ML13056A129
| ML13056A129 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 10/01/2008 |
| From: | Office of Information Services |
| To: | |
| References | |
| FOIA/PA-2012-0325 | |
| Download: ML13056A129 (4) | |
Text
-10/5/2012
_,"1
-Secu elat rm, Oconee Flood Issue Jocassee Dam Failure Frequency Division of Risk Assessment 10-1-2008
- U.S.NRC o
rity e Irma Jocassee Pumped Storage Proiect
-1
-10/5/2012 Derivation of the Random Jocassee Frequency Two Catastrophic Failures Derived from NPDP -
National Performance of Dams Program
- Frenchman (Montana)
- Failure in 1952 ft height (some literature shows it as 44-ft)
- Low hazard class
- Skagway (Colorado)
" Failure in 1965 ft height
" Significant hazard class USARC lus fl Scur e1~2 Parsing The Frequency Of Random
("Sunny Day") Failure Modes
- Pros Get higher resolution of failure modes contributing to the catastrophic failure frequency.
Concentrate on specific engineering/operational improvements to reduce overall frequency.
" Cons Cannot statistically justify if the overall catastrophic failure data is sparse.
Method is not well developed relying on the assumption that each failure mode is statistically independent without consideration of common cause.
Uncertainty needs to be addressed for each failure mode which might yield distributions that are too wide to be confidently used.
" Conclusion I'
This could be a viable approach to evaluating dam failure providing there is sufficient data. However, for this case, the scarcity of data makes this method unacceptable.
- =jU Na,,Zst seoni -s~ecuri
ýelate i for týon
-2
-10/5/2012 O~~l gs4 eeoaf teoaftio Reliability vs. Statistical Treatment-of Jocassee Dam Failure
" Reliability Approach Pros Detailed event and fault trees are developed for individual contributors to dam failure.
Evaluate which failure modes and scenarios are most likely.
Cons
- Difficult to properly categorize historical data in fault tree structure.
- Potential overlap and double-counting of data.
Requires large populations of detailed data.
" Statistical Approach Pros
- Develop an simple single frequency of dam failure.
- Incorporation of historical details of dam failure by use of Bayesian update tools.
Cons
- Does not lend itself to evaluate details of the initiating event.
,*U*.NRC: -
e" y
e a
Official Use Only - Security Related Information Comparison of Random Dam Failures by Type 2
3 4
6 7
8 Buttress Dams Over 50 Feet High Arch Dams Over 50 Feet High Concrete Dams Over 50 Feet High Earth Dams Over 50 Feet High Gravity Dams Over 50 Feet High Masonry Dams Over 50 Feet High Muati-Arch Dams Over 60 Feet High Rockfill Dams Over 50 fest high Total Dams over 50 Feet Tall 1.E-03 NRC Developed Random Failure Frequency for Jocassee 1.E-04 Licensee-Developed Random Failure Frequency for Jocassee (Taken From IPEEE Submittal) 1E-05 Me, 1
2 3
4 5
6 7
8 T
Type
ý4RC
'--okal "e-E ýhiy -
curiýAýeld I 5
-3
-10/5/2012
-DflmeftHI-Ye OnbL-SmonW~tvR~9,tmdnk~n A T (%1NRC
ý-Offi"aI On-ýoe6ffefyo 6
I Evaluation
" Uncertainty bands are narrow for rockfill (similar to Jocassee) and earthen dam types.
- Frequencies are in the 10-4 per year range.
- Discussion
(--*ýU.S.NRC 7
-4