC generic backfit process.
7.
D. Ericson discussed the sabotage analysis methodology in closed session. The central approach taken by SNL is to evaluate plant vulnerability to sabotage and then factor this risk contributor into the anlaysis methodology. Just how this "unquantifiable" factor will be evaluated is not completely clear to the NRC at this time.
8.
In open executive session, Mr. Marchese discussed the potential schedular iinpact of slipping the A-45 resolution schedule. There have been a number of schedule problems (Fig. 21) which necessitated a decision on whether or not to slip the overall A-45 program resolution plan by N 6 months. A set of program options were considered (Fig. 22).
NRC management decided to hold the existing schedule and develop the A-45 resolution package based initially at least on two plant analyses. The remaining plant analyses would be completed on an adjusted schedule. The
DHRS Meeting Minutes January 22, 1985 resolution package would be modified as necessary based on the results of the additional plant analyses.
Mr. Reed said he objected to basing the resolution package on no input from the B&W plant analysis. Mr. Ward noted that the variabilities in DHRS design call this effort into question.
Mr. Mark noted the illogic of resubmitting a second resolution package to CRGR based on information received from the later plant analyses.
Dr. Theofanous supported the NRC approach because he said it should force a coming to grips with the resolution of this issue and should help focus the methodology needed for the resolution Plan.
Mr.
Reed disagreed, noting tnat it is a half-baked solution that will lead to delay in overall program resolution. Mr. Ward said Dr. Theafanous' approach may be good for internal NRC delibera-tions, but will be awkward when taken to the public.
Mr. Sullivan agreed with Dr. Theofanous that the NRC approach may help jell the resolution plan.
9.
An overview of the Point Beach DHR Analysis results was discussed by W. Crammond (SNL) in closed session.
Figure 23 shows a schematic of the DHR analysis.
Figure 24 shows the initial insights gained from the Point Beach analysis. While the preliminary'results seem to indicate signfiicant improvement in core melt probability with minimal fixes, the non-quantifiable items such as sabotage still need to be factored into this analysis.
L
- 10. The Quad Cities preliminary analysis was reviewed by S. Hatch, SNL in closed session.
Figures 25-26 summarize the preliminary results and insights of the analysis. Mr. Hatch reviewed the plant vulnerabilities and proposed DHR modifications to address these
9-January 22, 1985 DHRS Meeting Minutes vulnerabilities. Mr. Davis questioned some of the credits given for some plant recovery sequences.
- 11. The. meeting was adjourned at 7:35 p.m.
NOTE:
Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, N.W., Washington, D.C., or one can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Washington, D.C. 20001 (202) 347-3700.
d e
1 6
8 b
J
-m-r
--+*r+----
T--, -'-- *-- -----*-
1--
w--e-
,----ra me w
,-r-e---
v---
-e--
e e-
P i
O FOUR KEY TASKS TO REACH M JECTIVES I DEVELOP CRITERIA FOR SCREENING PLANTS FOR DHR CAPABILITIES II APPLY CRITERIA TO PLANTS AND IDENTIFY THOSE WITH POTENTIAL D M PROBLEMS III ASSESS IN MORE QUANTITATIVE (AND QUALITATIVE) DETAIL THOSE PLANTS WITH 4 '
POTENTIAL EAKNESSES TO DETERMINE DHR CAPABILITIES IV PERFORM VALUE-IMPACT ANALYSES OF MODIFICATIONS DESIGNED TO IMPROVE DM RELIABILITY 9
0 9
e e
S 3
g<
-.~,
memtar m annissa 84E15 m assellait I
ILES M Mm W
..~ %}
mMT1 c..
assasss
~%
emmiswmte,emust sweettsis
- IstK
/
g amaan y
8EER885
-~
/
EWLW a auf, farm
- ~**#
- <sumes g"u an ur,7 mi a s % _,, -
yy anwiraim m,e,rg,mm -p j
mesamant monam muss aa same m asams 9
e aims a
-mmasmr
- muus -
e unanims III suas 6
m,ums, sawsr av en
._,,mmuseumwa
= = = = =
amanseems -
n.,,,,,,,es mawrrus
,uim mt, mums muens awm mnie wrM C'" H85 ** -
- 15
~ "" d" """
I"sem'"ia"s" O
s
,maat \\
ammuweimism i* seama up ASSESS lifett e naus asu uss e nas m f%
e
~
suum -
IMEIEEElt L
',,,,::; s R8EI c
t asseurs,
g
\\,,
f5fElat imaidm1 a 4 M s
g
<Qs N-s..
