Upload
hoangkhue
View
219
Download
5
Embed Size (px)
Citation preview
LIKELIHOOD THAT SHORT TERM STATION BLACKOUT SCENARIOS LEAD TO A
LARGE AND EARLY RELEASE Donald E. Vanover
September 27, 2017
www.jensenhughes.com 2
PRESENTATION TOPICS
Background Modeling approach Relevant considerations for a Large and
Early Release (LER) Integrated LER likelihood estimate results Conclusions
www.jensenhughes.com 3
LERF IN STSBO BACKGROUND
Large Early Release Frequency (LERF) risk metric used in risk-informed applications Mark I BWRs generally have assumed a high
conditional likelihood of LER in Short Term Station Blackout (STSBO) scenarios • Contribution from liner melt-through of
containment following vessel failure Considerable uncertainty associated with
exact timing and magnitude of release • Potential exists that release is not large
and/or not early
www.jensenhughes.com 4
LERF IN STSBO MODELING APPROACH
Identify key inputs to estimate LER likelihood Relevant considerations
• General Emergency (GE) declaration and Evacuation Time Estimates (ETEs)
• Core melt progression characteristics (e.g., stuck open relief valve, steam line rupture, timing of vessel failure, fission product location)
• Timing of “large” threshold release Provide uncertainty distribution for each key
input • Integrate using Monte Carlo uncertainty
evaluations
www.jensenhughes.com 5
RELEVANT CONSIDERATIONS FOR LERF
General Emergency (GE) declaration and Evacuation Time Estimate (ETEs) • The STSBO scenario will lead to an early
declaration of a GE • ETEs are available from recent emergency
planning study for a variety of conditions Conditions Evacuation
Advisory (Hours)
ETE for 100%
Population out to EPZ
(Hours)
Evacuation Time from Sequence Initiation (Hours)
Good to Poor 1.0 to 1.5 4.0 to 5.0 5.0 to 6.5
www.jensenhughes.com 6
RELEVANT CONSIDERATIONS FOR LERF
Assumed cumulative probability distribution for time to 100% evacuation
5.000 6.5005.0% 0.0%95.0%
0.0
0.2
0.4
0.6
0.8
1.0
4.8
5.0
5.2
5.4
5.6
5.8
6.0
6.2
6.4
6.6
Prob
abili
ty
Hours
Cumulative Distribution Mean for 100% Evacuation
www.jensenhughes.com 7
RELEVANT CONSIDERATIONS FOR LERF
Core melt progression characteristics Two key phenomena have first order impact on
fission product location at time of containment failure • Likelihood of a Stuck Open Relief Valve (SORV) • Likelihood of Steam Line Rupture (SLR)
NRC SOARCA study indicated very high likelihood of SORV (i.e., assumed SORV in base case analysis) • Large number of SRV cycles • Thermal degradation mechanisms
In our analysis, assume 95% likelihood of a SORV during core melt progression
www.jensenhughes.com 8
RELEVANT CONSIDERATIONS FOR LERF
NRC SOARCA study investigated likelihood of SLR (only assumed in sensitivity cases) • Extreme variations to the SORV failure criteria
were needed to obtain conditions that would allow SLR to occur − Prolonged SRV cycling, or − Partial opening of SORV such that additional SRV
cycles are precluded but RPV pressure stays high
In our analysis, assume 10% likelihood of a SLR during core melt progression • This is assumed to be a subset of the SORV
likelihood
www.jensenhughes.com 9
RELEVANT CONSIDERATIONS FOR LERF
Core melt progression characteristics – base case likelihood assumptions
Designator Description Likelihood
SORV Stuck Open Relief Valve Occurs
0.85
SLR Steam Line Rupture Occurs
0.10
HP RPV Remains at High Pressure
0.05
www.jensenhughes.com 10
RELEVANT CONSIDERATIONS FOR LERF
Timing of “large” threshold release SOARCA results from MELCOR for STSBO with
SORV and no injection Results indicate virtually no chance of a large
and early release
www.jensenhughes.com 11
RELEVANT CONSIDERATIONS FOR LERF
However, not all case results are in agreement Event MAAP4.06 MELCOR MAAP5.03 MAAP5.04
H2 Production Begins
36 Min. 55 Min. 50 Min. 42 Min.
SRV Sticks Open 1.8 Hrs. 1.8 Hrs. 1.8 Hrs. 1.1 Hrs.
Vessel Failure 3.8 Hrs. 8.2 Hrs. 8.0 Hrs. 3.6 Hrs.
Drywell Liner Mel-Through
3.9 Hrs. 8.5 Hrs. 8.2 Hrs. 3.8 Hrs.
Iodine or CsI release exceeds 1%
4.0 Hrs. (~4% shortly after that and slowly climbs)
9.7 Hrs. > 24 Hrs. 4.0 Hrs . (~2% shortly after that and slowly climbs
later on) Iodine or CsI release exceeds 10%
> 24 Hrs. 27 Hrs. N/A (at 48 Hrs.)
N/A (at 48 Hrs.)
www.jensenhughes.com 12
RELEVANT CONSIDERATIONS FOR LERF
Timing of “large” threshold release Assign probability distribution based on
uncertainty involved for SORV case
6.45 20.265.0% 5.0%90.0%
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0 5 10 15 20 25
Per H
our F
requ
ency
Hours
Time of Large Release
www.jensenhughes.com 13
RELEVANT CONSIDERATIONS FOR LERF
Timing of “large” threshold release Assign probability distribution based on
uncertainty involved for SLR and HP cases • Much more likely to have ‘large’ release occur
earlier in time
3.457 5.7175.0% 5.0%90.0%
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Per H
our F
requ
ency
Hours
Time of Large Release
www.jensenhughes.com 14
INTEGRATED LER LIKELIHOOD RESULTS
0.00 14.92
1.7% 5.0%93.3%
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
-5 0 5 10 15 20
Per H
our F
requ
ency
Hours
SORV Case: (LR - EV)0.000 0.618
72.5% 5.0%22.5%
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5 0.0
0.5
1.0
1.5
Per H
our F
requ
ency
Hours
SLR Case: (LR - EV)
SORV AND SLR MONTE-CARLO CASE RESULTS
Only 1.7% chance of LER is SORV scenario 73% chance of LER is SLR scenario Similarly, 59% of LER in HP scenario
www.jensenhughes.com 15
INTEGRATED LER LIKELIHOOD RESULTS
OVERALL MONTE-CARLO CASE RESULTS
Considering assigned uncertainty distributions, less than 12% LER likelihood estimated for STSBO scenarios
0.00 14.5811.6% 5.0%83.4%
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
-5 0 5 10 15 20
Per H
our F
requ
ency
Hours
Combined Case: (LR - EV)
www.jensenhughes.com 16
CONCLUSIONS
NRC SOARCA study indicates very high probability of SORV in STSBO scenario and much lower probability of SLR MELCOR and MAAP results both indicate time to
“large” threshold may be significantly delayed or not occur at all in SORV scenarios Time to “large” threshold will occur much sooner
in SLR and HP scenarios Integrating various assumptions leads to <12%
LER likelihood using Monte-Carlo analysis Recommend bounding value of 0.2 be used for
risk-informed applications
www.jensenhughes.com 17
Contact Donald E. Vanover +1 610-431-8260
For More Information Visit www.jensenhughes.com
QUESTIONS?