ML20132B893
Text
_
- y Y.AR 2 9 to?4 T'
MEtiORANDUM FOR:
Thcmas it. Novak, Assistant Director for Licensing FROM:
Frank Schroeder, Assistant Director
.for Generic Projects
SUBJECT:
REVIEW 0F MISSISSIPPI POWER AND LIGHT COMPANY'S PROPOSED STATI0ti BLACK 0UT RISK ASSESSMENT FOR GRAND GULF UNIT 1 4
In response to your memorandum dated liarch 8th, we have reviewed MP&L's
't proposed risk assessment for events involving a loss of AC power at Grand Gulf Unit I.
A written evaluation is attached.
In general, the overall scope of MP&L's risk assessment appears app,ropriate to assess the core melt frequency associated with loss of AC power events.
Staff comments on specific elements of' the a~ssessment, included in the attached evaluation, should be considered by MP&L to assure
' completeness of the assessment.
Results of the analysis should provide useful information and insights as to the risk associated with station blackout events at Grand Gulf Nuclear Station.
Frank Schroeder, Assistant birector for Generic Projects Division of Safety Technology t
Attachment:
As stated O
DISTRIBUTION i
Central Files
/
GIB Rdg p\\
7 GIB SER File
[5[g /yI f}J,sf' /
FSchroeder
?.
7]
./
DST c/f z
KKniel PNorian 0
nliut in
/f0 9.ke,,,c,
0,gy,
, DST,,: g,I B, g,
.,P.h,fG,1,B,,,,,,,,,.f.fg,,,,, G P,,,,,,
Nunuur)>
Ruh.1.a;,ch.....
tj,qn.,,,,,,,,
,KK n.i g,1,...........
..g gr....
....................w 0
... 3 /28/.84
......g/. 8 40 3 /...............
03/g..p/.84 03
/84 em>
s.a: roy m oommacu ano OFFICIAL RECORD COPY uscac i+ei-v..
~
Evaluation of Mississippi Power and Light Company's Proposed Risk Assessment of Station Blackout Events at Grand Gulf Nuclear Station Mississippi Power and Light Company's Proposed Risk Assessment
- Introduction and Scope 1k Mississippi Power and Light (MP&L) has proposed doing a risk assessment for I'
f station blackout at the Grand Gulf Nuclear Station (GGNS).
Station blackout
. is defined as the loss of all offsite and onsite emergency AC power.
The objective of this assessment is to determine ~the frequency of core melt associated with loss of AC power events. The assessment will be based on probability calculations, including event trees fo,r sequences involving total loss of AC power, and analyses of the relia'bility of the.two Transamerics T
DeLaval, Inc. (TDI) emergency diesel generators, offsit'e power reliability, and the capability of GGNS to withstand a station blackout.
~In addition to base calculations of core melt frequencies, studies will be' conducted to determine the sensitivity of the results to Varying a
reliabilities of components of the GGNS AC power system.
In particular, the sensitivity study will determine the required TDI diesel generator
' reliability to meet a numerical goal for core melt frequency.
- Reliability of AC Power Systems The reliability of the onsite emergency standby diesels will be evaluated based on overall industry data and the GGNS history of operation.
Both starting reliability (failure to start) and operating reliability (failure to run) will be considered.
Special emphasis will be placed on the TDI diesels where an assessment of recent changes will be included.
MP&L is. presently installing three gas turbines with a total capacity of 6,200 kilowatts (KW) for additional onsite emergency AC power supply. These gas turbine power sources will also be assessed for relfability based on
. - -. =
...--%~
-s
a.:
2 industry and plant-specific operating data.
Consideration of as turbine life will be included.
4 MP&L has indicated that, if no other AC power sources were available, the diesel generator dedic'ated to the high pressure core spray (HPCS) system, which has a capacity uf 3,300 Kw, could be connected manually to provide certain essential emergency AC power loads that normally would be supplied by l
the TOI diesel generators. A detailed evaluatica of this mcde of operation
.3 is in progress at MP&L.
e The GGNS offsite power system includes the following three power supplies to the nuclear plant:
two 500 kV transmission lines terminating at the switchyard, and one 115 kV line buried for a short distance near the switchyard and connected directly to its respective engineered safety feature transformer.
The basic assessment of the offsite ' power re].iability will
"-:;r consider these three transmission lines, their outage data, the reliability of the power supplied to the transmission system and the effects of external events.
- Station Blackout Capability A major element of the risk assessment by MP&L is an assessment of the capability of GGNS to withstand a station blackout event.
This is ~~
particularly important for assessing component or system repair probabilities and the time available for other operator actions.
To the extent,needed for the risk assessment, a detailed evaluation will be performed of the capability of GGNS to withstand and recover from a station blackout.
The evaluation proposed by MP&L will consider as a minimum the following possible limiting factors:
e g
- i...
~
c/
- containment temperature
- drywell temperature
- suppression pool temperature
- pump seal temperature limits
- instrumentation requirements and associated temperature limits
- control room and remote shutdown panel temp 5ratures
- battery life
- pump room temperatures
~,
. ;1
-f MP&L has done a preliminary evaluation of GGNS's station blackout 1
capability.
Results of this analysis indicate that GGNS could withstand a station blackout in excess of 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> without core damage.