3
~,.-
Cck./Eh7
,Jn c.w,(
~
I SPECIAL ISSUES:
PORVs for CE SYSTEM 80 PLANTS EFFECT of SAFETY GRADE COLD SHUTDOWN SYSTEMS S O
e 9
S e
/
p y :
SPECIFIC PLANT ANALYSES STATUS TURKEY POINT
- INITIAL PLANT VISIT SEPTEMBERp 1984
- QUANTIFICATION OF INTERNAL SEQUENCES IN PROCESS
- INITIAL FIRE AND INTERNAL FLOOD ANALYSIS COMPLETE
- EXTERNAL FLOOD AND SEISMIC IN PROCESS TROJAN
- INITIAL VISIT NOVEMBER 1984
- FAULT TREES BEING REVISED
- PLANT DATA BEING GATHERED COOPER
- INITIAL VISIT DECEMBER 1984
- SIMPLIFIED SCHEMATICS BEING PREPARED
- ADDITIONAL PLANT DATA BEING GATHERED
~
9 9
9
i 1%
i 1
k k
SPECIFIC PLANT ANALYSES STATUS ST. LUCIE
- INITIAL VISIT DECEMBER 1984
- DATA RECEIVED ANO 1
- INITIAL VISIT NOVEMBER 1984 l
- SCHEMATICS BEING RECHECKED
- FAULT TREES STARTED l
i SAN ONOFRE
- NO WORK To DATE
- INITIAL PLANT VISIT IN MARCH (TENTATIVE)
GRAND GULF
- NO WORK TO DATE i
- INITIAL PLANT VISIT IN MARCH (TENTATIVE) l 4
l l
l 1
%'k' I
i STEPS IN A SEISMIC PRA l
i
-1.
DETERMINE THE LOCAL EARTHOUAKE HAZARD.
i i
2.
IDENTIFY ACCIDENT SCENARIOS FOR THE PLANT WHICH LEAD TO RADI0 ACTIVE RELEASE.
(INITIATING EVENTS AND EVENT TREES).
3.
DETERMINE FAILURE MODES FOR THE PLANT SAFETY AND SUPPORT SYSTEMS.
(FAULT TREES) 4.
DETERMINE FRAGILITIES (PROBABILISTIC FAILURE CRITERIA) FOR THE i
Iif0RTANT STRUCTURES ME) C0tPONENTS.
)
{
5.
DETERMINE THE RESPONSES (ACCELERATIONS OR FORCES) 0F ALL STRUCTURES AND C0fFONENTS (FOR EACH EARTH 00AKE LEVEL).
i 6.
COPFUTE THE PROBABILITY OF CORE DAMAGE AND RADI0 ACTIVE RELEASE USING THE INFORMATION FROM STEPS 1 THROUGH 5.
7.
ESTIMATE UNCERTAINTY IN C0tFUTED RESULTS.
l j
)
COMPARIS0il 0F EASTERil HAZARD CURVES FROM RECEi4T PRAs 10-2 p i
\\
\\
10'3 -
~
/
n G
\\
SSE VAwEr 4
\\
b I 't. a/
I s
m g
\\
'N 10 '
i E
'g S
'4 w
c
\\,, - g a
' s4,-
N g
10-6 '-
.'k A'c&f.
gi
' q % # s.s, E
' s.
