ML20128L887
| ML20128L887 | |
| Person / Time | |
|---|---|
| Site: | 05000601 |
| Issue date: | 06/28/1985 |
| From: | WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML19304B194 | List: |
| References | |
| NUDOCS 8507110481 | |
| Download: ML20128L887 (50) | |
Text
__ . - - . - -
O 4.0 COREMELT QUANTIFICATION
, In this section, the plant coremelt frequency is quantified based on the
, modeling and data of Sections 1.0, 2.0, and 3.0. The plant coremelt category frequencies are individually quantified to provide in'put for .the plant risk analysis of Section 7.0. The coremelt is also broken down by contributions from initiating events and from support states. The dominant accident sequences are identified.
O 4.1 QUANTIFICATION OF EVENT TREE NODES In this section, the support state probabilities and the event tree node probabilities are quantified.
4.1.1 QUANTIFICATION OF SUPPORT STATE PROBABILITIES The support state model is given in Section 2.0. In this section, the support state probabilities are quantified for the four cases:
- 1. Transient Case (TRA)
This case applies to the transient initiating event tree.
- 2. Loss of Offsite Power Case (LSP) B i
This case applies to the loss of offsite power event tree.
- 3. LOCA Case (LCA)
This case applies to the following initiating events.
- 1. Steam Generator Tube Rupture l
- 2. Secondary Side Break
- 3. Small LOCA
- 4. Large LOCA
!O i
M APWR-PSS 4-1 June,1985 l
8507110481 850620 l PDR ADOCK 05000601 K PDR
l l
1
- 5. ATWS
- 6. Interfacing Systems LOCA
- 7. Vessel Failure .
- 4. Loss of Auxiliary Cooling Case (LC1) ~
This case applies to the Total Loss of Auxiliary Cooling event.
The data used to quantify the support state probabilities is shown in Table 4 .1 -1.
In support state quantification, the following modeling assumptions were made:
l 2. The recovery probability of 1 diesel generator within 40 minutes is taken as 30%. This is consistent with Zion Probabilistic Safety Study. The failure to recover both diesel generators is calculated as q
2 = (0.7)2= 0.49 The failure to recover only 1 diesel generator is calculated as q) = (0.3) (0.7) + (0.7) (0.3) = 0.42
- 3. Loss of offsite power probability is calculated as q = U = 3.3 x 10-4 365 for a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time. 0.12 is the loss of offsite power initiating event frequency.
O 4. Offsite recovery in 40 minutes is taken as 0.5. This is consistent with EPRI NP-2301.
O W APWR-PSS 4-2 June, 1985 7896Q:1D
CD S. e,on loss of SWS,CCW. the ioss of diesel c.oiing ieaeing to diesei generator failure is estimated as 0.5; this assumes that 40% of the SWS/CCW failures occur in the Service Water System, which cools the emergency diesel generators.
- 6. Loss of On-Site Emergency AC Power The loss of on-site emergency power probabilities are taken from Section 3.1. They are "both buses " E 3 )
N any one bus " E 3
- 7. Loss of IPS Signal The IPS failure probability is taken from Section 3.2. Since this is a highly redundant system with cross-connects for signal generation, if it
> fails, it is assumed that the failure will be total. Thus the failure probability of the system is q=[ ] (a,c) as estimated in Section 3.2. This failure is dominated by loss of Vital DC power. Note that this failure does not include the failure to trip the reactor. ATWS is treated as an initiating event.
- 8. Loss of SWS/CCWS Cooling The system failure probabilities are taken from Section 3.3.
The failure of the both trains is given for two cases:
{
O M APWR-PSS 4-3 June, 1985 l 7896Q:10
- 1) AC power available (a c) q=[ ] for transients (c,c) =[ ] for non-transients O ii) AC power is lost, but on-site emergency power is availab'le:
(a,c) q=[ ] for transients Ic.c) =[ ] for non-transients One train failure is also taken from Section 3.3. It is conservatively modeled that offsite power is initially lost so that SW pumps of the train must start. The model probability for failure of any train is calculated as (c,e) '4 " [ ]
The input data used for support state probability quantification is shown in Table 4.1-1. The calculated support state probabilities are given by Table 4.1-2. The major contributions to support states are listed in Table 4.1-3.
