ML033230225

From kanterella
Revision as of 09:35, 20 March 2020 by StriderTol (talk | contribs) (StriderTol Bot insert)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Submittal of WCAP-16084-NP, Development of Risk-Informed Safety Analysis Approach and Pilot Application.
ML033230225
Person / Time
Site: Beaver Valley, Millstone, Calvert Cliffs, Salem, Mcguire, Palisades, Palo Verde, Indian Point, Kewaunee, Catawba, Harris, Wolf Creek, Saint Lucie, Point Beach, Watts Bar, Sequoyah, Byron, Arkansas Nuclear, Braidwood, Summer, Prairie Island, Seabrook, Surry, North Anna, Turkey Point, Ginna, Diablo Canyon, Callaway, Vogtle, Waterford, Farley, Robinson, South Texas, San Onofre, Cook, Comanche Peak, Fort Calhoun, McGuire, PROJ0694  Constellation icon.png
Issue date: 11/12/2003
From: Schiffley F
Westinghouse Owners Group
To:
Document Control Desk, Office of Nuclear Reactor Regulation
References
WOG-03-601 WCAP-16084-NP, Rev 0
Download: ML033230225 (34)
v • d • e


Text

DOMESTIC MEMBERS WCAP-16084-NP AmerenUE Callaway Project 694 American Electric Power Co.

D.C. Cook I&2 Arizona Public Service Co.

Palo Verde 1. 2 &3 November 12,2003 Constellation Energy Group Calvert Cliffs 2 Dominion Nuclear Connecticut Millstone 2 &3 WOG-03-601 Dominion Virginia Power North Anna I &2 Surry 1 &2 Duke Energy U. S. Nuclear Regulatory Commission Catawba I &2 McGuire I &2 Attention: Document Control Desk Entergy Nuclear Northeast Indian Point 2 &3 Washington, DC 20555-0001 Entergy Nuclear South ANO 2 Waterford 3 Exelon Generation Company LLC Attn: Chief, Information Management Branch Braidwood1 &2 Byron I &2 Division of Program Management FirstEnergy Nuclear Operating Co.

Beave Valley 1 £ 2 FPL Group St. Lucie I 2

Subject:

Submittal of WCAP-16084-NP, "Development of Risk-Informed Seabrook Turkey Point 3 &4 Safety Analysis Approach and Pilot Application" Nuclear Management Co.

Kewaunee Palisades Point Beach I &2 Transmitted herewith are four (4) non-proprietary copies of the subject topical report.

Prairie Isand Omaha Public Power District The Westinghouse Owners Group (WOG) is submitting WCAP-1 6084-NP, Rev 0, Fort Calhoun Pacific Gas &Electric Co. under the Nuclear Regulatory Commission (NRC) licensing topical report program.

Diablo Canyon 1 &2 Progress Energy H. B. Robinson 2 Shearon Harris The Risk-Informed Safety Analysis (RISA) approach presented in WCAP-16084-NP PSEG - Nuclear Salen I &2 addresses the classification of specific transients and accidents into realistic event Rochester Gas &Electric Co.

R. E. Ginna categories by considering the frequency of occurrence of the overall event combination South Carolina Electric £ Gas Co.

V. C. Summer (i.e., the initiating event in combination with coincident occurrences and a single Southern California Edison SONGS 2 &3 failure). Correspondingly, the acceptance criteria proposed for use in assessing STP Nuclear Operating Co.

South Texas Project I &2 conformance to regulatory criteria would be those associated with the realistic overall Southern Nuclear Operating Co.

J. M. Farley 2 event frequency rather than the unrealistically higher frequency of the initiating event A. W. Vogtle 1 &2 Tennessee Valley Authority alone.

Sequoyah 1 &2 Watts Bar I TXU Electric Commanche Peak 1 &2 Consistent with the Office of Nuclear Reactor Regulation, Office Instruction LIC-500, Wolf Creek Nuclear Operating Corp.

Wolf Creek "Processing Requests for Review of Topical Reports," the WOG requests that the NRC INTERNATIONAL MEMBERS document the acceptance of this report for review, establish a target date for requesting Electrabel Doel 1. 2.4 any additional information, and the planned date for issuance of the Safety Evaluation Tihange i &3 Electriclt6 de France (SE). Also requested is an estimate of the staff hours needed to complete the review of KansaI Electric Power Co.

Mihania 1 WCAP-16084-NP.

Takahama 1 Ohi 1 &2 Korea Hydro &Nuclear Power Co.

Kori 1 -4 The South Texas Project Units 1 and 2 are used in WCAP-1 6084-NP to provide an Ulchin3&4 Yonggwangl -5 example pilot application of the RISA approach. It is expected that the South Texas British Energy plc Sizewell B Project, as well as other participating utilities, will be referencing WCAP-1 6084-NP on NEK Krsko their respective dockets in support of licensing actions. To this end, the WOG Spanish Utilities Asco 1 &2 respectfully requests that the NRC complete its review and issue its SE no later than Vandellos 2 Ajrraraz I &2 June 2004.

Rlnghals AB Ringhals 2 - 4 Taiwan Power Co.

lMaanshan 1 &2

.U. S. Nuclear Regulatory Commission November 12,2003 Aib WOG-03-601 Page 2 of 2 All correspondence and invoices related to the review of WCAP-16084-NP should be addressed to:

Mr. Gordon Bischoff Manager, Owners Group Program Management Office Westinghouse Electric Company Mail Stop ECE 5-16 P.O. Box 355 Pittsburgh, Pennsylvania 15230-0355 If you require further information, please contact Mr. Jim Molkenthin in the Owners Group Program Management Office at (860) 731-6727 Sincerely, Frederick. P. "Ted" Schiffley, II, Chairman Westinghouse Owners Group Enclosures xc: Management Committee Analysis Subcommittee Licensing Subcommittee Project Management Office C. B. Brinkman, fM, w/o enclosures M. C. Jacob, M3Y, w/o enclosures J. S. Galembush, (W), w/o enclosures V. A. Paggen, (M), w/o enclosures S. Dembek, NRC, w/o enclosures D. G. Holland, NRC, w/ enclosures (via Federal Express)

Westinghouse Non-Proprietary Class 3 WCAP-16084-NP September 2003 Revision 0 Development of Risk-informed Safety Analysis Approach and Pilot Application WOG Task MUHP1095 CEOG Task 2076 Westinghouse

Westinghouse Non-Proprietary Class 3 WCAP-16084-NP, Rev. 0 Development of Risk Informed Safety Analysis Approach And Pilot Application September 2003 Author: Aq If (L' Mathew C. Jacob Reliability & Risk Assessment Group Author: A P ,

oaseph

& ezendes Transients & Setpoints Analysis Group Reviewer: - i RobertRaetA4tAith Reliability & RiAsk'ssessment Group Reviewer: A'Ji4 -'-(

Michael L: Howard Transients & Setpoints Analysis Group Approved: I ary A. Wassart Reliability & Risk Assessment Group 02003 Westinghouse Electric Company, LLC 2000 Day Hill Road, P.O. Box 500 Windsor, Connecticut 06095-0500 All Rights Reserved

