ML19011A433

Lecture 6-2 Spatial Dependencies 2019-01-18

ML19011A433
Person / Time
Issue date:	01/16/2019
From:	Office of Nuclear Regulatory Research
To:
Nathan Siu 415-0744
Shared Package
ML19011A416	List: ML19011A417 ML19011A418 ML19011A419 ML19011A420 ML19011A421 ML19011A422 ML19011A423 ML19011A424 ML19011A425 ML19011A426 ML19011A427 ML19011A428 ML19011A429 ML19011A430 ML19011A431 ML19011A432 ML19011A433 ML19011A434 ML19011A435 ML19011A436 ML19011A437 ML19011A438 ML19011A439 ML19011A440 ML19011A441 ML19011A442 ML19011A443 ML19011A444 ML19011A445 ML19011A446 ML19011A447 ML19011A448 ML19011A449 ML19011A450 ML19011A451
References
Download: ML19011A433 (48)

Text

Spatial Hazards and Dependencies Lecture 6-2 1

Overview Key Topics

• Spatial dependencies - concept and potential importance

• General approaches for selected hazards

- Internal fires

- Internal floods

- Seismic events

- External floods 2

Overview Resources

• American Nuclear Society and the Institute of Electrical and Electronics Engineers, PRA Procedures Guide, NUREG/CR-2300, January 1983

• Electric Power Research Institute and U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research, EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities, EPRI 1011989 and NUREG/CR-6850, Electric Power Research Institute (EPRI), Palo Alto, CA and U.S. Nuclear Regulatory Commission, Washington, DC, 2005.

• K.N. Fleming and B. Lydell, Guidelines for Performance of Internal Flooding Probabilistic Risk Assessment, EPRI 1019194, Electric Power Research Institute, Palo Alto, CA, December 2009.

• V.M. Andersen, et al., Seismic Probabilistic Risk Assessment Implementation Guide, EPRI 3002000709, Electric Power Research Institute, Palo Alto, CA, December 2013.

• L. Shaney and D. Miller, Identification of External Hazards for Analysis in Probabilistic Risk Assessment: Update of Report 1022997, EPRI 3002005287, Electric Power Research Institute, Palo Alto, CA, October 2015.

• Subcommittee on Disaster Reduction https://www.sdr.gov/

3

Overview Other References

• Electric Power Research Institute and U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research, Fire Probabilistic Risk Assessment Methods Enhancements: Supplement 1 to NUREG/CR-6850 and EPRI 1011989, EPRI 1019259 and NUREG/CR-6850 Supplement 1, Electric Power Research Institute (EPRI), Palo Alto, CA and U.S. Nuclear Regulatory Commission, Washington, DC, 2009.

• M. Kazarians, N. Siu, and G. Apostolakis, Fire risk analysis for nuclear power plants:

methodological developments and applications, Risk Analysis, 5, 33-51, 1985.

• N. Siu, N. Melly, S. P. Nowlen, and M. Kazarians, Fire Risk Assessment for Nuclear Power Plants, The SFPE Handbook of Fire Protection Engineering, 5th Edition, Springer-Verlag, New York, 2016.

• Siu, N., K. Coyne, and N. Melly, Fire PRA maturity and realism: a technical evaluation, U.S. Nuclear Regulatory Commission, March 2017. (ADAMS ML17089A537)

• U.S. Nuclear Regulatory Commission, Workshop on Probabilistic Flood Hazard Assessment, Rockville, MD, 2013. https://www.nrc.gov/public-involve/public-meetings/meeting-archives/research-wkshps.html 4

Overview Other References (cont.)

• K.N. Fleming, Development of Pipework System Failure Rates: Where Do the Numbers Come From and Why Should We Believe Them?, CRA UK 5th Probabilistic Safety Analysis and Human Factors Assessment Forum, September 17-18, 2014.

• Lydell, B., K.N. Fleming, and J.-F. Roy, Analysis of possible aging trends in the estimation of piping system failure rates for internal flooding PRA, Proceedings of 14th International Conference on Probabilistic Safety Assessment and Management (PSAM 14), Los Angeles, CA, September 16-21, 2018.

• N. Siu, et al., Qualitative PRA insights from operational events, Proceedings of 14th International Conference on Probabilistic Safety Assessment and Management (PSAM 14), Los Angeles, CA, September 16-21, 2018.