l l
O O -
M APWR-PSS 4-4 June,1985 7896Q:10
i i
TABLE 4.1-1 DATA USED IN SUPPORT STATE PROBABILITY QUANTIFICATION N00E CONDITIONAL FAILURE PROBABILITY TRA LSP LCA LC1
-4 -4 -4 0FP None 3.3x10 1.0 3.3x10 3.3x10 ONP Both Buses Fail (a,c)
One of Two Fails OFR None 0.5 0.5 1.0 0.5 ONR Both DGs Fail 0.49 0.49 1.0 0.49 One of Two Fails 0.42 0.42 1.0 0.42 One of One Fails 0.21 0.21 1.0 0.21 SIG None [ ] (a,c)
S/C Both Trains Fail (a,c)
Offsite AC Available Both Trains Fail Offsite AC Failed One of Two Fails ,
One of One Fails l _
oGC None 0.5 0.5 0.5 0.5
- O M APWR-PSS 4-5 June, 1985 j 78960:10
_ .. . . - _ _ . . . . _ _ _ _ _ . . _ _ _ _ _ . . _ _ _ _ _. _ - . . _ _ . . _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ . _ _ _ - _ . _ . - . = . - - - _ _
i i
4 I
TABLE 4.1-2 1
, SUPPORT STATE PROBABILITIES SUPPORT STATES CASES !
I TRA LSP LCA ~ LC1 552 (a,c)
SS1 i
L SS12 :
SS11
- $50 S502 SS01 j SS00 i
i l i i
1 1
I, 1
i I
l' M APWR-PSS 4-6 June,1985 !
7896Q:10 .
[
r
,-,,--,--,,.----.=.,----,w.-r,-.,n,,,---,-,,--, *
~ , - _ - - , , . . - - . - - . - - , - . - - - . - - - -
l l
l l
i O TABLE 4.1-3 l
MAJOR CONTRIBUTORS TO SUPPORT STATES !
Support Transients: Probability State (a,c)
-Loss of offsite power: ._
(a,c)
_LOCAs: -
(a.c)
Loss of auxiliary cooling:
~ ~
(a,c)
PERCENT CONTRIBUTION OF SUPPORT SYSTEMS TO SUPPORT STATE 0:
Ca '.e Percent Contribution (Probabilitvl Onsite AC Power SW/CCW IPS TRA [ ] (a c)
- 1 (a,c)
LSP [ ] (a,c)
[ ] (a.c)
O[ LCA [ ]
]
(a,c)
(a,c)
LC1 -
100% -
lO l
i W APWR-PSS 4-7 June,1985 7896Q:10
O 4.1. 2 QUANTIFICATION OF EVENT TREE NODES The event tree node probabilities are quantified in this section for use in plant coremelt quantification and dominant accident sequence analysis.
4.1. 2.1 INITIATING EVENT FREQUENCIES The initiating event frequencies are calculated in Section 1.2 and are taken f rom Table 1.2.1.
4.1.2.2 SUPPORT STATE PROBABILITIES The support state probabilities are calculated in Section 4.1.2.1 and are given in Table 4.1-4.
4.1.2.3 OTHER EVENT TREE NODES The other event tree nodes are quantified as follows:
SECONDARY COOLING N0 DES:
The failure probability of secondary cooling nodes for SC1, SC2, and SC3 are taken directly from Table 3.7.2-2 for support states 2,1 and 0:
SC1:
42= (a,c) q) =
4 0"_ _
SC2:
q2 = [ ] (a,c)
M APWR-PSS 4-8 June,1985 7896Q:10
1 WESTINGHOUSE PROPRIETARY CLASS 2 4 (C,0) qi =
90 " _.
SC3:
92" O* c) 90"_ _
The SC4 node takes credit for the startup feedwater system (Section 3.7.1).
In this case, only partial credit will be taken for the SUFW systen for support state 2. The nodal probability will be calculated as 42"42 (SC1) x 0.1 l Where only 0.1 credit is taken for SUFW. No credit will be taken for SUFW for support states 1 and O.
(c 0) 42" l
l 43-9o " _. -
PRIMARY COOLING NODES: i The failure probabilities for primary cooling nodes are taken from Tables -
3 l
O ac.4-1 and 3.4-2 and modified for small LOCA cases to account for the operator tion (ORH) of depressurizing the primary system through PORVs and starting the RHR pumps after the unlikely event of conson cause failure of all fou- SI
, pumps. The ORH probability is taken from Table 3.11-1.
i
!O 1
l W APWR-PSS 4-3 June, 1985 78960:10 '
S11 : I i
i (a,c) 42*
qi = -
i 1 q = !
O _
SI2:- These failure probabilities are directly taken from Table 3.4-2.
42" (a,c) 41" !
q =
o _
! CONTAINMENT SPRAY NODES: !