=-
1 I
10~7 l-I 10-8'-
a i
i
~-
0.2 0.4 0.6 0.8 1.0 PEAK GROUllD ACCEL. (G) fib.4
M e1.,*.'%,$,
/
R
%,N
-df I
,a
=
c
= r s
1,- -
p f/
I-). _
3 3
'//
j
- ]
_3
/
i
) :'T '-
j y
m jE
- a g
ee f
/-
E' u
W
/,/
g'~
~
3 4
i i
i 3:
i 5
s 1
M 2:
g ;Q=
e 5
g=
=
C 8e 5
o s
x x x x a:
Q g
k W
N N
I E
w w=
=
.3*
e4 e
e
- ee ee
- e e-ee o eee e
e e e oo
= e oeo oo e e e see e o o o. m-w e se ee e
se -e
,e e
e
T LEIERARCHY OF METHOD _S__WILL BE USED TO ESTIMATE RFSPONSES GENERIC SCALING FOR ROCK / DEEP SOIL (HALF SPACE) SITES AND MAT FOUNDATIONS e
LLNL RESULTS TO RELATE FIXED BASE SPECTRA TO SPECTRA INCORPORATING SSI EFFECTS SIPFLE STRUCTURE MODELS (IF AVAILABLE) OR RIGID BLOCK MODEL (KAUSEL.
1978) WITH NEW BEST-ESTIMATE SOIL SPRINGS (FREQUENCY DEPENDENT) i
\\
\\w
m hkud!
b
.O v
W i
w a
i N
2 m
-lg g
N i
4.G a
w w
g W
M w
a
=
k G
g E
5 o
g g
Q w
=
=
W w
(
m W
$E C
A m
=
w W
w E
M Y
w g
m s
m O m
N Gg W
W d
i w
y a
s a
ya e
=w aW W
m f
M w
a W Wy W I
=
1 w
8 w
o g g a
i "gW zgiN i
W 55 o
MggEE 4
i E
a wE g-E w
au-i e
D
i l
FIRE ANALYSIS I
ASSUMPTIONS i
SINGLE AREAS EXPOSURE FIRES OF PRIME INTEREST.
ANALYZE FOR TRASH / POOL FIRES j
COMPARABLE TO UTILITY ANALYSIS COMPONENTS AFFECTED BY A
FIRE ARE j
I.
ELECTRICAL
/
ACTIVE COMPONENTS s'
OFFSITE POWER CONSIDERED AVAILABLE UNLESS THE FIRE CAUSES ITS LOSS j
THERMAL LAG TIMES NOT CONSIDERED FIRE GROWTH AND SPREAD INDEPENDENT f
OF SUPPRESSION CONTROL SYSTEM INTERACTIONS NOT ASSESSED N(
s NO SEISMIC INDUCED FIRES g
Ch l ~}
SCOPE OF ANALYSIS g,
ANALYSIS WILL NOT REPLICATE AN INTERNAL FLOOD
- PRA, ALTHOUGH A
SIMILAR APPROACH IS USED.
j.
UTILIZE EXISTING DATA FROM PUBLISHED INTERNAL FLOOD P R A's @.
h p.
LEVEL OF DETAIL DETERMINED BY A.
SYSTEMS ANALYSIS 8.
READILY OBTAINABLE DOCUMENTATION EVALUATE RESIOUAL RISK DUE TO INTERNAL FLOODS.
I s
4 a
t
\\l
f.
1 I
i INTERNA 4L' FLOOD ANALYSIS
'ASSUMPTIONS e
STRUCTURAL INTEGRITY OF WALLS MAINTAINED s.
IF COMPONENT IN LINE OF SIGHT OF SOURCE.
FAILS UNLESS SPRAY TIGHT ONLY SINGLE FLOOD SOURCES CONSIDERED PIPES BREAK IN ACCORDANCE WITH STANDARD REVIEW PLANS 3.6.1 AND 3.6.2 OFFSITE POWER CONSIDERED AVAILABLE UNLESS THE FLOOD CAN CAUSE ITS LOSS CABLES UNDAMAGED IN A FLOOD NO SEISMIC INDUCED, INTERNAL FLOODS CONTROL SYSTEM INTERACTIONS NOT ASSESSED FRONT LINE SYSTEMS NOT FLOOD SOURCES 9h 5
NI
N.
