Difference between revisions of "Seismic PRA Outputs (Task 14)"
Jump to navigation
Jump to search
SeismicAdmin (talk | contribs) m (1 revision imported) |
SeismicAdmin (talk | contribs) |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 44: | Line 44: | ||
* Completeness uncertainty is associated with risk contributors that are not included in the PRA model. | * Completeness uncertainty is associated with risk contributors that are not included in the PRA model. | ||
* Parameter uncertainty is associated with the input parameter values used to quantify the failure frequencies and probabilities of other events in the PRA (e.g. initiating events, human error probabilities) | * Parameter uncertainty is associated with the input parameter values used to quantify the failure frequencies and probabilities of other events in the PRA (e.g. initiating events, human error probabilities) | ||
| − | :- The parametric uncertainty propagation for a typical SPRA can be performed using the commercially available software UNCERT Version 3.0 (part of the EPRI | + | :- The parametric uncertainty propagation for a typical SPRA can be performed using the commercially available software UNCERT Version 3.0 (part of the EPRI [https://www.epri.com/research/products/000000003002023512 Phoenix Architect software suite]) or an alternative software. |
:- A Monte Carlo (or Latin Hypercube) evaluation of PRA logic can be performed using correlated or uncorrelated probability distributions to represent the inputs for the basic events. The probability density distribution describing the uncertainty in a component failure probability is characterized as a state of knowledge about an assumed fixed value. | :- A Monte Carlo (or Latin Hypercube) evaluation of PRA logic can be performed using correlated or uncorrelated probability distributions to represent the inputs for the basic events. The probability density distribution describing the uncertainty in a component failure probability is characterized as a state of knowledge about an assumed fixed value. | ||
:- The range factor for the sampled CDF (or LERF) can be calculated as the CDF (or LERF) at the 95<sup>th</sup> percentile divided by the median CDF (or LERF) at the 50<sup>th</sup> percentile (that is, 95%/50%). | :- The range factor for the sampled CDF (or LERF) can be calculated as the CDF (or LERF) at the 95<sup>th</sup> percentile divided by the median CDF (or LERF) at the 50<sup>th</sup> percentile (that is, 95%/50%). | ||
| Line 53: | Line 53: | ||
'''Sensitivity Analysis''' | '''Sensitivity Analysis''' | ||
| − | Example sensitivity cases to address SPRA modeling uncertainty include | + | Example sensitivity cases to address SPRA modeling uncertainty include: |
* Seismic hazard curve sensitivity | * Seismic hazard curve sensitivity | ||
* Fragility parameters sensitivity | * Fragility parameters sensitivity | ||
| Line 60: | Line 60: | ||
* HRA sensitivities | * HRA sensitivities | ||
* Treatment of SSLOCA sensitivity | * Treatment of SSLOCA sensitivity | ||
| − | * Impact on LERF sequence sensitivity | + | * Impact on LERF sequence sensitivity |
==Supplemental Guidance== | ==Supplemental Guidance== | ||
| Line 68: | Line 68: | ||
===EPRI Guidance=== | ===EPRI Guidance=== | ||
| − | ''Practical Guidance on the Use of Probabilistic Risk Assessment in Risk-Informed Applications with a Focus on the Treatment of Uncertainty'' (1026511). Section 5.10.2 provides a summary of generic sources of seismic PRA modeling uncertainty | + | ''Practical Guidance on the Use of Probabilistic Risk Assessment in Risk-Informed Applications with a Focus on the Treatment of Uncertainty'' ([https://www.epri.com/research/products/000000001026511 1026511]). Section 5.10.2 provides a summary of generic sources of seismic PRA modeling uncertainty |
| − | ''Seismic Probabilistic Risk Assessment Implementation Guide'' ( | + | ''Seismic Probabilistic Risk Assessment Implementation Guide'' ([https://www.epri.com/research/products/000000003002000709 3002000709]) provides general guidance to identify sources of uncertainty associated with the seismic hazards, seismic fragilities, and seismic PRA model throughout the document. Example sensitivity cases are provided regarding correlation and SPRA HEPs. |