The failure probabilities are taken from Section 3.5.3.1 as:
42" (***)
l -
gi =
- q =
o After total loss of AC power and coremelt, recovery of CSP is taken as !
q, = 0.5 CONTAINMENT FAN COOLER NODES: !
i l This failure does not include effects of degraded environment after coremelt.
l l
t Such effects are to be included in the containment event tree whenever I l applicable. The failure probabilities are taken from Table 3.6-3 as follows:
M APWR-PSS 4-10 June, 1985 7896Q:10
O 1) No loss of offsite power (c.o) q2 "
7.
q -
0 - _
ii) After loss of offsite power -
l l (c,0) 42" -
41" '
4o "L '-
LONG TERM COOLING NODES: _
Credit is taken for possible switchover to RHR pumps in the form of operator action ORH (Table 3.11-1). The probabilities are taken from Table 3.2-1 for failure of SI pump trains; (c,0) 1) Small LOCA q2 "
CFC available qg=
O s- -
ii) Small LOCA No CFC available e#) >
42" No " ._ --
W APWR-PSS 4-11 , ' June, 1985 7896Q:10 b I
l l
l l
O iii) Large LOCA q2 (a,c)
CFC available q, =
l go"_ a l
iv) Large LOCA q2" (*' )
No CFC available O 4) =
l.
=
4 o -
SLL: SEAL LOCA NODE:
The seal LOCA probability is calculated as a product of loss of BSI system probability and the occurrence of seal LOCA (after loss of BSI) probability.
BSI failure probability is taken from Table 3.8-2 as O AgSt = ( ) (a,c)
- 7. -
The probability of an appreciable seal LOCA following BSI failure is assigned as 3
4 seal " E (a.c)
Then the nodal probability is calculated as O q, =
(a,c)
LCO: CONSEQUENTIAL LOCA N0DE:
O This node is assigned a probability of 92* (a,c)
O Al"_ _
W APWR-PSS 4-12 June,1985
The q probability is assigned based on the discussion in Section 1.2.5:
2 a,0) the consequential LOCA adds [ ] to the small LOCA initiating event frequency. The probability of q) is conservatively taken to be an order of magnitude higher than q to account for additional challenges due to being 2
in a degraded support state. -
REC: RECOVERY OF AC/IPS This probability represents either the recovery of AC power or start-up of pumps by operators (in case only IPS is lost) after 40 minutes after a transient event and before coremelt. It is assigned a value of 0.5 based on time available before coremelt in providing opportunity to recover AC power (on-site or offsite) or manual start of pumps after loss of automatic signal.
ACR: RECOVERY OF POWER AFTER 40 MINUTES:
This probability represents power recovery after 40 minutes into the . accident and before coremelt. It is assigned a value of 0.5 based on availability of 2 to 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before coremelt after initiation of a small seal LOCA. Potentially the accumulators and CRTs may also be used to replenish water inventory in the vessel, thus extending this time further.
4 q = 0.5 3
ACC: ACCUMULATOR N00E:
The accumulator failure probability is taken from Table 3.4-8 as (c.o) q=[ ]
for a 2/3 success criteria.
W APWR-PSS 4-13 June,1985 78960:10
SOF: STEAM GENERATOR OVERFILL NODE:
The failure probability of the automatic SG drain feature (SOF) is calculated as O 4=[ ] (a c) in Section 3.9.3.1. If this automatic capability fails, the SG is assumed to O overfill and a SG safety is assumed to fail in an open position. Core melt can be prevented if the operator depressurizes the RCS by opening the pressurizer PORVs. thus lowering primary pressure. The probability of such an operator action is taken from Table 3.11-1 (OBL) as q = 0.01 Thus the nodal failure probability is calculated as:
42 I")
=
q) .-
4 =
0 -
PRR: PRESSURE RELIEF IN ATWS:
From previous work, the failure of this node is estimated to be q = 5.0 x 10
-3 This includes f ailure of relief and safety valves in the primary system to open to mitigate the pressure increases during early stages of an ATWS event.