5

Concept Some Well-Known Operational Events

• Browns Ferry (1975)

- Candle used to check penetration sealing ignites sealant (polyurethane foam)

- Fire spreads to multiple cable trays in Units 1 and 2

- Fire fighters reluctant to use water on electrical fire; fire burns 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />

- Complicated shutdown using non-safety injection source

• Fukushima Dai-ichi (2011)

- Earthquake trips operating reactors (Units 1-3)

- Subsequent tsunami causes SBO, eventual core melt and release

- Non-operating units (Units 5 and 6) also severely challenged

- Varying challenges (some severe) at other plants (Fukushima Dai-ni, Onagawa, Higashidori, Tokai Dai-ni) 6

Concept Some Other Notable Operational Events

• Gundremmingen (1977) - Cold-weather LOOP led to RCS overfill, flow through safety relief valves, 3m water in containment

• Narora (1983) - 17 hour1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> SBO caused by turbine blade failure, subsequent hydrogen explosion and fire

• Blayais (1999) - multi-unit LOOP and LOSW due to beyond-design basis hazard combination (high winds, wind-driven waves, storm surge, high tide)

• Maanshan (2001) - salt spray caused LOOP; subsequent HEAF led to 2-hour SBO

• Arkansas One (2013) - main generator stator drop caused multi-unit LOOP, auxiliary and turbine building flooding in Unit 2

• St. Lucie (2014) - local intense precipitation flooded auxiliary building through unsealed conduits 7

Concept Dependency => failure Spatial Dependencies events are not independent

• Multiple components and their supporting components (cables, pipes, etc.) can be vulnerable to shared environmental hazards

• Defenses against specific hazards might/might not be effective against others. Examples:

- Fire doors and seals might fail against hydrostatic loads

- Watertight doors designed against hydrostatic loads might not withstand dynamic loadings (e.g., from an incoming tsunami)

• Spatial interactions analysis identifies potentially important locations and combinations of locations (where failure of barriers is possible) 8

Concept Cautions

• Large variations in plant layouts, even for standardized designs (if designers are not thinking of spatial dependencies)

• Natural collection points (e.g., control room, cable spreading room, switchgear rooms, cable vaults, penetration areas) are of special interest

• Important risk contributors can come from detailed layout features (e.g., space between cable trays for redundant divisions, elevations and obstacles affecting likely flooding paths)

A well-documented walkdown is a critical element of internal and external hazards analyses 9

Concept Simplified Plant Layout (Schematic)

N Fuel & Radwaste Building N Containment Containment (Unit 1) (Unit 2)

Cable Spreading Main Control Room Room Main Control Room Auxiliary Building Switchgear Room Safety Pumps Turbine Building Section View Plan View 10

Importance Potential Importance - Old Studies NUREG-1407 11

Importance Potential Importance - IPEEEs 0.40 1.0E-03 1.0E-04 0.30 IPE IPEEE CDF (/yr)

IPEEE 1.0E-05 Fraction 0.20 1.0E-06 0.10 1.0E-07 0.00 1.0E-08 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.0E-08 1.0E-07 1.0E-06 1.0E-05 1.0E-04 1.0E-03 CDF (/ry) IPE CDF (/yr) 12

Importance Recent CDFs: External Hazards Effect All Initiators Internal Events 0.35 0.35 0.30 0.30 BWR BWR PWR PWR 0.25 0.25 Fraction 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0.00 0.00 10-6 10-5 10-4 10-3 10-6 10-5 10-4 10-3 Frequency (/ry) Frequency (/ry) 13

Concept Current Framework

• Internal hazards and external hazards

• Terminology and conventions

- External events => External hazards

- Fire: external event => internal hazard

- Internal flood: release point is within plant (even if ultimate source is outside of the plant)

• Caution: NPP PRA frameworks are plant-centric - hazards are treated as statistically-occurring threats to the plant 14

Concept Example Complexities

• A series of storms deposits an unusually heavy amount of snow in the mountains, which is subsequently melted by unusually warm weather which then leads to unusually high reservoir levels. To prevent dam failures, flood managers decide to open flood gates, causing extensive and extended flooding downstream that surrounds a U.S. NPP. [Intentional human action leads to flooding.]

• Salt spray caused a LOOP at Unit 1 of a 2-unit Taiwanese NPP.