EXECUTIVE

SUMMARY

Design basis analyses documented in Safety Analysis Reports (SARs) are performed deterministically using the guidance provided in Regulatory Guide 1.70 (Reference 1) and the Standard Review Plan (NUREG-0800, Reference 2). Regulatory Guide 1.70 suggests the use of risk informed safety analysis (RISA) for plant design basis events. It provides the guidance that the consequences of higher frequency events be evaluated against more stringent acceptance criteria in contrast to the guidance for the consequences of lower frequency events, which can be weighed against less restrictive acceptance criteria. The regulatory guidance also requires the use of coincident occurrences (COs) and single failures (SFs) in the analysis of the initiating event. When the frequency of the initiating event is combined with those of the CO and SF, the combined (or overall) frequency would be orders of magnitude lower than that of the design basis event defined by the initiating event alone. However, current regulatory guidance does not adequately take into consideration this impact when acceptance criteria are specified for an initiating event in combination with a CO and SF. Because the impact of the lower frequency combined event (i.e., IE+CO+SF) is measured against the same acceptance criteria as for the higher frequency event, there exists an inconsistency in terms of the real risk to public health and safety.

Unlike years past, the Nuclear Regulatory Comnmiission (NRC) and the nuclear industry now have sufficient staff members adequately trained in risk assessment methodology and its application to the design, operation and maintenance of nuclear power plants. It is, therefore, appropriate to now assess the RISA approach for use in future safety analyses. Consequently, the RISA Project was conceived and implemented to address the use of risk-informed considerations in conjunction with deterministic safety analyses. This is accomplished by:

(1) developing a risk-informed approach for event frequency re-categorization that considers events by their overall frequency and not just by that of the initiating event, (2) identifying a new event to replace the event being re-categorized, (3) analyzes the events, exactly as done today, using traditional NRC-approved deterministic analysis methods, (4) evaluates the consequences of the events using current regulatory acceptance criteria chosen, however, to be consistent with the re-categorized overall event frequency and not simply the initiating event frequency, and (5) applying the above approach to a pilot plant and an example design basis event scenario.

With regard to item (5), the viability and feasibility of the RISA approach in appropriately re-classifying design basis events and evaluating the results against applicable acceptance criteria were demonstrated by applying the approach to the Loss of Normal Feedwater (LONF) event analyzed for the South Texas Project plants. The results of the study suggest that (1) the LONF event in combination with a loss of offsite power (LOOP) and a failure of the Engineered Safety Features (ESF) signal has an overall frequency of less than 2.3E-07 per year, and (2) the acceptance criteria applicable for this event are those for the Limiting Fault 2 event category, instead of the current Moderate Frequency event category. Since the LONF in combination with a CO and SF was re-classified to the Limiting Fault 2 event category, a new Moderate WCA-1084N_ Re.0_Spebr20 WCAP-16084-NP, Rev. 0 i September 2003

-1 Frequency event would be defined to replace it. For the purpose of the current study, the definition of a replacement event and associated evaluation were not performed since they do not involve any new or different approaches from those traditionally used.

The results of the current study indicates that by applying the RISA approach, the categorization of plant transients and accidents can be accomplished more rigorously and systematically. In particular, the use of the approach allows the classification of specific transients and accidents into their proper (i.e., more realistic) event categories by considering the overall frequency of occurrence of the event combination, i.e., initiating event in combination with the CO and SF. Correspondingly, the acceptance criteria for the event combination generally becomes less restrictive than those currently used on the basis of the unrealistically higher frequency of the initiating event alone. The use of the RISA approach can lead to meeting the appropriate acceptance criteria for plant transients and accidents more readily, while maintaining the risk to public health and safety at a very low level consistent with the NRC's regulatory mandate. Since the RISA approach is independent of the nuclear steam supply system and/or fuel vendor, Westinghouse requests NRC acceptance of the RISA approach described herein for application in licensing actions for all plant and/or fuel types and for all design basis events.

_CP108 P Rev 0

_iSpebr20 WCAP-16084-NP, Rev. 0 ii September 2003

TABLE OF CONTENTS SECTION TITLE PAGE EXECUTIVE

SUMMARY

...................................................... i TABLES ..................................................... iv ACRONYMNS............................................................................................................v

1.0 INTRODUCTION

..................................................... 1 2.0 RISA APPROACH ...................................................... 3 2.1 Evolution of Event Categorization and Acceptance Criteria ................... ............... 3 2.2 RISA Procedure ....................................................... 4 3.0 APPLICATION OF THE RISA APPRAOCH ...................................................... 11 3.1 Selection of Pilot Plant and Initiating Event ..................................................... 11 3.2 Description of Initiating Event at STP ............................... ...................... 11 3.3 Regulatory Guidance for the Initiating Event ...................................................... 13 3.4 Probabilistic Consideration of the Initiating Event ............................................... 16 3.5 Replacement Event for the Re-categorized Event................................................. 16 4.0 THERMAL HYDRAULIC EVALUATION ...................................................... 18 4.1 Description of Safety Analysis of the Initiating Event ............................ ............. 18 4.2 Discussion of Assumptions and Single Failures ................................................... 19 4.3 Description of Results in Relation to Meeting Regulatory Requirements ............ 22 5.0 RESULTS OF RISA APPLICATION ...................................................... 23

6.0 CONCLUSION

S AND RECOMMENDATIONS .................................................... 24

7.0 REFERENCES

...................................................... 25 WCAP-16084-NP, Rev. 0 iii September 2003

I TABLES 2-1 Event Categorization Matrix .......................................................... 6 2-2 Evolution of Event Categorization ........................................................... 7 2-3 Evolution of Acceptance Criterion on Radiological Release Based on Categories ...... 8 24 Event Categorization Matrix Based on RISA .......................................................... 9 2-5 Categorization Probabilities/Frequencies and Radiological Release Acceptance Criteria ......................................................... 10 3-1 Recommended Revised Event Acceptance Criteria .................................................... 15 4-1 Assumptions Used in Pilot Plant Total Loss of Normal Feedwater Analysis ............20

_CP108 ,v P Rev 0 Setme 2003, WCAP-16084-NP, Rev. 0 iv September 2003

ACRONYMNS AFW Auxiliary Feedwater AFWS Auxiliary Feedwater System ANS American Nuclear Society ANSI American National Standards Institute AOO Anticipated Operational Occurrence AOR Analysis of Record CE Combustion Engineering CFR Code of Federal Regulations CO Coincident Occurrence DBE Design Basis Event DNB Departure from Nucleate Boiling DNBR Departure from Nucleate Boiling Ratio ECCS Emergency Core Cooling System ESF Engineered Safety Features GDC General Design Criteria IE Initiating Event LF Limiting Fault LOCA Loss-of-Coolant Accident LOESF Loss of Engineered Safety Features signal LONF Loss of Normal Feedwater LOOP Loss of Offsite Power NRC Nuclear Regulatory Commission OG Owners Group PC Plant Condition PORV Power Operated Relief Valve PRA Probabilistic Risk Assessment PSE Public Service Electric PSV Primary Safety Valve RCP Reactor Coolant Pump RCS Reactor Coolant System RISA Risk Informed Safety Analysis RRA Reliability and Risk Analysis SAFDL Specified Acceptable Fuel Design Limit SAR Safety Analysis Report SBDG Standby Diesel Generator SBLOCA Small Break LOCA SF Single Failure SRP Standard Review Plan (NUREG-0800)