| − | ''Surry Seismic Probabilistic Risk Assessment Pilot Plant Review'' (1020756) identifies sources of uncertainty consistent with EPRI and ASME/ANS guidance. The report documents the parametric uncertainty analysis. In addition, a set of sensitivity cases were performed based on a review of the dominant contributors to seismic risk. | + | ''Surry Seismic Probabilistic Risk Assessment Pilot Plant Review'' ([https://www.epri.com/research/products/000000001020756 1020756]) identifies sources of uncertainty consistent with EPRI and ASME/ANS guidance. The report documents the parametric uncertainty analysis. In addition, a set of sensitivity cases were performed based on a review of the dominant contributors to seismic risk. |
| − | '' | + | ''Phoenix Architect v2.0 Suite'' ([https://www.epri.com/research/products/000000003002023512 3002023512]) |
===Other Guidance=== | ===Other Guidance=== | ||
| − | NUREG-1855, Rev1, Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decision making, provides an industry-recognized methodology for identifying sources of plant-specific or generic uncertainty including external events. | + | [https://www.nrc.gov/reading-rm/doc-collections/nuregs/staff/sr1855/index.html NUREG-1855, Rev1], Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decision making, provides an industry-recognized methodology for identifying sources of plant-specific or generic uncertainty including external events. |
| − | + | [https://www.iaea.org/publications/8235/development-and-application-of-level-1-probabilistic-safety-assessment-for-nuclear-power-plants IAEA Safety Guide SSG-3], Development and Application of Level 1 Probabilistic Safety Assessment for Nuclear Power Plants, provides general guidance on the identification of sources of uncertainty consistent with EPRI and ASME/ANS guidance. IAEA SSG-3 does not provide any specific guidance on the performance of parametric uncertainty analysis or sensitivity studies. | |
| − | |||
| − | IAEA SSG-3, Development and Application of Level 1 Probabilistic Safety Assessment for Nuclear Power Plants, provides general guidance on the identification of sources of uncertainty consistent with EPRI and ASME/ANS guidance. IAEA SSG-3 does not provide any specific guidance on the performance of parametric uncertainty analysis or sensitivity studies. | ||
Latest revision as of 09:48, 30 July 2024
Task Overview
Objective
The objective is to provide final and intermediate results to provide insights gleaned from the SPRA analysis.
Purpose
This section describes the following expected outputs for an SPRA.
- Typical quantitative results
- Conclusions and insights
- Uncertainty and sensitivity analysis guidance
Guidance
Typical Quantitative Results
Typical results from an SPRA include the following:
- Important seismic initiators
- - CDF contribution by seismic ground motion bin
- - Conditional core damage probability values for the seismic interval initiators
- Importance contributors to CDF
- - FV importance identifies the fractional contribution to the CDF risk metric for the cutsets containing the given basic event. Typical SSCs with high FV importance include structural failures of buildings and events related to loss of station power (that is, offsite power).
- - RAW characterizes risk by comparing the increase in risk if a given component is assumed to be completely unavailable (that is, failed). Typical SSCs with high RAW importance include structural failures of buildings and multi-train, common-cause failure events. RAW values are typically lower for SPRA results than they would be for other PRA hazard assessments.
- Cutset evaluation
- - By identifying sequences of events in cutsets with high frequencies that lead to core damage, it becomes possible to take preemptive mitigating actions to reduce operational risk.
- Dominant accident sequences
- - The summary of the top accident sequences provides insights to the significant CDF contributors.
- Accident classes
- - The CDF distribution among the contributing functional accident sequence classes should be summarized to identify the accident sequence progression and timing characteristics for the dominant CDF scenarios. The dominant CDF accident classes may also provide insights to the Level 2 (LERF) characterization.