This analysis.
assumed the postulated loss of AC power occurs with the plant operating at 75; power.
~
- Assessment of Core Melt Frequency
~
The calculation of core melt frequency will be based on probability
' calculations using event trees in which a loss of'offsite and onsite emergency AC power is the initiating event.
Results~of the analyses of the-reliability of the offsite and onsite AC power systems, the probable time to restore AC power, and the ability of the plant to cope with a station blackout will be used as inputs to the core melt frequency calculations.
Sensitivity analyses will be conducted to determine variations in the results due to different input assumptions.
- q.
Staff Commentf The overall scope of MP&L's risk assessment for station blackout appears to be appropriate to meet the stated objective.
Upon review, the staff has one major comment related to tne assessment of the plant's capability to withstand a station. blackout and the assumed power level before a postulated station. blackout. MP&L did a preliminary evaluation assuming loss of AC power S
,,n,-.a
-.m
- ~ ~
'sx-
.4
- {.
occurs at 75% power. For the proposed risk assessment, MP&L a s to evaluate the GGNS station blackout capability assuming "a variation of time after shutdown that the station blackout occurs." Once GGNS receives a full power license, it is expected that the plant will be operating at 100% power most of the time. The'refore, the staff recommends that the time GGNS can withstand and recover from a station blackout be determined based on a postulated loss of all AC power occurring while the plant is operating at full power.
- s In addition to the above, the staff has the following comments on specific elements of the proposed risk assessment.
- Reliability of AC Power Systems Onsite Emergency Diesel Generators Two reports, NUREG/CR-0660, " Enhancement of On-Site Emergency Diesel Generator Reliability," and NUREG/CR-2989, " Reliability of Emergency AC Power Systems at Nuclear Power Plants,"'d'iscuss operating history.
'and significant causes of diesel generator failures a't nuclear power.
plants. Although no single subsystem was identified as the dominant contributor to d.iesel failures, a number of failure modes, including hardware and human error, were discussed.
These reports should be considered by MP&L in their reliability study of the GGNS emergency diesel generators.
Gas Turbine AC Power Sources Since MP&L calculated the essential AC loads during a station blackout 'to be 2,931 Kw, if no other AC power sources were available, two out of the three gas turbines would be required to supply this 4
load.
This requiremnt shoulU 3e considered in the AC power reliability analysis. Also, faults or failures in the gas turbine system which could be propagated to cause failure of the normal emergency onsite AC power system should b.e considered in the rel,iability analysis.
m
" v eus s po9 4a ue$4Wemwww m,9%4
.s 6
e-mens-H',
h eb mp6-4 4-4
r
- HPCS Diesel Generator Procedures and training should be developed to manually connect the HPCS diesel to supply essential emergency AC power. This would provide a level of confidence that this mode of operation could be achieved, if included in the reliability analysis as an alternate l
means of providing AC power. Operating in this mode assumes that the steam turbine-driven reactor core isolation cooling (RCIC) system i
), k '
pump is available to maintain necessary coolant inventory.
's
- Offsite Power System In their reliability analysis, MP&L mentioned external events but did not specifically discuss weather-related causes of loss of offsite power. Based on loss-of-offsite-power operating experie'nce, severe 9
weather has been shown to be a potentially signfic' ant contributor to long-duration losses of offsite power.
Therefore, MP&L :hould pay special attention to this factor in their risk assessment, and the effects of external events should include consideration of hurricanes, tornadoes, and snow and ice storms.
~
~
- Common Cause Failures,-
NUREG/CR-2989 indicated that common cause failures can be a significant contributor to the unavailability of onsite emergency AC power systems. However, common cause failures were not mentioned in MP&L's AC power reliability assessment.
The staff recommends that MP&L assess single point vulnerabilities that can cause failures of redundant offsite and onsite AC power sources as part of the AC power system reliability analysis.
Potential common' mode failures in design, operation or maintenance should be identified to determine their contrib.utions to core melt frequency.
..w-e..
-s
',i;,
6-l
/
J'
- Station Blackout Capability-In addition to the items mentioned by MP&L, the staff recommends that the following factors be considered in the assessment of GGNS's ability to withstand a station blackout:
a
- Condensate storage tank capac~ity 5
- Compressed air capacity
. :. - f
- Procedures and training for actions to restore AC power and for
. ).
d]g other operator actions that may be necessary to withstand and recover i-from a station blackout (e.g., load shedding to extend DC battery life)
- Assessment of Core Melt Freqency
.u NUREG/CR-3226, " Station Blackout Accident Analysis'," uses event
'\\
tree / fault tree analysis to evaluate the dominant core damage sequences and their associated frequencies for four " generic" nuclear power
'~
plants. One PWR and three BWRs are modeled,' including a BWR with HPCSL and RCIC. MP&L should review this report for specifit ' accident
.a.
sequences to be included in the assessment of core melt frequency at GGNS as well as other insights on station blackout accident analysis, s
g Summa ry J
The staff believes that results of the risk assessment proposed by MP&L, q
if it includes the staff comments listed above, should provide j
information and useful insights as to the risk associated with station blackout events at.GGNS.
s a
A 4
o 9
4
--.-.p, rex-me me
=m.
=i e r
%,v.=4-=a-
- p ez h
. - g e. s i.es.
wee,=
,e-aw,
,****N