STP South Texas Project TMI Three Mile Island UFSAR Updated Final Safety Analysis Report USNRC United States Nuclear Regulatory Commission WOG Westinghouse Owners Group WCAP-16084-NP, Re.0_Spebr20

_CP108-P Rev. 0 - v September 2003

1.0 INTRODUCTION

Design basis analyses documented in Safety Analysis Reports (SARs) are performed deterministically using the guidance provided in Regulatory Guide 1.70 (Reference 1) and the Standard Review Plan (NUREG 0800, Reference 2). Regulatory Guide 1.70 suggests the use of risk informed safety analyses for plant design basis events. It provides the guidance that the consequences of higher frequency events be evaluated against more stringent acceptance criteria in contrast to the guidance for the consequences of lower frequency events, which can be weighed against less restrictive acceptance criteria. The underlying principle of this guidance is that the risk to public health and safety from postulated plant transients and accidents, which is a function of the event frequency and consequences, should be maintained at a very low level.

The regulatory guidance also requires the use of coincident occurrences (COs) and single failures (SFs) in the analysis of the initiating event. When the frequency of the initiating event is combined with those of the CO and SF, the combined (or overall) event frequency would be orders of magnitude lower than that of the design basis event (DBE) defined by the initiating event alone. However, current regulatory guidance does not adequately take into consideration the significance of this impact when acceptance criteria are specified for an initiating event in combination with a CO and SF. Because the impact of the lower frequency combined event (i.e., LE+CO+SF) is measured against the same acceptance criteria as for the higher frequency initiating event alone, there exists an inconsistency in terms of the real risk to public health and safety.

In the late 1970's, Combustion Engineering risk-informed the safety analyses for two plant designs: St. Lucie Unit 2 and the CE Standard Plant design - System 80. Design basis events were categorized into five (5) groups based on the frequency of occurrence and the corresponding acceptance criteria were established consistent with the guidance of Regulatory Guide 1.70 (Reference 1) and the Standard Review Plan (SRP) (Reference 2). Complete safety analyses performed using this categorization scheme and acceptance criteria were documented in the SARs for these designs. The SARs were submitted to the Nuclear Regulatory Commission (NRC) for review. Although supportive of the approach used in performing these safety analyses, the NRC was not fully prepared to actually review such risk-informed applications and would have required significantly longer time periods for completing the safety evaluations of the SARs, thus potentially causing licensing delays for the designs involved. Therefore, the risk-informed safety analyses for these plant designs were withdrawn and replaced with SARs containing conventional deterministic safety analyses.

Unlike years past, the NRC and the nuclear industry now have sufficient staff members adequately trained in risk assessment methodology and its application to the design, operation and maintenance of nuclear power plants. It is, therefore, appropriate to now assess the RISA approach for use in future safety analyses. The RISA Project was conceived and implemented to address the use of risk-informed considerations in conjunction with traditional deterministic safety analyses. This is accomplished by:

I September 2003 WCAP-16084-NP, Rev.Rev. 0 0 I September 2003

(1) developing a risk-informed approach for event frequency re-categorization that considers events by their overall frequency and not just by that of the initiating event, (2) identifying a new event to replace the event being re-categorized, (3) analyzing the events exactly as done today, using traditional NRC-approved deterministic analysis methods, (4) evaluating the consequences of the events using current regulatory acceptance criteria chosen, however, to be consistent with the re-categorized overall event frequency and not simply the initiating event frequency, and (5) applying the above approach to a pilot plant and an example design basis event scenario.

Since the RISA approach is independent of the nuclear steam supply system and/or fuel vendor, Westinghouse requests NRC acceptance of the RISA approach described herein for application in licensing actions for all plant and/or fuel types and for all design basis events.

WCP108 WCAP-16084-NP, P Re. 02Spebr03 Rev. 0 2 September 2003

2.0 RISA APPROACH The governing principle of the RISA approach can be stated in one simple word 'consistency'.

That is, the risk associated with a plant transient or accident is a function of the overall frequency of occurrence and the measure of its consequences when compared to consistently selected regulatory acceptance criteria. The conventional deterministic approach in current use is to the contrary 'inconsistent' in its application and, therefore, cannot by definition present an accurate measure of the risk to the public health and safety. Current regulatory guidance stipulates that the risk to public health and safety as a result of transients and accidents be maintained at a very low level. This guidance, therefore, implies that the consequences of a high probability event should be small. Conversely, the consequences of low probability events can have a higher acceptance threshold in comparison to higher probability events.

As indicated in Table 2-1, the current regulatory guidance (References 1 and 2) groups plant transients into three (3) broad classes:

1. Moderate Frequency Events,
2. Infrequent Events,
3. Limiting Faults References 1 and 2 also provide the guidance that each initiating event should also consider coincident occurrences along with the initiating event (e.g., loss of offsite power, iodine spike, etc.) and the single failure (SF) of an active component or system (e.g., loss of a Diesel Generator, loss of a HPSI pump, etc.). The regulatory guidance does not, however, adequately account for the significant change in the overall frequency that results when the initiating event is combined with a CO or SF, since the only change in the acceptance criteria is the allowance for fuel failure for initiating events "in combination with any single active component failure, or single operator error...". No loss of function of any fission product barrier other than the fuel cladding is allowed. In addition, the accidents are grouped into a single category, namely, Limiting Fault, which is shown to have a wide frequency range (1.OE-6 to L.OE-2 per year).

In contrast, the RISA approach would re-classify the Design Basis Events (DBEs), including those with COs and/or SFs, and their corresponding (i.e., congruent) acceptance criteria by the overall event frequency. Additionally, when a DBE is re-categorized to a lower frequency class (e.g., Moderate to Infrequent), a new event will be defined to replace it in the category from which the original DBE was removed. Further, the newly defined event will be one that truly belongs in that category based on its overall frequency of occurrence. That is, it would not become a candidate for re-categorization in the future.

2.1 Evolution of Event Categorization and Acceptance Criteria The evolution of event categorization schemes over a period of time can be seen from Table 2-2, which shows the categorization based on Title 10 of the Code of Federal regulations (10CFR)

(References 3 and 6), American National Standards Institute (ANSI) 18.2 (Reference 4), NRC Rev 0 WCAP-16084-NP,~~~~~ .etmbr20 WCAP-16084-NP, Rev. 0 3 September 2003

I Regulatory Guide 1.70 (Reference 1), ASME Code and ANSI 51.1-1983 (Reference 5). In particular, it shows the relationship between the various categorization schemes.

