ML081840534

06 - 06/20/2008 NFPA-805 Workshop Handouts - Change Evaluation for Non-Pilot Diablo Canyon

ML081840534
Person / Time
Site:	Diablo Canyon
Issue date:	06/20/2008
From:	Afzali A Pacific Gas & Electric Co
To:	Office of Nuclear Reactor Regulation
Shared Package
ML081840360	List: ML081840398 ML081840534 ML081840731 ML081840735 ML081840744
References
NFPA 805
Download: ML081840534 (15)

Text

NFPA 805 Public Workshop Non-pilot Fire PRA Transition Experiences June 19 & 20 Bethesda, Maryland Risk Assessment Process in Support of Change Evaluation Amir Afzali Diablo Canyon Power Plant 1

Change Evaluation Risk Assessment Scope To present a process for the determination of delta risk (defined in terms of CDF and LERF) in support of change evaluation.

Contributors:

Diablo Canyon Power Plant (Amir Afzali, John Pyo, Nathan Barber)

Scientech EPM 2

Definitions Risk Informed/Performance Based (RIPB)

Compliance Strategy- Includes non-approved strategies/features in the CLB and/or non-deterministic criteria that can be demonstrated to provide adequate protection under the NFPA 805 licensing basis using performance-based/risk-informed methods- e.g, An AFW pump power cable associated with decay heat removal which has only a 1 hr wrap but no automatic fire suppression protection A charging injection motor operated valve associated with the inventory control safety function which may spuriously close and has been resolved by a non-approved manual recovery action 3

Definitions (Cont.)

A RIPB Compliance Set- Represents a group of safe shutdown strategy features which are to be addressed within a single change evaluation.

A Post-Transition Safe Shutdown Strategy- Represents the plant configuration together with those credited manual actions which can be demonstrated to provide adequate protection under the NFPA 805 licensing basis through compliance with the required deterministic or risk-informed/performance-based criteria.

A Virtual Deterministic Compliance Strategy- represents a virtual plant configuration where, together with approved or allowed manual actions and all approved exemptions or deviations, would meet a NFPA-805 prescriptive compliance criteria for a particular safety function within a given fire area.

4

Risk Assessment Steps Step 1. Develop a Preliminary Post Transition Fire Risk Model and Quantify the Risk for Each Fire Area 1.1 For each fire area, analyze the safe shutdown compliance strategies based on the prescriptive requirements of the NFPA-805 standard, including multiple spurious actuation requirements, identifying functions that do not meet the prescriptive requirements.

1.2 For each safety function in an area, identify potential RIPB strategy that could be/is credited to address the non-compliance issue for the function. These may include newly identified strategies (e.g., non-approved operator action that may be credited to respond to a potential MSO) or an existing one (e.g., an existing non-approved operator action).

1.3 Develop a preliminary transition fire risk model. This risk model represents all passive and active safe shutdown strategies and features that are credited in the pre-transition program plus any required enhancements that may have been identified to address safe shutdown issues following completion of Step 1.2.

1.4 Quantify and assess the preliminary post transition model developed in Step 1.3 to identify those safe shutdown strategies/features that are of very low risk-significance.

1.5. Based on the results of Step 1.4, by eliminating low-risk significant features, refine the preliminary post transition safe shutdown model and re-quantify. This model will be used to evaluate the RIPB compliance risk for each fire area.

5

Risk Assessment Steps (Cont.)

Step 2. Develop and Quantify the Virtual Deterministic Compliance Risk Model for Each Fire Area and Safety Function 2.1 For each fire area, using results of step 1.2, identify the changes which would be required to achieve a prescriptive compliance for each safety function individually.

2.2 Treating each safety function separately, modify the preliminary post transition risk model developed in Step 1.5, assuming the changes identified in Step 2.1 are in place and quantify the fire area risk in terms of CDF and LERF.

6

Risk Assessment Steps (Cont.)

Step 3. Determine the Change in Risk 3.1 Calculate the change in risk for a safety function change set by subtracting the CDF/LERF estimate derived for the RIPB compliant strategy obtained in Step 1.5 from the fire area CDF/LERF estimates obtained from the virtual deterministic compliant area derived in Step 2.2.

