ML22080A258

RIC 2022 TH21 NRC Extreme Weather Presentation

ML22080A258
Person / Time
Issue date:	03/10/2022
From:	Christopher Hunter NRC/NRR/DRA
To:
Lane J
References
Download: ML22080A258 (14)

Text

Weather-Related Loss of Offsite Power Trends and Risk Insights Chris Hunter Office of Nuclear Regulatory Research Division of Risk Analysis Performance and Reliability Branch

Overview

• Evaluations of recent events caused by severe weather conditions have provided important risk insights.

• Severe weather events are most likely to lead to losses of offsite power (LOOPs).

- A review of current LOOP data and trends can be used to evaluate whether more extreme weather events affect nuclear power plants.

• Key Questions

- Are the frequency and duration of weather-related LOOPs changing?

- What do the results and insights from recent weather-related LOOPs tell us?

LOOP Data and Trends LOOP Frequencies

• The overall frequency of all LOOPs is decreasing over the past 15 years (2006-2020).

• There is no statistically significant trend for weather-related LOOPs during the past decade (2011-2020).

Weather-Related LOOPs (2006-2020)

Power Ops/ Duration Event Date Plant Cause Shutdown (hours) 8/10/2020 Duane Arnold Hurricane Power Ops 25.3 8/3/2020 Brunswick 1 Hurricane Power Ops 14.0 3/13/2018 Pilgrim Snow and Wind Shutdown 12.0 10/8/2016 Shearon Harris Hurricane Shutdown 7.4 1/27/2015 Pilgrim Snow and Wind Power Ops 0.03 2/8/2013 Pilgrim Snow and Wind Power Ops 21.0 10/29/2012 Oyster Creek High Winds Shutdown 14.4 4/27/2011 Browns Ferry 1 Tornado Power Ops 0.03 4/27/2011 Browns Ferry 2 Tornado Power Ops 0.03 4/27/2011 Browns Ferry 3 Tornado Power Ops 0.03 4/16/2011 Surry 1 Tornado Power Ops 5.1 4/16/2011 Surry 2 Tornado Power Ops 5.1 8/19/2009 Wolf Creek Lightning Power Ops 2.2 2/24/2007 Duane Arnold Ice Shutdown 17.5

LOOP Duration

• The duration of all LOOPs showed an increasing trend during the 1997-2020 period.

• There is no statistically significant trend for the duration of weather-related LOOPs during this same period.

Durations by LOOP Type Parameter Plant Switchyard Grid Weather LOOP event count 16 32 17 19 Mu (µ) 0.70 0.44 0.35 1.37 Standard error of µ 0.53 0.29 0.33 0.51 Sigma () 2.13 1.66 1.36 2.24 Standard error of 0.38 0.21 0.23 0.36 Fitted median, hour 2.00 1.56 1.41 3.94 Fitted mean, hour 19.17 6.21 3.55 48.23 Fitted 95th percentile, hour 66.10 24.03 13.19 156.51 Error factor 32.97 15.42 9.33 39.70

Recent Events Brunswick LOOP during Hurricane Isaias

• Storm-generated debris resulted in a LOOP to Unit 1 in August 2020.

• The LOOP lasted approximately 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />.

• The mean conditional core damage probability (CCDP) was 2x10-5.

- LOOP transient scenarios dominated risk; station blackout (SBO) risk was minimal.

Duane Arnold LOOP during Derecho

• Severe winds (100-130 miles per hour) during a derecho resulted in a LOOP in August 2020.

- The storm caused severe damage to nonsafety-related cooling towers and minor damage to a few buildings.

- The high winds also resulted in increased debris loading to the essential service water system and resulted in a clogged strainer.

• The LOOP lasted approximately 25 hours2.893519e-4 days <br />0.00694 hours <br />4.133598e-5 weeks <br />9.5125e-6 months <br />.

• The mean CCDP was 8x10-4.

- SBO scenarios were the dominant risk contributors.

Waterford LOOP during Hurricane Ida

• The high winds, heavy rain, and localized flooding resulted in damage to both sources of offsite power.

• The supplemental diesel generator experienced a failed battery due to rapid discharge after the LOOP occurred.

• The LOOP lasted approximately 53 hours6.134259e-4 days <br />0.0147 hours <br />8.763227e-5 weeks <br />2.01665e-5 months <br />.

• The preliminary analysis indicates a mean CCDP from mid 1x10-4 to low 1x10-3.

- SBO scenarios were the dominant risk contributors.

General Risk Insights

• SBO risk is dominant for two emergency diesel generator (EDG) plants for long-duration LOOPs.

- Multiunit sites with shared EDGs typically have much lower risk.

- Having an EDG not included in the same common-cause component group as the other safety-related EDGs can be a significant benefit.

- Modeling of common-cause failures across the units introduces significant uncertainties because the data do not support this modeling.

• LOOP duration has significant impact on plants that have dominant SBO risk.

• FLEX credit can have significant impact on the results.