Table 2-3 shows the evolution of the acceptance criteria on radiological releases/fuel performance for Title 10 CFR, ANSI 18.2, and the Standard Review Plan (Reference 2). It is seen that for Title 10 CFR, the acceptance criteria are simply for two broad categories of events, namely, Anticipated Operational Occurrences (AOOs) and Accidents, whereas for the SRP, the evolution of the acceptance criteria has resulted in three broad categories of events. The last category, Limiting Faults, is divided into three subcategories with appropriate acceptance criteria attached to each. This approach leads to the breakdown of the Limiting Fault category (which covers a wide range of frequencies, 106 to 10.2) into smaller frequency groups and results in more meaningful application of the concept of a constant, acceptably low risk due to applicable accident scenarios.

Note that other acceptance criteria (e.g., criteria on peak primary system and secondary system pressures) can also be dealt with in a similar manner as was done for the acceptance criteria for radiation release. Examples are the limits on the peak RCS and secondary side pressures. The peak pressure limit for the Moderate Frequency events would be significantly lower than that for the Limiting Fault 3 category of events.

The systematic application of the RISA approach would start off with a listing of the types of events and their categorization into the various frequency categories using a matrix such as shown in Table 24. The limiting events in each type of event class are listed along with the frequency category to which they belong. Based on the frequency categorization, the appropriate acceptance criteria for the limiting event can be identified.

Table 2-5 shows a typical categorization scheme and relevant radiological release acceptance criteria for each category of events.

2.2 RISA Procedure The procedure for risk-informing the categorization of the DBEs involves the following:

1. Review the existing Analysis of Record (AOR) for the specific DBE for which it is desired to risk-inform the event categorization. Clearly identify the initiating event (IE),

and the limiting CO and SF.

2. Using an industry database of event frequencies and probabilities, determine the frequency of occurrence for the IE, and the conditional probabilities for the CO and SF.
3. Calculate the overall frequency of the DBE using the individual frequency of occurrence for the IE, and the conditional probabilities for the CO and SF.
4. Compare the overall frequency of occurrence for the DBE under consideration against the frequency of occurrence for the three (3) broad classifications of events (moderate September 2003 WCAP-16084-NP, Rev.Rev. 00 44 September 2003

frequency, infrequent and limiting faults) provided in the event categorization matrix (e.g., Table 2-1). Assign the DBE to the appropriate frequency category based on the comparison.

5. If the selected DBE can be re-categorized, define a new replacement DBE for the frequency category from which the original DBE is removed (See Section 3.5 for guidelines for defining the replacement event).
6. Identify the congruent acceptance criteria that must be met for the re-categorized DBE and the new replacement event using the criteria defined in the regulatory guidance (e.g.,

Table 2-3) for the calculated overall frequency of occurrence.

,CP104N, Re. 0 _ Septevber 200 WCAP-16084-NP, Rev. 0 5s September 2003

Table 2-1 Event Categorization Matrix Other Categorization Schemes Event Frequency US NRC ANS/ANSI Range RG 1.70 (per reactor-year) 10CFR RG 1.48 ASME Rev. 3 N18.2 51.1-1983 (References 3 & 6) Code* (Reference 1) (Reference 4) (Reference 5)

Planned Operations Normal Normal Normnal Condition ! PC-I Anticipated Moderate Frequency Condition II PC-2

1OE-I------- Operational Occurrences Upset _-_-_--

Infrequent Condition mII PC-3

-. Incidents

IOE-2------- _- - - _ _ ___

Emergency Condition IV l--IOE-3------

~ ~PC-4 Limiting l1E4----Accidents Faults

1OE-5------- Faulted PC-S

IOE-6------- _- -

Not Considered

  • Information extracted from Reference 5.

WCAP-16084-NP, Rev.0 6 September 2003

Table 2-2 Evolution of Event Categorization 10 CFR ANSI 18.2 Reg. Guide 1.70 ANSI 51.1 Normal Operation Condition I PC-1 Normal Operation Condition II Anticipated Operational Moderate Frequency Moderate Frequency PC-2 Occurrences (AOOs) "...may occur during a

"... are expected to occur one or calendar year..."

more times during the life..." Condition HI Infrequent Incidents Infrequent PC-3

"...may occur during a lifetime..."

Accidents Condition IV PC-4

... exceedingly low probability of Limiting Faults Limiting Faults PC-5 occurrence..."

PC: Plant Condition 7S pt m er2 0 WPC16 8 -N , Re.

WCAP-16084-NP, Rev.0 7 September 2003

Table 2-3 Evolution of the Acceptance Criteria on Radiological Release Based on Categories 10 CFR ANSI 18.2 Section 15

_________________________ Standard Review Plans Anticipated Operational Condition II Moderate Frequency Occurrences (AOOs) Moderate Frequency . SAFDLs are not exceeded

  • Specified Acceptable Fuel
  • No loss of function of any not exceeded barrier to radioactive product escape Condition III Infrequent Infrequent Incidents
  • Event dependent
  • Damage to small fraction of fuel elements Accidents Condition IV Limiting Faults
  • 1) Small fraction of 10 CFR

Table 2-4 Event Categorization Matrix Based on RISA Limiting Faults LF-1 LF-2 LF-3 Moderate Type of Events Frequency Events Infrequent Events Incidents Not Incidents of Low Incidents of likely to Occur Probability Exceedingly Low (Small Fraction of (Well within 10 Probability 10 CFR 100) CFR 100) (10 CFR 100) l 2 _____-__ ___ X___

3 4

6 8

Note: This table is provided as an example of an event categorization matrix that displays potential event categories. Entries under column headings are not provided, since the purpose of the table is to identify only the event categories.

9 September 2003 WCAP-16084-NP, Rev.O WCAP-16084-NP, Rev.0 9 September 2003

Table 2-5 Categorization Probabilities/Frequencies and Radiological Release Acceptance Criteria 10 CFR AOOs Accidents Reg. Guide 1.70 Moderate Infrequent Events Limiting Faults (Ref. 1) and Frequency Event SRP (Ref. 2)

Incidents Not Incidents of Low Incidents of Likely to Occur Probability Exceedingly Low

____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __________ _ _ _ _ _ _ _ _ _ _ _ _ Probab ility Probability (per 1 -0.5 0.5 - 1.7E-2 1.7E 1.OE-3 1.OE 1.0E4 1.OE 1.0E-6 reactor year)

Frequency (per 20.693 0.693 - 1.7E-2 1.7E 1.OE-3 1.OE 1.0E-4 1.0E 1.OE-6 reactor year) I Acceptance Criteria Appendix I 1% IOCFRI1O 10% of 10CFR1OO 25% of 10CFR1OO 10CFRIOO DNBRŽSAFDL DNBR]31.0 I I I 0S pe b r2 0 WC P 1 0 4 N , R v0 WCAP-16084-NP, Rev.0 10 September 2003

3.0 APPLICATION OF THE RISA APPRAOCH The viability and feasibility of the RISA approach in classifying DBEs and evaluating the results against congruent acceptance criteria will be best understood by applying the approach to a specific DBE and plant. The objective is to use the RISA approach to:

(1) identify the current classification of the specific DBE, (2) identify the CO and SF used in the analysis, (3) determine the frequencies for the IE, CO and SF, (4) calculate the overall event frequency by combining the individual frequencies, (5) reclassify the selected DBE into a new frequency category and define a replacement event for the selected DBE and the original frequency classification, and (6) specify the new congruent acceptance criteria that the selected DBE would be subjected to based on the revised frequency classification.

