ML23093A162

From kanterella
Jump to navigation Jump to search
Automating the Level 2 - Level 3 Interface: Lessons Learned from Soarca and the NRC Site Level 3 PRA Project
ML23093A162
Person / Time
Issue date: 04/14/2023
From: Hathaway A
NRC/RES/DSA
To:
References
Download: ML23093A162 (1)
v • d • e


Text

Automating the Level 2-Level 3 Interface:

Lessons Learned from SOARCA and the NRC Site Level 3 PRA Project Trey Hathaway, PhD U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research Accident Analysis Branch Alfred.Hathaway@nrc.gov 14th European MELCOR User Group Meeting Ljubljana, Slovenia 12-14 April, 2023

MACCS Overview Early (Emergency) Phase Intermediate Phase Long-Term Phase Primary Offsite

  • Protect public from exposures
  • Protect public from
  • Protect public from Accident Response to passing plume and exposures to deposited exposures to deposited deposited materials materials materials Objective(s)
  • Plan for long-term
  • Conduct long-term cleanup and recovery cleanup and recovery activities activities Typical Duration ~1 week, starting at the time of Weeks to months, starting Months to years, starting at and Time Frame the accident initiation at the end of the early the end of the intermediate phase phase Exposure Pathways
  • Inhalation
  • Groundshine
  • Groundshine
  • Skin Deposition
  • Inhalation of
  • Inhalation of
  • Cloudshine resuspended materials resuspended materials
  • Groundshine
  • Food and water ingestion Protective Actions
  • Sheltering
  • Relocation
  • Decontamination
  • KI ingestion
  • Interdiction
  • Evacuation
  • Condemnation
  • Relocation 2

MACCS Overview MACCS is a fully integrated, engineering-level severe accident consequence computer code developed to analyze the offsite consequences of a hypothetical release of radioactive material to the environment 3

SOARCA - UA Process 4

Level 3 PRA

  • Numerous Release Categories (16) with representative source terms and release category frequencies
  • Representative accident sequences have specific Emergency Action Level (EAL) timings
  • Five possible Emergency Plans depending on source term characteristics 16.1 km [0-10 mile] EPZ evacuation (e.g. ~7 cohorts) 16.1 km [0-10 mile] EPZ evacuation with 16.1-24.1 km [10-15 mile] expanded evacuation with/without a schools cohort (e.g. ~12/13 cohorts) 16.1 km [0-10 mile] EPZ evacuation with 16.1-32.2 km [10-20 mile] expanded evacuation with/without a schools cohort (e.g. ~17/18 cohorts) 5

SOARCA vs Level 3 PRA Source Terms SOARCA Sequoyah Level 3 PRA - IEIF 6

Level 3 PRA

  • Want to rely on WinMACCS to create and execute MACCS calculations

- Challenge

  • 16 source terms
  • Perform a base set of calculations along with potential sensitivities
  • MACCS models have a lot of release category dependent input which could be error prone if building multiple models
  • Goal: Minimize potential input error by creating a base model which defines global input and change only emergency plan and source term input

- Ideally a user could define an emergency plan and add it to a MelMACCS source term and execute the calculation as a cyclic file set

  • Challenge: Users must define the number of cohorts at the beginning of the problem, which must be acknowledged by the user

- Want to create a text file which defines source term and emergency plan

  • Import model changes and execute the calculation with WinMACCS 7

WinMACCS Methodology

  • Create a calculation pipeline to minimize user interaction during calculations

- Excel macro created to execute pipeline

  • Three phases to calculation methodology employed:

- MelMACCS processing of MELCOR Source Terms

- MACCS Non-Evacuating Calculations

- MACCS Consequence Calculations

  • Including additional sensitivities 8

WinMACCS Methodology

  • Automatic generation of inputs minimizes time needed for input file development, improves traceability, and minimizes chances for user error by reducing manual data entry 9