CON: INTERFACING SYSTEMS LOCA OCCURS IN CONTAINMENT:
O W APWR-PSS 4-14 June,1985 7896Q:10
The failure of the return path to the EWST in case of the most credible interfacing systems LOCA event (RHR suction path) is estimated to be o) q=[ ]
This estimation is carried out as follows:
q= MOV along the + Pipe or RHR EWST path is closed Pump Seal Break The first contribution is dominated by misposition after maintenance (c,o) [ j. The second contribution is taken to be the same magnitude as the first; thus the total failure is:
(c,o) q=[ ]
OPERATOR ACTIONS:
The operator action failure probabilities are taken from Table 3.11-1 as
-3 OFB q2 = 5 x 10 gj = 0.01 q,= N/A OST q = 0.01 q) = 0.01 q, = 0.1 2
ORT q = 0.01 q) = 0.01 q, = 0.01 2
OLT q2 = 0.01 gj = 0.01 q,= N/A O
M APWR-PSS 4-15 June,1985 7896Q:1D
O- TABLE 4.1-4 EVENT TREE NODE QUANTIFICATION HQDE CONDITIONAL SSI SS2 SS3 0 S01 None
(***}
S02 503 504 SOS S06 S07 S08 509 S10 SC1 None SC2 O SC3 SC4 SIl None SI2 ACC CSP None Small Loca l CFC No loss of offsite power After loss of offsite power LTC Small LOCA with CFC ;
Small LOCA no CFC i Large LOCA with CFC Large LOCA no CFC O ~
W APWR-PSS 4-16 June,1985 7896Q:1D I
1
1 i
4 d
TABLE 4.1-4 (Cont)
, EVENT TREE NODE QUANTIFICATION
- .NQRL CONDITIONAL SS1S1 553
-3 -2 f 0FB 5 x 10 1 x 10 _
! OST 1 x 10 -2 1 x 10 -2 1 x 10 ~I
~
ORT 1 x 10 1 x 10~ 1 x 10~ OLT 1 x 10 -2 1 x 10 -2 , LCO (a,c) SLL ACR - SOF PRR CON I i' t. j O n ~;l) M APWR-PSS 4-17 June, 1985 7896Q:10
O 4.2 QUANTIFICATION OF COREHELT FREQUENCY j The coremelt frequency is quantified by using the initiating event frequencies of Section 1.0, event trees of Section 2.0, and the nodal probabilities of Section 4.1. The results are tabulated by , I
- 1. Coremelt Categories
- 2. Initiating Events
- 3. Support States i
in Table 4 . 2-1. Note that this table also contains the conditional probability of coremelt, given the occurrence of the initiating event. This l conditional probability is a measure of the plant's ability to withstand a given initiating event. O O O l O W APWR-PSS 4-18 June,1985 7896Q:1D
l I. . ,
- i i.
I q , 't ! I i- I u n !:. TA8LE 4.2-1 l 4
-PLANT COREMELT FREQUENCY AND CONTRIBJTORS 'I
(
. COREMELT FREQUDICf BY CORDIELT STATE :
- t. -
(a.c) j i._ ! I i e e t l I i 6 i l l i i i _. I~ : i i . l9 i-
\
l . 1 l l 1 j I i 9 r i-i I i r- [- M APtNt-PSS 4-19 June,1985 799M):18 t
O 4.3 ANALYSIS OF COREMELT CONTRIBUTORS t The dominant accident sequences contributing to coremelt are listed in Table I 4.3-1. This table also shows the failed event tree nodes in, each event 0 Q sequence. The percent contribution of the top 20 event sequences to the total coremelt is given by Table 4.3-2. By examining Table 4.2-3, one notes that the coremelt risk is driven by [ (a,c)
}
An importance ranking analysis of the major event tree nodes is also presented. The importance (IMP (X)) of a system (X) is defined to be N IMP (X) = I q g (X)/q T i=1 where N = Number of dominant sequences; qq(X) = Frequency of the i sequence if it contains the failure of the system X; qg(X) = 0 if the i th sequence does not contain the failure of the system X; qT
= Total plant coremelt frequency Table 4.3-3 ranks the systems and operator actions by their importance.
O f O W APWR-PSS 4-20 . lune, 1985 7896Q:10
( !