3.1 Selection of Pilot Plant and Initiating Event A survey of Westinghouse Owners Group (WOG) plants was conducted to identify the pilot plant for the study and to narrow the list of IEs to be considered. Responses received from WOG members identified a suggested list of Ms that included the loss of normal feedwater, steam line break, boron dilution, and inadvertent ECCS actuation as example events. These events were suggested based on the potential difficulty faced in satisfying the acceptance criteria based solely on the IE frequency classification. The goal was to gain relief from the potential difficulty in satisfying the incongruent acceptance criteria by reclassification of the overall event frequency and use of the resulting congruent acceptance criteria.

Based on a review of the IEs recommended and discussions with the WOG members, the Loss of Normal Feedwater (LONF) event was chosen as the example IE and the South Texas Project units as the pilot plant. This IE benefits substantially from the application of the RISA approach.

3.2 Description of Initiating Event at STP The following is a brief description of the example initiating event for the South Texas Project Units 1 and 2. The 1E under consideration is a total loss of normal feedwater that can be caused by valve malfunctions or pump failures that result in a reduction in the capability of the secondary system to remove heat generated in the reactor core. A loss of offsite power can also cause a total LONF. However, for the analysis considered, a LOOP subsequent to reactor/turbine trip caused by low-low steam generator level resulting from the LONF is assumed. If an alternate supply of feedwater were not provided to the plant, core residual heat 11 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 0 0 I1I September 2003

following the trip would heat the primary system to the point where water relief from the pressurizer would occur, resulting in a substantial loss of water from the RCS. Since the plant is tripped well before the steam generator heat transfer capability is reduced, the primary system variables never approach a DNB condition.

In the short term, the event is either less limiting than other Condition II (moderate frequency) events or of little concern with respect to the SRP criteria on primary and secondary pressure and DNBR. The event is thus analyzed to ensure the pressurizer does not go solid (i.e., fill with water) and subsequently discharge water from the power operated relief valves (PORVs) and/or the Primary Safety Valves (PSVs). This ensures long term satisfaction of the above mentioned acceptance criteria. The PORVs are conservatively assumed to function in order to maximize the swell in pressurizer water level, thus maximizing the potential for pressurizer fill. Pressurizer fill can result in primary coolant being discharged to the containment due to a rupture of the pfessurizer relief tank rupture disk. The rupture disk would perform its pressure relief function if sufficient water is discharged from the pressurizer via the PSVs and/or the PORVs. Further, an uncontrolled release of primary fluid to the containment may occur in the event the PSVs or PORVs fail to reseat due to the discharge of subcooled water. Failure of these valves to reseat is likely in this case since neither of these valves is qualified to discharge water. Note, however, that in some plants the PORVs have been qualified to discharge water.

The above situation leads to a Condition II event becoming a more severe, potentially a Condition III or Condition IV event, since in this case a Small Break Loss of Coolant Accident (SBLOCA) would be in progress. For the STP units, the worst postulated LONF event with respect to potentially overfilling the pressurizer with eventual discharge of water is one in which a LOOP occurs coincident with the reactor trip. This is due to rapid depletion of the steam generator secondary inventory prior to the reactor trip on steam generator low-low level followed by reactor coolant pump (RCP) coastdown which further degrades the capability of the reactor coolant system to remove residual core heat.

The following events occur upon a LOOP coincident with a reactor trip:

1. Plant vital instruments are supplied from emergency DC power sources,
2. Increasing secondary pressure following a reactor trip results in the steam generator PORVs opening automatically to the atmosphere. Turbine bypass to the condenser is not available due to the loss of power. If steam flow through the PORVs is not available, the steam generator safety valves may lift to dissipate the sensible heat of the fuel and coolant in addition to the residual decay heat produced in the reactor.
3. As the no-load temperature is approached, the steam generator PORVs (or the safety valves if the PORVs are not available) are used to dissipate the residual decay heat and to maintain the plant at the hot standby condition.
4. The Standby Diesel Generators (SBDGs), started on loss of voltage to the plant emergency busses, begin to supply plant vital loads.

The auxiliary feedwater system (AFWS) for the STP units consists of four AFW pumps, three motor-driven and one turbine-driven, along with the necessary piping and valves to deliver WCP104NRv 2Spebr20 WCAP-16084-NP, Rev. 0 12 September 2003

feedwater flow to the four steam generators. The AFW pumps, which take suction from the auxiliary feedwater storage tank, are started automatically by an actuation logic that is controlled by the steam generator water level. The turbine-driven AFW pump utilizes steam from the secondary system and exhausts it to the atmosphere. The motor driven AFW pumps receive power from the SBDGs. The pumps take suction directly from the auxiliary feedwater storage tank for delivery to the steam generators. The subject event analyzed for the STP units assumes a SF in the AFW system. The failure assumed is that of a single train of AFW actuation logic causing the failure of two out of the four AFW pumps to automatically start. Hence, two motor driven pumps are started automatically delivering flow to two of the four steam generators.

Operator action is relied upon to manually start a third AFW pump resulting in flow being delivered to a third steam generator. Hence, three out of the four steam generators eventually receive AFW. Note that one AFW pump out of service is not allowed by Technical Specifications for the STP plants.

Upon a loss of power to the RCPs, coolant flow necessary for core cooling and removal of residual heat is maintained by natural circulation in the reactor coolant loops. In the long term and subsequent to AFW actuation, the addition of feedwater is manually controlled to maintain proper steam generator water level.

3.3 Regulatory Guidance for the Initiating Event The Regulatory guidance are best summarized in SRP Section 15.2.7 "Loss of Normal Feedwater Flow" (Revision 1, 1981).

The General Design Criteria (GDC) which are imposed on the acceptability of this event are:

GDC 10 - Reactor design: Requires that SAFDLs are not exceeded during AO0s, GDC 15 - Reactor coolant system design: Requires that the RCS be designed with appropriate margin to assure that the pressure boundary will not be breached during AO0s, GDC 26 - Reactivity control system redundancy and capability: Requires reliable control of reactivity changes to assure that SAFDLs are not exceeded during A0Os, TMI Action Plan (NUREG-0737) Items II.E.1.l and II.E.1.2 require that the design and performance of the auxiliary feedwater system should be such that it (1) can automatically start-up, (2) is capable of adequately removing the decay heat, and (3) has protection from a SF.

Based on the above guidance, the NRC has developed the following specific acceptance criteria:

a. Primary and main steam pressure must be less than 110% of design values,
b. Fuel cladding integrity must be maintained by ensuring DNBR remains above the 95/95 DNBR limit, 13 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 0 0 13 September 2003

I I

c. An incident of moderate frequency should not generate a more serious plant condition without other faults occurring independently,
d. An incident of moderate frequency in combination with any single active component failure, or single operator error, shall be considered and is an event for which the amount of fuel failure must be calculated for radiological dose calculations. There shall be no loss of function of any fission product barrier other than the fuel cladding.
e. Regulatory Guide 1.105, "Instrument Spans and Setpoints" positions are used with regard to their impact on plant response.
f. The most limiting plant systems SF, as defined in the "Definitions and Explanations" of Appendix A to 10 CFR Part 50, shall be identified and assumed in the analysis and shall satisfy the positions of Regulatory Guide 1.53, "Application of the Single-Failure Criterion to Nuclear Plant Protection Systems".
g. The analysis of the LONF transient should be performed using an acceptable analytical model.