MelMACCS

  • Rely on batch processing of MELCOR source terms
  • Create a *.mel file which defines flowpath and plume segment information

- Defined for all flowpaths in a MELCOR problem

  • Execute MelMACCS from Windows Command Prompt

- Create batch files to perform calculations Demo.mel from MelMACCS distribution

>melmaccs.exe ProjectFile.mel -i SourceTerm.ptf -o MACCSSourceTerm.inp -r 10

Non-Evacuating Calculations

  • Used to select base-case emergency response models
  • Simplified analysis with single non-evacuating cohort

- Used subset of weather trials based on SOARCA methodology

- Estimates the size of population subject to normal and hotspot early-phase relocation (i.e. exceeding the early phase PAG levels of 10-50 mSv [1-5 rem] in 4 days) as a function of distance from the site

- Estimate distance range encompassing 90%

of affected population to select emergency plan

- Estimate the size of the affected population to determine the intermediate- and long-term phase durations

  • Data from Fukushima suggests most
  • Example: Fraction of population affected by recovery actions would be focused on the emergency phase protective actions curve area adjacent to residences, farmland, and public spaces - EP Model 2 selected 11

WinMACCS Level 3 PRA Base Consequence Model

  • Create a Base Consequence Model containing all possible cohorts (18 cohorts)
  • 8760 weather trials
  • Tabulated results include:

MACCS Measure Output Type Spatial Interval Units Collective total effective dose 5 0-50 and 0-100 mi person-rem Total latent fatality cases 1 0-50 and 0-100 mi persons Population-weighted individual latent fatality risk 8 0-10 mi individual risk (unitless)

Total early fatality cases 1 0-50 mi persons Population-weighted individual early fatality risk 8 0-1.8 mi individual risk (unitless)

Area exceeding 555 kBq/m2 Cs-137 D 0-50 and 0-100 mi mi2 Population relocated during intermediate phase 14 0-50 and 0-100 mi persons Total economic costs 10 0-50 and 0-100 mi 2015$

12

WinMACCS Emergency Plan Consequence Models

  • SUMPOP file

- Provides spatially dependent population distributions for individual cohorts

- Create a SUMPOP file for each Emergency Plan (EP)

  • 5 files

- If a cohort is not present in an emergency plan, its population distribution is set to 0 for all grid sectors

  • Population of neglected cohorts is shifted into remaining non-evacuating cohort
  • Base model was copied and a EP specific SUMPOP files provided to define EP specific consequence model 13

Emergency Plan Definition

  • OALARM - reference point from which protected actions are initiated

- Set to 0 seconds for this analysis

  • Delay-to-Shelter (DLTSHL) - time from OALARM to until cohort begins to shelter

- Time from accident initiation until cohort begins to shelter since OALARM is set to 0 seconds

- Covers time for a protective action recommendation to the offsite response organization, notification of the public to evacuate by the offsite response organization, and the time needed to begin sheltering

  • Delay-to-Evacuation (DLTEVA) - length of the sheltering period from the time a cohort enters the shelter until the cohort begins to evacuate
  • Cohorts are assumed to travel a fixed average distance during evacuation phase

- DURBEG - Beginning phase of the evacuation

  • Typically a few minutes as evacuees begin to enter the evacuation network
  • Has accompanying travel speed: ESPEED1

- DURMID - remaining time required to safely exit the evacuation zone

  • Travel speed assumed to decrease due to congestion on evacuation network
  • Has accompanying travel speed: ESPEED2
  • ESPEED3

- Cohort speed from the end of the evacuation zone to the end of evacuation network

  • DLTEVA, DURBEG, DURMID, ESPEED1, ESPEED2, and ESPEED3 assumed to be independent of release category 14

Emergency Plan Definition

  • Mobilized

- Time until cohort is ready to Evacuation Time Estimate (ETE) begin evacuation

  • Exited

- Time cohort has fully exited the evacuation zone

  • Cohort 1 +

- 30% of the EPZ population (0-30%)