! i . f i
t 1 i- - i t TABLE 4.3-1 I
- DOMINANT ACCIDENT SEQUENCES I FRE%ECY PEEEhi C3EMELT EVENT TREE SEQ SUF Ft.lLEt
- no STATE RCOEA
. .. . . - - - - .- -- j j - - ,
i - (a c) ; i t t i d .i 5 I L9 1' ] i l i i 1- j i i 1 i ; t i j i i +1 I 1 , 1 1- i i i i i 1 I
~
i. 1 I 1 i i i, 4 1 1 i ~ i j g APWR-PSS 4-21 . lune
- 1985 i 1
7094:10 i'
j. i i i i l TABLE 4.3-2 PERCENT CONTRIBUTION OF DOMINANT ACCIDENT SEQUENCES i 4 i 4 Event Secuence 5 Contribution . m ! (a,c) i i 1 i. d I i f i i i i l 4 1 1 i l i i 2 l 1 1 1 t i i j i i i i . -- _ i
! M APWR-PSS 4-22 June,1985 18960:10 t_.._,._.,-,,-..____
. . . . . . . _ . - . . - . - . . _ - . . - - . . - - - . - . . _ _ - - - . . _ _ - . - - . ~ . - - _ _ _ = . _ -
f 1 4 f i TABLE 4.3-2 (cont.) l i- ' 4 PERCENT CONTRIBUTION OF DOMINANT ACCIDENT SEQUENCES 1 \ Event Secuence 5 Contribution ) ! (a.c) l. 1 l i i 1 r I i i ) i i '! l Y ll . I i i i i i 4 i l 1 i, .. 1 i a W APWR-PSS 4-23 , June,1985 , 7896Q:10 i i 1
1-. ! i 4 i i ! t i f l_ TABLE 4.3-2 (cont.)
- PERCENT CONTRIBUTION OF DOMINANT ACCIDENT SEQUENCES
; Event Seouence 5 Contribution -
1 - - 4
- (a.c) t i
i l i i i i' ! i ! i ., O i 4 i 4 1 .i i 1 r 5 1 l i W APWR-PSS 4-23 June, 1985 i i i 18960:1D I 1 l
O TABLE 4.3-2 (cont.) PERCENT CONTRIBUTION OF DOMINANT ACCIDENT SEQUENCES Event Seauence 5 Contribution O- . (a,c) l l 0 t TOTAL , [98.6] (a e) O l i O P O f I W APWR-PSS 4-24 June, 1985 18960:10 l
1 J 4 i i i i TABLE 4.3-3 j IMPORTANCE RANKING OF EVENT N0 DES i s ! System Description (Event Tree Node) Importance h- - (a.c) , i 1 I i 1 1 i i , t
- i i
i l t t I. {' t i 1' i i ! ! l ! I
- t i
i 4 1 6 i i i 1 1, i l i > lO i } l i a 1 I I i ( 4 j l I i ! 4
- 3 i
W APWR-PSS
- 4-25 June, 1985 I
4 1896Q:10 l
}
i r __..,c
l l l l
, 4.4 SENSITIVITY OF PLANT COREMELT FREQUENCY TO SYSTEM RELIABILITIES T'his section presents an analysis in which several event tree node probabilities are changed to assess the sensitivity of the plant coremelt O frequency to system reliabilities.
compared with the WAPWR base case results. Eight cases are studied below and are Some of these cases attempt to simulate ef fects of new systems on plant coremelt frequency. Note that these cases are strictly for sensitivity analysis purposes. They are not backed-up by detailed analysis of the implied system changes. The results of these cases are sununarized in Table 4.4-1. CASE 1 4 DGS
; In this case, the failure probability of both main emergency bases is decreased by an order of magnitude:
q=[ ] for DNP node in support state event tree. (a,c) This case simulates a four-diesel on-site power design. As expected from the system importance analysis, the plant coremelt frequency is cut in half as a l result of this change. The sununary of the analysis is given by Tables 4.4-2 i and 4.4-3. CASE 2 2 BSI PUMPS In this case, the failure probability of the back-up seal injection system is decreased by an order of negnitude: thus the event tree node SS2 has q.[ (a,c)
]
! This case simulates a 2-Pump BSI system. The results of the analysis are summarized in Tables 4.4-4 and 4.4-5. The plant coremelt frequency is cut in half as expected from the system importance analysis. W APWR-PSS 4-26 June,1985 7896Q:10
O CASE 3 PASSIVE STEAM CONDENSER l In this case, the failure probability of the secondary cooling nodes for support state 0 are reduced by an order of magnitude. This case simulates the passive steam condenser design for EFWS. The plant coremelt frequency is (c e) reduced by [ ]%. The nodal probabilities used are c) q, = for SC1; q = for SC2; q = for SC3; q o
= for SC4.