The above criteria are thus applicable to the analysis of two specific event combinations, those being the LONF and the LONF with a SF.

One of the objectives of this topical report is to recommend a set of congruent acceptance criteria consistent with the above requirements for the specific event combination under consideration based on the frequency of the specific event combination. Hence, it may be determined that the IE in combination with certain COs and SFs can deviate from certain currently established acceptance criteria if the frequency of the event combination is sufficiently low. Table 3-1 provides the recommended revised congruent acceptance criteria that are based on the SRP criteria presented above. The frequency definitions are comparable to the Regulatory Guide 1.70 (Reference 1) definitions for Moderate Frequency events, Infrequent Events and Limiting Faults with the exception that the Limiting Fault category has been further divided into three subcategories: Limiting Fault I (LF-1), Limiting Fault 2 (LF-2) and Limiting Fault 3 (LF-3).

The table also provides the associated frequencies in terms of events per reactor year.

WCAP-16084-NP, Rev. 0 14 September 2003~~~~~

WCAP-16084-NP, Rev. 0 14 September 2003

Table 3-1 Recommended Revised Event Acceptance Criteria Parameter Moderate Infrequent Events+ Limiting Faults Frequency+ 0.693 - 1.7 x 10.2 LF-1+ LF-2+ LF-3+

2 0.693 1.7 x 10.2 3 10 104 104 - 106 RCS Pressure < 110% of Design < 110% of Design < 110% of Design < 110% of Design < 120% of Design Secondary Pressure < 110% of Design < 110% of Design < 110% of Design < 110% of Design < 120% of Design Fuel Performance DNBR > 95/95 DNBR > 95/95 Maintain Coolable Maintain Coolable Maintain Coolable Limit, Limit, Geometry Geometry Geometry No Fuel Melting No Fuel Melting Radially Averaged Enthalpy <

Licensed Limit*

Radiological Appendix I Very Small Fraction Small fraction of Well Within 10 CFR 100 Release I _I of 10CFR100 (1%) 10CFR100 (10%) IOCFR100 (25%) _

Since changes have been considered for this parameter based on industry research.

' Frequencies are in units of events per reactor year.

epe b r2 0

_C P 1 08 - P Rev. Re .01 WCAP-16084-NP, 0 15 September 2003

I 3.4 Probabilistic Consideration of the Initiating Event The LONF event is identified as a Moderate Frequency event in the SRP (Reference 2). From plant specific PRA information for the STP plants, the frequency of occurrence for this event (FLONP) is 5.08E-02 per year (mean value) and 1.13E-01 per year (95h percentile value). The CO is the LOOP on assumed turbine trip. This CO has a conditional probability (PLOOP) of 6.21E-04 (mean value) and 1.50E-03 (95h percentile value) for the STP plants. The SF is assumed to be the failure of an ESF signal that results in two out of the four AFW pumps not starting up based on a low steam generator water level signal. For the STP plants, this event has a conditional probability (PLoEsp ) of 6.05E-04 (mean value) and 1.33E-03 (95h percentile value). Thus, the overall frequency of this DBE is:

Fovcal = F LOP

  • PLOOP
  • PLOESP

= 5.08E-02

  • 6.21E-04
  • 6.05E-04 = 1.911E-08 (mean)

=1.13E-01

  • 1.50E-03
  • 1.33E-03 = 2.254E-07 (95h percentile)

This suggests that the LONF analyzed here has a significantly lower frequency than that for the "Moderate Frequency" classification due principally to the addition of the CO and the SF.

Consequently, the acceptance criteria for the event are justifiably less restrictive than those for the Moderate Frequency event, based on maintaining a very low risk to public health and safety.

Currently, the acceptance criteria to be met for the LONF event are the ones for a Moderate Frequency event.

Consistent with the approach described in Section 3.2.3 of Reference 5, a comparison of the overall frequency value calculated above with the frequencies shown in Table 2-5 was made. It suggests that the LONF 1E in combination with the CO of a LOOP and the SF of the ESF signal would fall into the outer fringes of the Limiting Fault-3 category. This is a shift of four categories (shift from moderate frequency event to Limiting Fault-3 category). To avoid excessive change in categorization that is simply based on the change in the frequency of occurrence, the RISA approach also considers the characteristics of the event under consideration in the categorization process. For the LONF event in combination with the given CO and SF, the pressurizer can fill up and potentially discharge two phase fluid or liquid through the PORV and the relief tank, if no mitigating operator action is assumed. This event would thus represent a subset of the SBLOCA event which has been categorized as a LF-2 event.

This leads to correspondingly higher acceptance criteria on the radiological releases as shown in Table 2-5. The analyses documented in the STP UFSAR show that the acceptance criteria on relevant parameters for a SBLOCA are met with adequate margins.

3.5 Replacement Event for the Re-categorized Event The shifting of the LONF IE in combination with a CO and SF into the Accident category, based on the rationale described above, would require the identification of another event to replace it in the Moderate Frequency event category. This would be accomplished using the same approach previously employed to group the "decrease in heat removal by the secondary system" (SAR Re.01.

WCP-604.P epebr20 WCAP-16084-NP, Rev. 0 16 September 2003

section 15.2) events into the various frequency categories. In general, the procedure for identifying the replacement event involves the following steps.

(1) Identify/compile initiating events in the "decrease in heat removal from the secondary system" classification whose frequencies fall within the Moderate Frequency category.

(2) Evaluate/order these events to identify the events that would be limiting within the Moderate Frequency event category. The objective is to reduce the number of events that need to be quantitatively analyzed. Qualitative evaluations/comparisons may be sufficient to identify the Moderate Frequency event that leads to the most limiting consequences.

(3) Perform limited amount of quantitative analyses to determine the event which leads to the most limiting consequences if qualitative evaluations do not identify the initiating event that results in the most limiting consequences.

(4) Choose the initiating event that gives the most limiting consequences as the replacement Moderate Frequency event. The event chosen, including the CO and SF, will have a frequency of occurrence that truly falls in the Moderate Frequency classification. That is, the replacement event will not become a candidate for re-classification at a future date.

As a consequence of applying the above guidelines, an appropriate replacement Moderate Frequency event for the LONF IE in combination with a CO and SF may be the LONF event by itself. This is because the SRP specifically requires the analysis of the LONF event whose frequency of occurrence falls within the Moderate Frequency event category.

For the purpose of the study documented in this topical report, the definition of a replacement event and associated evaluation were not performed since they do not involve any new or different approach from those traditionally used.

17 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 0 0 17 September 2003

I 4.0 THERMAL HYDRAULIC EVALUATION 4.1 Description of Safety Analysis of the Initiating Event As mentioned in Section 3.2, the LONF event for the STP plants is such that one criterion is imposed (pressurizer does not go solid) to ensure long term satisfaction of Criteria "a" and "b" of Section 3.3, i.e., fission product barrier integrity. The short term satisfaction of these criteria is ensured either (1) by being bounded by other events or (2) by being of minor concern for this event. The pressurizer fill criterion also ensures satisfaction of Criterion "c": i.e. the LONF event does not progress to a more severe event, that being a SBLOCA.