3.2 Repeat Step 3.1` for each function in an area.

3.3 Obtain the total change in risk for each fire area by summing up the results obtained for each safety function in Step 3.2. This is the change evaluation delta CDF and LERF result to be compared with the RG1.174 criteria 3.4 If the change in risk derived in step 3 is smaller than RG1.174 criteria then the "RIPB strategies/features is (are) acceptable means of NFPA 805 compliance.

If the change is risk is not acceptable, then address in Step 4.

3.5 Repeat Steps 3.1 through 3.4 for each fire area and safety function and identify all potential strategies/features that either meet the prescriptive requirements or the "non-prescriptive" compliance requirements 7

Risk Assessment Steps (Cont.)

Step 4. Re-evaluate Deterministic Compliance and Post Transition Risk Model-4.1 Propose enhancements to the preliminary transition strategy. If such enhancements would result in compliance with the NFPA deterministic criteria then no further action is required.

4.2 If not then repeat steps 2.2, 3.1, 3.3, 3.4 and 4.1 until fire area is RIPB compliant.

8

Example Unit 1 12 kV Switchgear Room-Step 1 Results:

Affected Safety Function - Electrical Support- Potential unavailability of offsite power and all three EDGs (due to fire-induced cable damage).

Current Safe Shutdown Strategy- Operator manual action to verify S/U feeder breaker is open and manually align two EDGs.

Detailed fire modeling identified no credible fire scenario where more than one EDG would be inoperable (Since there is no exposed combustible in the compartment none of the fires propagate outside of the switchgear cabinets although inter cabinet propagation is possible. Furthermore due to the size of the room and large ventilation pathways, hot gas layer formation will not occur even in the event of multiple cabinets being affected).

Current Safe Shutdown Operator action can potentially be eliminated.

CDF is calculated for the following scenarios:

Full area burn up-up- Assuming non-non-recoverable LOOP, loss of all EDG, and loss of recovery action (HEP = 1.0), CDF = 7.6E-7.6E-3 per year Detailed area analysis-analysis- Assuming non-non-recoverable LOOP, loss of one EDG at a time (with no recovery of the lost EDG), CDF = 1.2E-1.2E-5 per year 9

Example Unit 1 12 kV Switchgear Room-Step 2 Results:

Virtual deterministic compliance strategy was to exclude cables which prevent EDG start and load.

CDF is calculated a non-recoverable LOOP, loss of one EDG. CDF = 1.2 E-5 per year (note that this is the same as the detailed analysis results)

Step 3 Results: CDF (between the virtual case and the detailed model case) = 0 Currently credited manual action is a candidate to be eliminated from the Post Transition fire protection program.

10

An Actual Unit 1 12 kV Switchgear Room Fire- LER 200-004 On May 15, 2000, Diablo Canyon Unit 1 experienced an electrical short and fire in a 12kV bus duct which resulted in a turbine trip and a reactor trip.

The electrical fire damaged the class 2 4kV bus ducts and both sources of offsite power were unavailable to the vital 4kV system in Unit 1.

The emergency diesel generators started and loaded properly, supplying power to the vital 4kV buses and their associated equipment. All Unit 1 emergency equipment had power.

Unit 1 buses were powered from the offsite power by backfeeding from the 500kV system approximately 33 hours3.819444e-4 days <br />0.00917 hours <br />5.456349e-5 weeks <br />1.25565e-5 months <br /> later. However, DCPP had concluded that the offsite power to the emergency buses could have been recovered in about 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />.

Unit 2 remained at 100% power throughout the event.

11

An Actual Unit 1 12 kV Switchgear Room Fire- LER 200-004 12

An Actual Unit 1 12 kV Switchgear Room Fire- LER 200-004 13

An Actual Unit 1 12 kV Switchgear Room Fire- LER 200-004 14

An Actual Unit 1 12 kV Switchgear Room Fire-LER 200-004

. Estimated CCDPs-Turbine Trip = 1.834E-05 LOSP Sensitivity = 7.17E-05 Translates to (at worst case) to about 5.5E-7 per year Differences between actual and postulated events Turbine trip Initiating event vs LOOP Initiating Event No damage to EDGs vs damage at least to one EDG Recoverable consequential loss of power to emergency bosses vs non recoverable one 15