The results are sununarized in Tables 4.4-6 and 4.4-7. CASE 4 BSI NOT PRESENT In this case, the BSI system is removed f rom the design to assess its impact on plant coremelt frequency. The event tree node SLL has the failure probability of J q,= 0.5 (c,s) The plant coremelt frequency increases by [ ] fold. The results are sununarized in Tables 4.4-8 and 4.4-9. CASE 5 NO AUTOMATIC SOF SYSTEM In this case, the automatic steam generator overfill protection system is removed from the design. The event tree node is now driven by the operator action of feed and bleed and has the probability i O W APWR-PSS 4-27 June,1985 7896Q:10
l i l O 4 - 4) = 0.01 4, = 1.0 0 The resulting plant coremelt frequency has [ ). Th'e results are (a c) summarized in Tables 4.4-10 and 4.4-11. CASE 6 LESS RELIABLE ECCS In this case, the failure probabilities of event tree nodes involving ACC, SI and LTC are increased to simulate a less reliable (less redundant) ECCS. The accumulator failure probability is increased by an order of magnitude to simulate a 3/3 success criteria; the short term cooling (SI) node failure probability is also increased by an order of magnitude. The long term cooling node (LTC) probabilities are increased by three orders of magnitude (for ' support state 2) to simulate a no EWST case where switchover to recirculation is needed and only two SI pumps are present for this purpose. The failure probabilities used are given below: ACC: 4 = (- ) (a,c) SIl-g,,,) 42" 4j= 40" - SI2: __ 42" C*) Aj=
=
4 0 _ .,_ M APWR-PSS 4-28 June, 1985 7896Q:10 e --w -e - - , - - ,-
l V LTC: q 4 4 2 1 0
-4 Small LOCA with CFC 1.1 x 10 1.1 x 10 -3 1.0
~4 -3 Small LOCA no CFC 2.7 x 10 1.6 x 10 . 1.0 D -4 -3 Large LOCA with CFC 1.3 x 10 1.5 x 10 -
1.0
-4 -3 Large LOCA no CFC 2.8 x 10 3.1 x 10 1.0 The plant coremelt f requency becomes [2.5 x 10-6] af ter this change. The
results are sununarized in Tables 4.4-12 and 4.4-13.
A more detailed analysis of Long Term Cooling in a standard four-loop Westinghouse PWR yields an unavailability of LTC for Small LOCA with CFC of c,0) roughly [ ] and LTC For Large LOCA with CFC of about [ ). The unavailability of these alternative system designs is impacted by piggy-back operation (RHR to HHSI) and by the existence of two RHR trains of equipment. Analysis of such a system in the APWR yields a core melt frequency of about c) [' ]. This frequency is dominated by failure of Long Term Cooling following either Small LOCA or transient with consequential LOCA. CASE 7 LESS RELIABLE OPERATOR ACTIONS In this case, the sensitivity of the plant coremelt frequency to operator actions (in event tree nodes) is studied. The failure probabilities of event tree nodes containing operator actions (such as OA, SOF, LTC and SI1) are increased by a factor of 3. This is the error factor usually associated with
~4 HEPs in NUREG-1278 for probabilities in the range q > 10 . The plant hV coremelt frequency is not appreciably af fected by this change. The results are sunenarized in Tables 4.4-14 and 4.4-15.
CASE 8 CONVENTIONAL M PWR DESIGN O V In this case the following changes to the event tree node probabilities are made to simulate a conventional M PWR design: O W APWR-PSS 4-29 June, 1985 7896Q:10
O 1. BSI is not present;
- 2. Automatic Steam Generator Overfill Protection System is not present;
- 3. Startup feedwater system is not present; *
- 4. Interfacing systems LOCA occurs outside the containment.
O 5. ECCS failure probabilities are increased as in Case 6 above for ACC, SI and LTC nodes.
- 6. Secondary cooling failure probabilities are increased by an order of magnitude.
The plant coremelt frequency becomes [ ]/ year; this value is (a,c) consistent with coremelt frequencies obtained in recent PRA's for conventional M PWR plants. The analysis is sununarized by Tables 4.4-16 and 4.4-17. I l l O O i i i O 4-30 June, 1985 W APWR-PSS 7896Q:10 l t__ _ _ , _ _ _ . . , _ . . . . _ _ _ _ . _ , _ _ _ _ .