The safety analyses supporting the UFSAR thus focuses on selecting parameters and utilizing assumptions that exacerbate the potential for pressurizer fill. The analysis was performed utilizing the RETRAN computer code. The event combination analyzed was as follows:

(1) Initiating Event Total Loss of Normal Feedwater (2) Reactor Trip Credited Low-Low Steam Generator Water Level (3) Coincident Loss of Offsite Power, Two (2) Seconds Following Occurrence (CO) Reactor Trip (Rods Begin to Drop)

(4) Single Failure Failure of an Engineered Safety Features (ESF) Signal:

Two (2) Out of Four (4) Auxiliary Feedwater (AFW)

Pumps Do Not Start (5) Automatic AFW Two Motor Driven AFW Pumps Start 60 Seconds Following the AFW Actuation Signal (6) Third AFW Pump A Third AFW Pump is Assumed to be Started Manually 15 Minutes Following AFW Actuation The inclusion of a SF addresses Criteria "d" and "1" of Section 3.3. The remaining Criteria "e" and "g" are not relevant to the current study, since this study focuses on the change in the overall frequency of the LONF event in combination with a CO and SF.

Section 4.2 details the assumptions utilized to establish the limiting case. The most limitin combination was used in the final analysis. The pressurizer volume of the STP plants is 2100 ft .

The limiting case as presented in the UFSAR yielded a maximum pressurizer water volume of 2040 ft3 for the long term analysis of the LONF event with a LOOP. In order to achieve this acceptable maximum water volume it was necessary to require that two (2) motor driven pumps be available to supply auxiliary feedwater to the steam generators automatically and a third auxiliary feedwater pump be available to be manually started 15 minutes following auxiliary feedwater actuation. This requirement prevents the Technical Specifications from being relaxed to allow a single motor driven auxiliary feedwater pump to be out of service indefinitely. This Rev 0 WCA-1084N. 8Spebr20 WCAP-16084-NP, Rev. 0 18 September 2003

condition along with an assumed SF of the auxiliary feedwater turbine driven pump would result in only two (2) motor driven pumps being available to supply auxiliary feedwater to the steam generators, an unacceptable condition that will result in the pressurizer going solid.

The RISA approach, however, has the potential to allow relaxation of applicable Technical Specifications in this case depending on the frequency of the event resulting in the pressurizer going solid.

4.2 Discussion of Assumptions and Single Failures The assumptions utilized in the performance of the STP plant LONF analysis with a LOOP are detailed in Table 4-1. As mentioned in Section 3.2, the SF assumed was the failure of an ESF train to actuate auxiliary feedwater which results in only two out of the four auxiliary feedwater pumps starting automatically.

These assumptions are specific to the STP Units I and 2 plants. Future analyses for other plants, that utilizes the RISA approach discussed within this document, will need to define the assumptions that are specific to the plant being analyzed as was done here for the STP plants.

Rev. 00 19 September 2003 WCAP-16084-NP, Rev. 19 September 2003

Table 4-1 Assumptions Used in the Pilot Plant Total Loss of Normal Feedwater Analysis ITEM ASSUMPTION COMMENT 1 Both Minimum Tube Plugging (0%) and Maximum Tube Plugging (10%) considered. Minimum tube plugging Is limiting.

2 A-AFW pump (loop 1)no longer 'out of service' Indefinitely (T.S. 3.7.1.2 ACTION a.)______________________

3 Limiting Single Failure is failure of ESFAS actuation 'Train A7 which precludes automatic delivery of AFW flow to steam generators A and D.

4 AFW delay time for auto start: 60 seconds Assumes diesel start time as well; hence very conservative for cases without LOOP.

S Operator action time to start third AFW pump: 15 minutes following reactor trip on Low-Low Steam Generator Water Level.

6 A94 Steam Generators.

7 Credit taken for a LOOP resulting in 50% of the Backup Heater Capacity (as opposed to 100%) being Applicable to LOOP cases only.

available due to Emergency Diesel Generator loading.

8 Backup Heaters actuate on both pressurizer level deviation and pressure effects. ______

9 Low Steam Generator Water Level Trip Setpoint of 10.1% Narrow Range Level Span (NRS). This was reduced from previous analyses to account for the pressure drop across the steam generator mid-deck plate.

10 Credit for thick metal masses associated with the reactor vessel and steam generator Inlet and outlet plenums was not taken in this analysis. This differs from the previous analysis.

11 LOOP subsequent to reactor trip. LONF event analyzed with and without LOOP.

12 LOOP is assumed with a 2 second delay following reactor trip.

13 Initial NSSS Power 3821 MWt + 2% uncertainty 14 Pump Heat- maximum assumed 24 MWt Value assumes 4 RCPs operating.

Upon RCP trip pump heat will coast down proportional

___________________________________________________ to rotational speed of the pumps.

15 Both low and high nominal Tavg considered with + 5.10 F uncertainty. High nominal minus uncertainty limiting.

16 Both low and high Initial Pressurizer Pressure considered nominal 2250 psia + 46 psi uncertainty Nominal plus uncertainty limiting.

17 Initial RCS Flow = Thermal Design Flow =98,000 gpmAoop 18 Initial Pressurizer Level = 64.1% NRS (high Tavg, full power nominal level (57% NRS) plus 7.1% NRS High Tavg Program plus uncertainty limiting.

uncertainty) and 47.1% NRS (low Tavg. full power nominal level (40% NRS) plus 7.1% NRS uncertainty) 19 Initial Steam Generator Level = 76.3% NRS (full power nominal level (70.7% NRS) plus 5.6% NRS uncertainty) 20 Main Feedwater Temperature: 390 TF and 440 'F 440 OF limiting 21 Low-Low Steam Generator Water Level Trip delay time = 2.0 seconds 22 Physics Parameters: Minimum Reactivity Feedback Note: for the acceptance criterion of Interest, a stuck rod is of little consequence since pressurizer level Is governed by long term heat removal. Decay heat rate would have a first order effect on meeting this criterion.

20 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 00 20 September 2003

Table 4-1 Assumptions Used in the Pilot Plant Loss of Main Feedwater (Continued)

ITEM ASSUMPTION COMMENTS 23 Decay Heat: ANS 1979 + 2a_ 2a uncertainty Included on decay heat 24 Pressurizer spray assumed to actuate via the pressurizer pressure control system Spray actuation more adverse for maximizing pressurizer level. Hence, the relevant control systems are assumed to operate.

25 Pressurizer proportional and backup heaters actuate as part of the pressurizer pressure control system. Heater actuation more adverse for maximizing Backup heaters also actuate on high pressurizer level deviation 5% NRS. pressurizer level. Hence, the relevant control systems are assumed to operate.

26 Pressurizer PORV actuation Is assumed to occur on an uncompensated signal setpoint of 2350 psia PORV actuation swells the pressurizer level; hence is (for an initial pressurizer pressure of 2296 psia). an adverse assumption for this acceptance criterion.