Emergency Diesel Generator (EDG) A started but tripped. Heavy smoke from a high energy arcing fault (HEAF) occurring during plant response prevented access to the switchgear room to align EDG B, resulting in a station blackout. [Model as a LOOP with possible subsequent HEAF, or model as HEAF with possibility of LOOP?]

15

Internal Hazards Notable Internal Hazards Analyses

• Internal Fires

- Long history with NPP PRA

- With regulatory application (Lecture 8-3), strong input from fire protection community

- Performed for many plants

- Can be an important or even dominant risk contributor; analysis realism a major source of debate

• Internal Floods

- Also long history

- Often tied with internal events

- Less controversial than fire 16

Internal Fires Internal Fire PRA

• Cable spreading room analyses: WASH-1400 and General Atomic HTGR PRA (1978)

• Current framework developed after 1975 Browns Ferry fire, used in Zion (1981) and Indian Point (1982) studies (Lecture 8-3).

• Uses information from operational experience, models, and experiments

• Involves fire protection engineering, fire science, PRA as integrator

• Focused on Level 1 PRA (CDF):

- Includes high energy arc faults (HEAF) as well as flames

- Includes fires involving transient as well as in situ combustibles 17

Internal Fires Fire PRA Methodological Framework

• Elements mirror NPP fire protection defense-in-depth

• Basic methodology developed and applied in early 1980s

• Refinements added over time (NUREG/CR-6850)

• Analysis is iterative

• Current work focused on improving data and specific models 18

Internal Fires Fire Frequency Analysis

• Objectives

- Identify and characterize potentially significant fire scenarios

- Estimate scenario frequencies

• Data: historical fire events

• Estimation

- Generic

- Plant-specific 19

Internal Fires Equipment Damage Analysis

• Objectives

- Identify potentially significant combinations of equipment that can be damaged by a fire scenario

- Estimate conditional probabilities of equipment failure modes, given a fire scenario

• Underlying model: competition between damage and suppression processes Damage occurs if tdamage < tsuppression 20

Internal Fires Equipment Damage Analysis Elements 21

Internal Fires Equipment Damage Analysis (cont.)

• Prediction of fire environment

- Correlations

- Zone models

- CFD models

• Equipment response/component fragility

- Temperature and/or heat flux thresholds

- Empirical data and probabilistic models for specific failure modes (e.g., spurious operation, high-energy arc faults)

• Fire suppression

- Historical data

- Fire brigade drills 22

Internal Fires Plant Response Analysis

• Objectives

- Identify potentially significant fire-induced accident scenarios

- Estimate fire-induced core damage frequency (CDF)

• General approach: propagate fire-induced losses through event tree/fault tree model

- Start with internal events model

- Modify to include effects on equipment availability and operator actions 23

Internal Floods Internal Flood PRA

• Includes all wetting mechanisms (including spray, dripping, steam), not just inundation

• Includes floods from external sources (e.g., intake canals, rivers, lakes) that L. Armstrong, Internal Flooding Background, Regulatory Meeting, Internal enter plant through a plant system Flooding Risk Reduction Activities, November 30, 2006. (ADAMS ML063460495)

(e.g., failed expansion joint)

• Analysis approach analogous to treatment of internal fires

- propagation physics simpler

- minor amounts can cause trouble

• Can be an important or event dominant risk contributor 24

Internal Floods Internal Flooding Analysis Process K.N. Fleming and B. Lydell, Guidelines for Performance of Internal Flooding Probabilistic Risk Assessment, EPRI 1019194, Electric Power Research Institute, Palo Alto, CA, December 2009 25

Internal Floods Internal Flooding Frequencies PIPExp Data* Pipe Rupture Model*

• Adapted from K.N. Fleming, Development of pipework system failure rates: where do the numbers come from and why should we believe them?, CRA UK 5th Probabilistic Safety Analysis and Human Factors Assessment Forum, September 17-18, 2014.

• • Adapted from B. Lydell, K.N. Fleming, and J.-F. Roy, Analysis of possible aging trends in the estimation of piping system failure rates for internal flooding PRA, Proceedings of 14th International Conference on Probabilistic Safety Assessment and Management (PSAM 14), Los Angeles, CA, September 16-21, 2018.