Conclusions and Insights
Typical conclusions from a seismic PRA model may include the following:
- Develop an appreciation of severe accident behavior under seismic challenges.
- Understand the most likely seismically induced severe accident sequences that could occur at the plant.
- Gain a more quantitative understanding of the overall frequency of core damage due to seismic events.
- Identify insights to reduce the overall probabilities of core damage and fission product releases by modifying hardware and procedures that would help prevent or mitigate severe accidents.
Uncertainty Analysis
There are two types of uncertainty in probabilistic risk models: aleatory and epistemic.
- Aleatory uncertainty is associated with the random nature of events. These include initiating events (seismic initiating events) and component failures.
- Epistemic uncertainty is associated with lack of knowledge of information. This includes statistical inferences from data.
Epistemic uncertainties may be classified into three different categories:
- Completeness uncertainty is associated with risk contributors that are not included in the PRA model.
- Parameter uncertainty is associated with the input parameter values used to quantify the failure frequencies and probabilities of other events in the PRA (e.g. initiating events, human error probabilities)
- - The parametric uncertainty propagation for a typical SPRA can be performed using the commercially available software UNCERT Version 3.0 (part of the EPRI Phoenix Architect software suite) or an alternative software.
- - A Monte Carlo (or Latin Hypercube) evaluation of PRA logic can be performed using correlated or uncorrelated probability distributions to represent the inputs for the basic events. The probability density distribution describing the uncertainty in a component failure probability is characterized as a state of knowledge about an assumed fixed value.
- - The range factor for the sampled CDF (or LERF) can be calculated as the CDF (or LERF) at the 95th percentile divided by the median CDF (or LERF) at the 50th percentile (that is, 95%/50%).
- Model uncertainty is associated with the aspects of the PRA model which may be represented by different modeling approaches.
Specific areas of uncertainty identified for seismic PRA model include HRA, earthquake caused correlations or dependencies, relay chatter and recovery analysis. More details regarding uncertainty approaches are found in the guidance listed below.
Sensitivity Analysis
Example sensitivity cases to address SPRA modeling uncertainty include:
- Seismic hazard curve sensitivity
- Fragility parameters sensitivity
- Correlation assumption sensitivity
- Number of hazard curve intervals sensitivity
- HRA sensitivities
- Treatment of SSLOCA sensitivity
- Impact on LERF sequence sensitivity
Supplemental Guidance
Related Element of ASME/PRA Standard
Part 5, Seismic Plant-Response (SPR)
EPRI Guidance
Practical Guidance on the Use of Probabilistic Risk Assessment in Risk-Informed Applications with a Focus on the Treatment of Uncertainty (1026511). Section 5.10.2 provides a summary of generic sources of seismic PRA modeling uncertainty
Seismic Probabilistic Risk Assessment Implementation Guide (3002000709) provides general guidance to identify sources of uncertainty associated with the seismic hazards, seismic fragilities, and seismic PRA model throughout the document. Example sensitivity cases are provided regarding correlation and SPRA HEPs.
Surry Seismic Probabilistic Risk Assessment Pilot Plant Review (1020756) identifies sources of uncertainty consistent with EPRI and ASME/ANS guidance. The report documents the parametric uncertainty analysis. In addition, a set of sensitivity cases were performed based on a review of the dominant contributors to seismic risk.
Phoenix Architect v2.0 Suite (3002023512)
Other Guidance
NUREG-1855, Rev1, Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decision making, provides an industry-recognized methodology for identifying sources of plant-specific or generic uncertainty including external events.
IAEA Safety Guide SSG-3, Development and Application of Level 1 Probabilistic Safety Assessment for Nuclear Power Plants, provides general guidance on the identification of sources of uncertainty consistent with EPRI and ASME/ANS guidance. IAEA SSG-3 does not provide any specific guidance on the performance of parametric uncertainty analysis or sensitivity studies.