. - . _ - . . . --.- - _ _ - . _ -. . - - . . . _ _ _ _ . ~ .-
l i l 4.4.1 SENSITIVITY ANALYSIS
SUMMARY
l Plant Coremelt ' , .C.411 Frecuency i ._ WAPWR (BASE) CASE (a,c) l CASE 1: 4 DGS O CASE 2: 2 BSI PUMPS CASE 3: PASSIVE STEAM CONDENSER CASE 4: NO BSI SYSTEM CASE 5: NO SOF SYSTEM CASE 6: LESS RELIABLE ECCS
- O CASE 7
- OPERATOR ACTION FAILURES 3 TIMES CASE 8: CONVENTIONAL WPWR DESIGN i
{ O lO
- W APWR-PSS 4-31 June, 1985 ,
! 7896Q:10
i l l TA8LE 4.4-2 i ACCIDENT SEQUENCES FOR CASE 1 FREQUEh:Y PERCENT CORENELT EVENT TREE SEE SUP FAILED No STATE NODES (a.c) 1 d
-r e i
I l h i M APWR-PSS 4-32 June, 1985 78960:10 ; h
_ _ _ . ._ _ _ _ _ _ _ _ _ . _ . _ . _ . . _ _ _ _ _ . _ . . . _ . _ _ _ _ - - _ _ _ _ _ _ _ - _ _ - - - - . . - - - = - e l 1 1 TABLE 4.4-3 ' 1 COREMELT CONTRIBUTORS FOR CASE 1 l COREMELT FREOUEhCY BY COREMELT STATE , (a,c) l I i e, i ! r ? i . ! l
+
i 6 i 1, - t i a I i f 1
> i i
i 1 I !O1 i j - j W APWR-PSS 4-33 June, 19P5 I j 70960:10 i t d
- i
)
I t i ! 1 l
- i j i a !
l 1 TA8LE 4.4-4 l ACCIDENT SEQUENCES FOR CASE 2 i
- FREGl ECf FEK;ENT CDREPE&T EVEhi TREE SE2 SUF FAILE:
NC STATE m::E5 4 l . (a,c) ! t
- r 4 i
- \
t I
- i i
i- . t' 4 i ! t
\
i ; ! I i i i i l l 1 t I i l l 1
- i t
l, i j - i l W APWR-PSS 4-34 June,1985 , l 78960:10 I l'
p I i
@ TABLE 4.4-5 f COREMELT CONTRIBUTORS FOR CASE 2 {
CT EMELT FREGUE C Y B1 COREMELT STATE l (a,c) f i 4 G l i' l I j l i, 1 I I r i 9 t i t
. I ~@
l l W APWR-PSS 4-35 June,1985 , 70%Q:10
<. t , . 1 1 ! a - i i ! + 4 i o TABLE 4.4-6 i
- i ACCIDENT SEQUENCES FOR CASE 3 i FRE;;EN;f PEREENT COREnEt! EVEmiTREE SE3 SUF FAILED I no STATE N0;EE ;
- ... .. . . .. . . . . . . ~ . . . . . . . . . - ,
t ,, i (a,e) ;
\
I s
- - l t t I
t t 4 j t , 4 <> l I ! f l- t
,t c
4 t l l I i , t 1 .I l , t 4 t s
! i i .
t i 1 !@i,
, P
, c t L i ! i9J l
'r'
?
1 1 ,,,,) , i W APtNt-PSS 4-36 June, 1985 { l i 4 7096Q:10 1 i i k
. = -
l' i j I t i 1 , ! l TABLE 4.4-7 , COREMELT CONTRIBUTORS FOR CASE 3 l COREMELT FREGJihCV Bf COREMELT STATE l 4 i.
~ ~
l (a.e) ! t I f 1 ! I i e e f I i l i i i O - i h i W APWR-PSS 4-37 June,1985 7et60:10 l
l
./
A e TABLE 4.4-8 __ ACCIDENT SEQUENCES FOR CASE 4 FREC' jet:t PERCEhi CLREMELT EVENT TREE ' SEE S# ' Fake! NQ $T HE NC:ES
~
(a,e) V
..i e
l O .-
/
I
/
e
- 1
- d-r A
h _ _ M APtNt-PSS ,4-38 June,1985 78960:10 '
/
TABLE 4.4-9 COREMELT CONTRIBUTORS FOR CASE 4 COREMELT FREC Xh:Y Bf CN.. MELT STATE
~
i (a.c) O O O O O M APWR-PSS 4-39 June,1985 78960:10
i t t TA8LE 4.4-10 ACCIDENT SEQUENCES FOR CASE 5 FRIDEk:1 PEACEE CCAEMILT EVENTTAEE SED 55 FARE: 5 NO
.......-.. . .. ST ATE..N00E5 .. ,,,,,
l' (a,e)
, l , i i
i i l l l t l l i l i l : .I 1 i i i ' I i 4 > i
- i h
t i :
~
~
E APWR-PSS 4-40 June,1985 4 7896Q:10 I t 1 ? i l
- _ . . - - - - . - . - _ . . . . - . - - . - . - _ . _ - _ - . . - - _ . . - - . _ _ _ - _ - . . . - - . _ . _=_
l i ! 1 ! l 1 TA8LE 4.4-11 COREMELT CONTRIBUTORS FOR CASE 5 i COREMELT FREMN:T If COREMELT STATE i (a,c) ; r i i k 1 I i i i I , 1 i i j i l l r i l ? I I i I i i J i 1 L l h b M APWR-PSS 4-41 June,1985 78960:10 i t I
i 1 1 I i l' t i 1
; TABLE 4.4-12 l l
- ACCIDENT SEQUENCES FOR CASE 6 l FRE0JEk