+

  • Cohort 2

- 30% of the EPZ population (30-60%)

  • Cohort 3

- 30% of the EPZ population (60%-90%)

  • Tail

- 10% of the EPZ population (90%-100%)

15

Emergency Plan Definition

  • Relocation time (Hotspot and Normal)

- Non-evacuating individuals relocated based on dose projections

- Amount of time affected non-evacuating individuals are exposed prior to relocating

  • Based on the longest evacuating cohort
  • Assumes 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> to identify areas subject to evacuation following plume arrival
  • Hotspot relocation

- 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> plus 90% ETE

  • Normal relocation

- 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> plus 100% ETE 16

Consequence Calculations

  • Excel macro:

- Creates a directory for the release category consequence model

- Copies emergency plan specific base consequence model into the release category directory

  • Selection based on simplified non-evacuating calculation

- Determines if school cohort is necessary based on timing of the accident progression

- Creates text file defining source term and release category emergency plan

  • User individually imports source term and emergency plan file then executes MACCS models 17

Output Post-Processing Range/Level Mean Output Type Output Subtype Example output from WinMACCS model using SOARCA Sequoyah Realization 554 source term

  • A script may be used to post-process the model1.out files and convert them into a database format
  • Post-processed files for each source term are concatenated into one file

- Enables quick parsing of the data

  • Scripts were written to automatically create plots based on user input

- User supplies the Cohort, Output Type, Output Subtype, and Range/Level to define a specific figure

- A graphical user interface was constructed to enable the exploration of data 18

Output Post-Processing

  • WinMACCS 3.11 can output results to tab delimited text file

- Change DEBUG_BIN_RESULTS flag in WinMACCS.ini file to TRUE

  • Binary file contents saved as text file in RESULTS_DB folder
  • tbl_outStat.txt and tbl_outCCDF.txt Model1.out Example output from WinMACCS model using SOARCA Sequoyah Realization 554 source term tbl_outStat.txt
  • One data block-must break into subcomponents 19

Conclusion

  • Creating a calculation pipeline for consequence calculations requires upfront effort but assists in future analysis efforts

- Increases efficiency for future analyses

  • e.g. sensitivities, alternative initiating events, potential reruns due to calculation errors

- Increases quality assurance of results by minimizing user interaction 20

References

April 2022. ADAMS Accession Number: ML22067A215.

  • U.S. Nuclear Regulatory Commission. NUREG/CR-7270. Technical Bases for Consequence Analyses Using MACCS (MELCOR Accident Consequence Code System). Washington, DC.

October 2022. ADAMS Accession Number: ML22294A091.

  • U.S. Nuclear Regulatory Commission. NUREG-2254. Summary of the Uncertainty Analyses for the State-of-the-Art Reactor Consequence Project. Washington, DC. October 2022. ADAMS Accession Number: ML22193A244.
  • U.S. Nuclear Regulatory Commission. NUREG/CR-7155. State-of-the-Art Reactor Consequence Analyses Project: Uncertainty Analysis of the Unmitigated Long-Term Station Blackout of the Peach Bottom Atomic Power Station. Washington, DC. May 2016. ADAMS Accession Number:

ML16133A461.

  • U.S. Nuclear Regulatory Commission. NUREG/CR-7245. State-of-the-Art Reactor Consequence Analyses (SOARCA) Project: Sequoyah Integrated Deterministic and Uncertainty Analyses.

Washington, DC. October 2019. ADAMS Accession Number: ML19296B786.

  • U.S. Nuclear Regulatory Commission. NUREG/CR-7262. State-of-the-Art Reactor Consequence Analyses Project: Uncertainty Analysis of the Unmitigated Short-Term Station Blackout of the Surry Power Station. Washington, DC. December 2022. ADAMS Accession Number: ML22194A066.

21

Retrieved from "https://kanterella.com/w/index.php?title=ML23093A162&oldid=3535869"