N.
t y
l c
e n
M eu e
q r
e o
r C
F Dj n
Dj o
i tac i
f i
tnau Q
y k
I t
i s
s i
R ne g
P n
I draz la l
Dl m
s isy l
an A
sm n
e t
o ss s
y i
y t
t ee y
t i
/a ee S
s yu i
rr s
TT t
n tl
\\
s n
n e
ia
'\\
i e
t l v s
tt a
n iE t
y l
n nl l
i g
eu P
ay a
I va A
d EF q
d rt n
r Fi r
a l
z ti d
a a
nb r
z a
la ea l
g eE%51 l
eh b.
nr l
a J
oe l
o D53Em fxiIuiJ pn l
ml
~
ou CV e-l a n
bt n
o an w
i l e o
m d
t i
a au k
tz vc l
ni ao a
ar d
w l a w
Pi et t
nn n
l i
i vao a
m eli l
a R pt P
F e
e
- fWEYrEE[
g
. d FE
, 4s
l a,
Ipg = 1.0 1
u 3
}
.p 3
E sp.
E E
"1
~
.p y t
s p2 g
1 u
Peak Discharge Peak Discharge 5
5 I I p q) = 1.0 j
2, s,
ij 9
(
1 N
P411 D
E 92 o
I P412 N
I q) = 1.0 E922 d
Peak Discharge Peak Stage STEPS TO EVALUATE THE PROBABILITY DISTRIBUTION ON THE FREQUENCY OF FLOOD STAGE Jack R. Benjamin & Associates,Inc.
Consulting Engineers B
f/6. /5
RISK 00ANTIFICATION i
j INPUT FLOOD HAZARD CURVES FRAGILITY DATA SYSTEMS MODEL I
_ ANALYSIS STEPS /DRODUCTS STEP:
FOR EACH SEQUENCE, COMBINE THE COMPONENT FRAGILITY YALUES ACCORDING TO THE SYSTEMS LOGIC, AT EACH FLOOD INTENSITY VALUE 1
PRODUCT:
CONDITIONAL FRACTION OF SYSTEM FAILURE DEFINED IN TERMS OF FLOOD INTENSITY 1
STEP:
COMBINE FLOOD HAZARD DATA AND SYSTEM l
FRAGILITY CURVE PRODUCT:
FREQUENCY OF CORE DAMAGE llNCERTAINTY AN'ALYS1.S PR0v! DES THE PROBABILITY DISTRIBUTION ON THE FREQUENCY OF CORE DAMAGE Jock R. Ben)omin & Assocle'es,Inc.
Consulting Engineers B
Risk Quantification H
d Hazard Curve
_1. 0 System Fragility
- 0.8 Frequency of Failure, Pf
?3 Yu 8
b Y
- 0.6 w
o E
L E
E
- 0.4 y
w
{
u.
- 0.2
\\\\\\
0 s
Flood Depth Intensity I
Jack R. Bonjomin & Associates,Inc.
9 Consulting Engineers
o
'o
.C e
~
o
=
=
=
t s
E N
.f.
'O 4E O
so C
Z W
=
W O~
=
d DJ b
=
g d
W
=
ao "o
8 f
CE M
W O
o K
-E I
E
.J
=
D E
Zd
~
l T
= e
.e e
m to
'O I
I I
I 8
C D
D.
w o*
e w
d A.Li'llsV80Wd Jock R. Benjamin & Associales,Inc.
Consulting Engineers 9
Frt/
rl
i jI lI dny y
es e
s it t
Wns l ti ons ri ics i i di i dus say t as nt y ht y nsy aas spl isl ma grl i
oe a ie a We a rrl rn MI n Tcn rnA PA A
acn nA tU U
S r
r no S
i es I
t lt S
a bn Y
d c
ae L
Cnr if rn A
R aa f oeo N
N ez i
np m A
SMa y
t l
U H
t n
uo i
e VC YT l
d I
d i
r L
i g
I a
a G
z r
A a
F R
H F
DNA DRA ZAH e
D t s N
it I
wSn W
e e
i&m v
u et c RnoaD
^
l P
no i
t maz tno ad l k a
Pl i
a l
W ima F
l I
W g6E, e3 g5 > k f
8-W ogg 5 j-ws% k%
7 IDENTIFICATION OF VULNERABLE COMPONENTS (FOR MISSILE & WIND PRESSURE) 9 EXPOSED E0ulPMENT TANKS PIPES AIR INTAKES EXHAUST STACKS 9
EQUIPMENT PROTECTED BY CONCRETE BARRIERS LESS THAN 12 INCHES THICK 9
EQUIPMENT EXPOSED TO MISSILES THROUGH OPENINGS DOORS LOUVERS
)
flETAL BUILDINGS
)
9 STRUCTURES NOT DESIGNED FOR PRESSURE DROP i
IllTERIOR f! ALLS (MASONRY)
O Jock R. Benjamin & Associates,Inc.