- T PEREhi CORIPAT riVENTTREE SEC SP FARED NC STATE BODES j -. ,,,,,,,,
t (a c) 1 4 i 1 i i }! f l l M APWR-PSS 4-42 June,1985 ( 7896Q:10 l l l l . .-... - - .. - , . - - _ - _ ____ _ _ __ _ _ _ _ _ _ _ _ _ _
l \ i l i TABLE 4.4-12 i
- t ACCIDENT SEQUENCES FOR CASE 6 I
t i I - ~ (a,c) j l l i i i @ l i
- l
- 9 i
! t i
i P l i 9 l 4 I
. W,APWR-PSS 4-43 '
June,1985 a j 7896Q:10 1 l
TABLE 4.4-13 3 COPEMELT CONTRIBUTORS FOR CASE 6 C0nEMELT FREG'aECf if ;0REPELT STATE 4 (a,c) , l l l 1 i, i i, 1 l l t l l M APWR-PSS 4-44 June, 1985 i i 78960:1D l l
- l
i TABLE 4.4-14 ACCIDENT SEQUENCES FOR CASE 7 FRE;;Eh;f FERCIh! CDAEni;T EVEhi TREE SE: SUP FA! LED i N: STATE k;;ES
-(a,c) i O
4 l l' i i i d i i M A NR-PSS 4-45 June, 1985 70960:10
i i t TABLE 4.4-15 l COREMELT CONTRIBUTORS FOR CASE 7 CORER.E.T FRE;UEN:f H COREME.T STt.TE
~
i - c..e l 1 . ] l 3 i W APWR-PSS 4-46 June,1985 78960:10 t
=i l
__ . . . . _ _ _ _ . _ __. _ . . _ ._ . ._ _ _ . . . _ _ . _ _ = _ . __ _ __ i I i 1 , I l E TABLE 4.4-16 ACCIDENT SEQUENCES FOR CASE 8 l FEOUEN;Y PE E ENT COREMELT EVENT TE E KO SUP FAILED '
- RD STATE RODES i
L
* (a.c) [
k l 1
. l i
l P t - l l M APWR-PSS 4-47 June,1985 78960:10
i, i TABLE 4.4-16 ACCIDENT SEQUENCES FOR CASE 8 (a,c) e i
- i l
l 1 1 I 1 4 4 - 4
- l t i
i M APWR-PSS 4-48 June, 1985 78960:10 i i
~ e- -..- ~ -,._. - ,_ _ . _ .. % - mm._..,._.~. _ . _ _ . , - , _ . _ _ . _ . _ . _ . . _ . . , ~ . _ _ _ _ . . _ . - - - - _ - -
4 4 i 9 i i TABLE 4.4-17 COREMELT CONTRIBUTORS FOR CASE 8 CCREMELT D EQUEkh BY COREMELT STATE (a,c)
- l. -
i i i . 1 c 4
- M APWR-PSS 4_49 June,1985 78960
- 10
e ! l
\
l l I O- 4.5 CONSERVATISM IN COREMELT STATE CLASSIFICATION The point estimate analysis indicates that the coremelt state with Transient-Early Melt-No Containment Safeguards (TE) is the major' contributor to the plant coremelt frequency. However, a more detailed analy' sis could be performed to point out that some of the event sequences included in the TE corenelt state are actually either late coremelts (TL) or coremelts with containment safeguards available (TEFC or TLFC). The reasons for the above contention are briefly described below:
- 1. For the support states, a 24-hour mission time is used for loss of offsite power and SW/CCW cooling. Thus, failure of a support system during the 24 hour period is treated as if it occurs at the beginning of the event sequence. This leads to a conservative classification of many event sequences as early melts.
l The same argument also applies to secondary cooling where a 24-hour mission time is used. Thus, many potential late failures are classified as early ones. 1
- 2. The coremelt progress and vessel failure in many TE event sequences will take a long time considering the WAPWR's larger vessel, ACC, and CRT volumes which make recovery of support systems realistic. Thus, in some event sequences, the time may be available for at least recovery of containment cooling (fan coolers and/or containment sprays), even if coremelt and vessel failure occurs. Thus, many TE sequences might become TEFC.
O - 0 0 W APWR-PSS 4-50 June,1985 78960:10}}