Pipe Aging** Plant-Level Data*

26

Internal Flood Propagation K.N. Fleming and B. Lydell, Guidelines for Performance of Internal Flooding Probabilistic Risk Assessment, EPRI 1019194, Electric Power Research Institute, Palo Alto, CA, December 2009 27

Simulation from Idaho National Laboratory research supported by the U.S. Department of Energy https://safety.inl.gov/public/

Internal Flood Propagation Internal Flooding Video 28

External Hazards Notable External Hazards Analyses

• Seismic

- Long history with NPP PRA; strong input from geotechnical and structural engineering communities

- Performed for all plants (full SPRA or margins analysis)

- Can be an important or even dominant risk contributor

• External Floods

- Explicit analyses and important contributors for some plants

- IPEEE guidance allowed screening based on deterministic grounds; reviews focused on seismic and fire, treated floods as part of HFO (high winds, floods, and other)

- Renewed interest post-Fukushima

• High Winds

- Similar history as external floods

- Need to consider wind-driven missiles => simulation analysis 29

External Hazards External Hazards - General Approach

• Probabilistic Hazards Analysis

• Fragility Analysis

• Plant Response Analysis Adapted from NUREG/CR-6042 30

Seismic Events Probabilistic Hazards Analysis - Seismic NRC HQ

• Source strength

• Propagation to site

• Site response

• Structural response

• Multiple hazards

- Acceleration North Anna NPP

- Displacement V.M. Andersen, et al., Seismic Probabilistic Risk Assessment Implementation Guide, EPRI https://earthquake.usgs.gov/earthquakes/

3002000709, Electric Power Research Institute, Palo Alto, CA, December 2013 31

Seismic Events Fragility Analysis - Seismic

• Sources

- Models

- Shake table data

- Expert judgment

• Informed by post-earthquake investigations

• Considers frequency and failure mode

• Addresses both aleatory V.M. Andersen, et al., Seismic Probabilistic Risk Assessment Implementation Guide, EPRI 3002000709, Electric Power Research Institute, Palo Alto, CA, December 2013 and epistemic uncertainties

• Considers correlation 32

Seismic Events Plant Response Analysis - Seismic

• Modify internal events model to address effects of different magnitude earthquakes

• Seismic Equipment List (SEL)

• Induced hazards (internal floods and fires) Example SEL Headings

• Solution considers V.M. Andersen, et al., Seismic Probabilistic Risk Assessment Implementation Guide, EPRI 3002000709, Electric Power Research Institute, Palo Alto, CA, December 2013

- Correlation between SSCs

- Relatively high conditional probability of events => cant use rare event approximations 33

Seismic Events Seismic PRA Notes

• Technical community is generally comfortable with state of analyses

• Need to consider induced effects*

- Fires

- Floods

- Human (distractions, access limitations, worker safety, psychological impacts)

• Need for expert judgment

• Dominant risk not from biggest earthquakes.

• Example: pipes moved aboveground following the 2007 Kashiwazaki-Kariwa earthquake were swept away by the 2011 seismically-induced tsunami at Fukushima Dai-ichi. 34

External Floods Probabilistic Hazards Analysis - Flooding

• Flooding is a potential effect of multiple phenomena, sometimes in combination.

Examples:

- Wind-driven waves, storm surge, intense precipitation

- Seiche

- Tsunami

- Floods from upstream flood management decisions

• Multiple hazards, e.g.,

- Water levels (low and high)

- Dynamic forces

- Debris Example Tsunami Propagation Prediction

• Important considerations From V. Titov, et al., Tsunami Hazard Assessment Based on Wave Generation, Propagation, and Inundation

- Timing: warning, duration Modeling for the U.S. East Coast, NUREG/CR-7222, July 2016.