Consulting Engineers S
REVIEW PLANT AND SITE DOCUMENTS e
BUILDING CONSTRUCTION DRAWINGS e
FSAR e
DESIGN IOPICAL PEPORTS DAST WIND HAZARD AND STRENGTH STUDIES e
e NATIONAL SEVERE STORM FORECAST CENTER (NSSFC)
TORNADO STATISTICS AND VIOLENT STORM CLIMAT0 GRAPHY tlUREG/CR-3670 (F-Li & F-5 TORNADOS 1880 TO 1982) e STRAIGHT WIND DATA e
NUREG/CR-2980 (GREAT LAKES) e NUREG/CR-2639 (GULF 3 ATLANTIC COAST) e NBS 118 (U.S.)
e NBS 124 (HURRICANE)
Jock R. Benjamin & Associalet,Inc.
Consulting Engineers l#/LM
r l
TORNADO UNCERTAINTY ANALYSIS e
PROCEDURE DESIGNED BY REINHOLD:
TORNADO DAMAGE RISK ASSESSMENT NUREG/CR-2944 e
CONSIDER UNCERTAINTY IN FOLLOWING AREAS:
STRIKE MODEL PLANT SITE FIDTH TORNADO ANALYSIS DATA TORNADO DAMAGE AREA AND LENGTH VARIATION F-SCALE I!!ND SPEED CONVERSION e
PRODUCT IS 5 TORNADO HAZARD CURVES WITH MEAN CURVE EQUAL TO USNRC CURVE Jock W. Benjamin & Associates,Inc.
Consulting Engineers A
RISK OUANTIFICATION INPUT TORNADO AND STRAIGHT WIND HAZARD CURVES FRAGILITY DATA PROBABILITY OF MISSILE IMPACT SYSTEMS MODEL ANALYSIS STEPS / PRODUCTS STEP:
FOR EACH SEQUENCE, COMBINE THE COMPONENT FRAGILITY VALUES AND MISSILE IMPACT PROB-ABILITIES ACCORDING TO THE "YSTEMS LOGIC, AT EACH WINDSPEED VALUE PRODUCT:
CONDITIONAL FRACTION OF SYSTEM FAILURE DEFINED IN TERMS OF WINDSPEED STEP:
COMBINE WIND HAZARD DATA AND SYSTEM FRAGILITY CURVE PRODUCT:
FREQUENCY OF CORE DAMAGE UNCERTAINTY OHALYSIS PROVIDES THE PROBABILITY DISTRIBUTION ON THE FREQUENCY OF CORE DAMAGE Jack R. Benjamin & Associates,Inc.