For low Initial pressurizer pressure a compensated signal setpoint of 100 psid was assumed (I.e. for an initial pressurizer pressure of 2204 psla, 2304 psla was assumed). However, high initial pressurizer pressure Is limitina.

27 Vessel Mixing; Perfect mixing is assumed for LONF with LOOP. Design mixing Is assumed for LOW cases.

28 Fuel Heat Transfer Data: Minimum UAs. As opposed to maximum.

29 Auxiliary Feedwater Enthalpy a maximum; corresponds to 120 0F.Umits FCS heat removal.

30 Maximum AFW purge volume: 40 ftSlAoop.

31 Minimum AFW flow 500 gpm/pump __mits RCS heat removal.

32 Rod control system: Off No credit for this control system 33 Pressurizer Level Control System: Off No credit for charging and letdown is assumed.

34 Initial Steam Generator Conditions Consistent with Initial NSSS Power, Thermal Design Flow, Initial GENF code generates IGOR input for eventual use In T., ,, Initial Pressurizer Pressure, Nominal Steam Generator Level + Uncertainty, and Steam Generator RETRAN tube Plugging.

35 Tech Spec MSSV setpoints Increased by. 3% (tolerence + drift) and a 21.8 psI AP between the steam Minimize heat removal generator and the MSSV sensing point.

36 Secondary PORV actuation credited. Opening pressure Is 1282.0 psla which includes a 3% tolerance. Credited with customer consent. One secondary PORV The assumed full open pressure Is 1345.35 psla which Includes 5% accumulation. per steam generator was modeled.

21 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 00 21 September 2003

I 4.3 Description of Results in Relation to Meeting Regulatory Requirements The criterion of interest for the analysis of the LONF event is the maintenance of the pressurizer water volume to less than 2100 ft3 , which is the volume of a filled pressurizer. Satisfaction of this criterion ensures no water relief via the PSVs or PORVs. The discharge of water from either of these valves can result in a stuck open valve with subsequent rupture of the pressurizer relief tank rupture disk, causing primary coolant to be discharged to the containment. This would result in violation of two (2) NRC acceptance criteria as defined by the SRP: 1) the event would progress to a more serious event (i.e. a LONF would propagate to a SBLOCA); and 2) for the event with a SF, the criterion which states that there shall be no loss of function of any fission product barrier other than the fuel cladding would be violated since the primary coolant barrier would be bypassed. The event as analyzed for the pilot plant meets the 2100 fW requirement, however, this requires overburdening restrictions on auxiliary feedwater system availability via the plant Technical Specifications. The Utility's preference would be to allow a single auxiliary feedwater pump to be out of service such that a SF in conjunction with a LONF event would result in only two (2) auxiliary feedwater pumps delivering flow to two (2) steam generators. In this scenario, the SF would be assumed to be in the turbine driven pump, and since the remaining motor driven pump is out of service, AFW delivery would then be limited to two (2) pumps.

Exploring the frequency of the event being analyzed helps to alleviate the burdensome restriction on the AFW system availability since the event with all its conservatism (LONF, LOOP, SF in a safety system etc.) is shown to be a Limiting Fault rather than an AOO. The pressurizer becoming water solid with water being discharged to the containment is an acceptable consequence for a Limiting Fault event, provided all regulatory acceptance criteria for the consequences of the event are met.

Section 3.4 addresses these possibilities from a risk perspective. It indicates that the overall frequency of the LONF event in combination with a LOOP and a failure of the ESF signal is 2.254E-07 based on 95th percentile values of EB frequency and conditional probabilities and 1.91E-08 based on mean values. These exceedingly low frequency values place this event in the Limiting Fault 2 category as discussed in Section 3.4. This categorization would allow the specific LONF event being considered to meet less restrictive acceptance criteria (e.g., those applicable to a SBLOCA).

22 September 2003 WCAP-16084-NP, Rev.

WCAP-16084-NP, Rev. 00 22 September 2003

5.0 RESULTS OF RISA APPLICATION The application of the RISA approach resulted in the reclassification of the EE (i.e., LONF) in combination with a CO and SF into a frequency category having frequencies that are several orders of magnitude smaller. This category is Limiting Fault 2. Consequently, the acceptance criteria for this event are shifted to those for a SBLOCA from the current acceptance criteria that are applicable to the Moderate Frequency event. The SBLOCA event was chosen, since the increasing RCS pressure and pressurizer level could potentially lead to a SBLOCA via the PORVs/PSVs and the pressurizer relief tank if no operator actions are assumed.

604NRv WCP- 3Spebr20 WCAP-16084-NP, Rev. 0 23 September 2003

I

6.0 CONCLUSION

S AN]D RECOMMENDATIONS By applying the RISA approach, the categorization of plant transients and accidents can be accomplished more rigorously, systematically, and more reasonably. In particular, the use of the RISA approach allows the classification of specific transients and accidents into more meaningful event categories by considering the frequency of occurrence of the IE and conditional probabilities of COs and SFs. Correspondingly, the acceptance criteria for an 1E in combination with a CO and SF would become less restrictive than those resulting from the use of the current deterministic approach. Current regulatory guidance suggests that the l! in combination with a CO and SF should use acceptance criteria that are essentially the same as those applicable to the initiating event by itself, even though the frequencies for both scenarios differ by orders of magnitude.

The use of the RISA approach can more easily lead to acceptable results for plant transients and accidents, since the re-categorization based on this approach would support the use of less restrictive acceptance criteria. The risk to public health and safety would be maintained at a very low level consistent with the regulations due to the fact that the event consequences would still be judged commensurate with the frequency of occurrence.

The viability and feasibility of the RISA approach to bring event categorization and acceptance criteria into congruence have been demonstrated by specific application to a pilot plant and an example event. The application of the RISA approach to other PWR designs would be analogous, although the plant system/component configuration, alignment, COs and SFs are very plant dependent.

24 September 2003 WCAP-16084-NP, Rev. 00 WCAP-16084-NP, Rev. 24 September 2003

7.0 REFERENCES

1. US NRC Regulatory Guide 1.70, "Standard Format and Content of Safety Analysis Reports for Nuclear Power Plants (LWR Edition)", USNRC, Revision 3, November, 1978.
2. NUREG-0800, "Standard Review Plan", Revision 2, USNRC, July 1981.
3. Title 10, Code of Federal Regulations, Part 100, "Reactor Site Criteria".
4. ANSI N18.2, "American National Standard, Nuclear Safety Criteria for the Design of Stationary Pressurized Water Reactor Plants", August 6, 1973.
5. ANSIIANS-51.1-1983, "American National Standard Nuclear Safety Criteria for the Design of Stationary Pressurized Water Reactor Plants", April 29, 1983.
6. Title 10, Code of Federal Regulations, Part 50, "Domestic Licensing of Production and Utilization Facilities".

25 September 2003 WCAP-16084-NP, Rev. 0 WCAP-16084-NP. Rev. 0 25 September 2003

Retrieved from "https://kanterella.com/w/index.php?title=ML033230225&oldid=2638218"