- Site location and design

• Multiple sources (historical, paleoflood, simulation models) 35

External Floods Probabilistic Hazards Analysis - Flooding Tsunami Video Simulation from Idaho National Laboratory research supported by the U.S. Department of Energy https://safety.inl.gov/public/

36

External Floods Fragility Analysis - Flooding

• Multiple hazards Levee Failure Modes

• Multiple damage mechanisms (not just overtopping)

• Need to consider barrier Overtopping Slope Instability elements (not just reactor systems)

- Permanent (e.g., dikes, doors, penetration seals, drainage PIping Erosion systems)* Adapted from T. Schweckendiek, Dutch approach to levee reliability and flood risk, Workshop on Probabilistic Flood Hazard Assessment,

- Temporary (e.g., sand bags, Rockville, MD, January 29-31, 2013.

inflatable barriers)

• States can change over time 37

External Floods Plant Response Analysis - Flooding

• Modify internal events model to address flooding effects

• For unscreened floods, assume instantaneous maximum hazard levels

• Potential effects on operators

- Ability to access areas

- Psychological impacts

• Mitigation systems

- Drainage

- Pumping 38

External Floods External Flood PRA Notes

• Multiple technical communities

- Growing agreement on meaningfulness of and need for quantitative risk assessment

- Performing analyses not focused on but relevant to NPPs

- Varying viewpoints on meaningfulness of frequency of very rare events

• Cliff edge characterization potentially misleading

- Damage mechanisms beyond overtopping

- Progressive damage states

- Unlikely confluence of likely events can be more important than overwhelming floods

• Non-stationarity concerns

- Climate

- Human-induced changes to landscape => runoff

• Should consider correlated (and possibly concurrent) non-flooding effects (e.g., LOOP due to high winds) 39

External Hazards Other Hazards - Example List*

Aircraft impact Local intense precipitation Avalanche Low lake or river water level Biological events Low winter temperature Coastal erosion Meteor or satellite strike Drought Onsite chemical release External fire Pipeline accident External flooding River diversion Extreme winds and tornadoes Sandstorm Fog Seiche Forest fire Seismic activity Frost Severe temperatures Hail Snow High summer temperature Soil shrink-swell High tide Space weather Hurricane Storm surge Ice cover Transportation accident Industrial/military facility accident Tsunami Internal flooding Turbine-generated missiles Landslide Volcanic activity Lightning

• See ASME/ANS PRA Standard for current list. 40

External Hazards A Structured View

• Unstructured lists

- Can have potentially important gaps (e.g., heavy load drops)

- Can have overlaps (e.g.,

external flooding and tsunamis)

- Include slowly developing conditions as well as events

- Dont show connections between phenomena (e.g.,

multiple storm-related hazards)

• Explicit display of causality might help

- Gaps

- Dependencies

- Screening 41

Observations

• Results highly plant specific (e.g., location of major equipment, cable routings, natural hazards occurrences and plant design)

• Maturity and realism a long-running issue; increased importance with current approaches to RIDM (e.g., per Regulatory Guide 1.174) 42

Cautions

• Overly rapid dismissal based on personal intuition (e.g.,

potential magnitudes and consequences) - Lecture 2-3

• Potential violations of fundamental assumptions (e.g.,

aleatory model and concept of frequency)

- Non-stationary processes

- Observation-based predictions (e.g., Near-Earth Objects, earthquakes?)

• Implementation assumptions

- Environmental qualifications

- Barrier existence, integrity

- Effectiveness of mitigation features (e.g., pumping, drainage) 43

Current Challenges

• New hazards (space weather, high-energy arc faults - HEAF, )

• Combinations of hazards

• Changing conditions (non-stationarity)

• Different technical disciplines, views on important issues, and heterogeneous analyses Subcommittee on Disaster Reduction, Space Weather www.sdr.gov 44

Knowledge Check At one plant, an unfortunate rodent caused a loss of offsite power by bridging two phases of a 3-phase AC power bus. For the purpose of NPP PRA, should this be considered a dependent failure?

45

Thought Exercise: Emergency Diesel Generator (EDG) Redundancy

• NPPs have two or more redundant EDGs to supply power if offsite power is lost.

• How might redundancy be threatened by spatial hazards?

USNRC, Diesel Generators as Emergency Power Sources (ADAMS ML11229A065) 46

Thought Exercise - EDG Addition N

A plant is planning on adding a new, air-cooled EDG to supplement its water-cooled EDG Cable Spreading Main Control Room Room (located in the EDG switchgear (existing)

Turbine Building). air-cooled EDG (new)

Switchgear water-cooled EDG From a spatial Room (existing)

EDG switchgear (new) hazards viewpoint, Safety Pumps what are some pros and cons of the proposed Section View update?

47

Thought Exercise In a recent news story, scientists from LANL have indicated that they are on the path to predicting earthquakes (using Big Data and AI). Should they be successful, should this change the way we approach seismic PRA? If so, how?

48