I Consulting Engineers m
/
h
SCHEDULE PROBLEMS PLANT VISITS (9 SITES) HAVE TAKEN LONGER TO ARRANGE THAN EST!. riED M
INDIVIDUAL PLANT FAULT TREE AND EVENT TREE ANALYSES ARE REQUIRING MORE DETA (SUPPORT SYSTEMS, CONTAINMENT SYSTEMS, RECOVERY ACTIONS) AND'TAKING LONGER TO COMPLETE (BASED ON UPDATED PLANT INFORMATION) THAN ORIGINALLY ESTIMATED INFORMATION REQUIRED TO PERFORM AN INTEGRATED SPECIAL EMERGENCY EVALUATION (FIRE, FLOOD, SEISMIC AND SABOTAGE) HAS INCREASED SIGNIFICANTLY AND HAS BEEN DIFFICULT TO GET FROM LICENSEES TIME FOR SANDIA TO PLACE SUPPORTING SUBCONTRACTS (BASED ON COMPETITIVE BIDS) H TAKEN LONGER THAN ANTICIPATED STAFF REVIEW 0F INTERIM MILESTONE REPORTS HAS NOT BEEN TIMELY s,
s ADDITIONAL COMMISSION, ACRS, AND STAFF COMMENTS AND CONCERNS HAVE ADDED TO THE Os d[.I SCOPE OF THE A-45 PROGRAM s
PROGRAM OPTIONS
- COMPLETE PROGRAM AS STRUCTURED WITil SLIP OF ABOUT SIX MONTHS
- SIGNIFICNATLY REDUCE NUMBER OF PLANTS ANALYZED WITH SCHEDULE SLIP OF ONE TO TWO MONTHS
- REDUCE NUMBER OF PLANTS TO FIVE AND DO ANALYSES AND ASSESSMENT OF ADD-ON DEDICATED SYSTEM ONLY
- COMPLETE ANALYSIS ON AT LEAST TWO PLANTS BY FEBRUARY AND COMPLETE BALANCE ON ADJUSTED SCHEDULE
.x N
eb M S
4 e
a _
u >-
NW =
3W m
W U M W
= < >
=
.J b J 4 E 4 M
> w 3 N
M>
J 4z4 a4 m
w3 C
W -
E ua ap
.J
_a W
2 I4 j a
- =
W 4W 2
UW C
E W M 3
- E O
E~
l n
W a 2
> a Z
$e M
I
)
aCm am 4 -
<=W 3 e4
== W to k >
U 49 >
I W
.J W E J W 4
& W 4 i
a55 lF/6. al _
m
- W
,e.
F e
Q.z N
5 a
w h
G-M a
a w
W x
O 8
E o-a g
E g
a ws w
.a e
o h
m E
n e
g E*
E a
g W
E W
=
E m
m O
5
=s m
E VJ g
E g
~
h k
w w
ME g
E i
I' 5
E if
(
5 w
W g
5 g
o t
E a
=
m
=
r r
Y o
g
=
o u
x w
u-z g
x 5
E g
m a.
2
=
r E
M w
H E
r r
z b
k z
e 5
H W<
w C
o x
E a
g w
W-b a
W"m W
3 5
w Y
wMm 5
W E<
==
a W
W M
M
(
g
>M w
W n.
a a 5 g
a E
a W
w W
E 0~
r 3
E w
a w
B
=
o e
w e
d
=
o w
W o
U 9
~
a
- }w gg
=
8 9
g 9
'*'**%m.
v-
~ =.
[.
~
l..
l' UAD CITIES
~
SUMMARY
OF PREUMINARY QRES j,l l
TO ENCES DOMINANT PRIOR i..
4 :-
1NTERNAL EVENT SEQU ER RECOVERY
- l RECOVERY QUENCES DOMINANT AFT
- 11 POMER AND ECl
- 5 INTERNAL EVP.NT SEWITH LOSS O
~
S MOST REMOVAL
- SEQUENCESFAILURE WITHIN 4 HOURS O
- i. *
- SEQUENCEG WITH LOS(1W) VVERE IMPORTANT l
r f
i m
e
-x
'x -
3 l
1 i..
i.
I I
4
SUMMARY
(CONTINUED)
(
p.0
- QUAD CITIES HAS GOOD PHYSICAL REDUNDANCY IN DHR l..,
e '2 SYSTFMS 1.1 I.':
l
- ELECTRICAL FAILURES DOMINATE INTERNAL EVENT ANALYSIS
?:
I
~
a
- 3 DIESELS PERAUNITS f
1 DC POWER TRAIN PER UNIT
- LOSS OF OFFSITE POWER SEQUENCES CAN AFFECT BOTH UNITS t.
1 i
- WIND-RELATED ACCIDENTS DOMINATE SPECl AL E!VERGENCIES 1
i-i i
1' i
i i
i
\\
hsDdl
---