defining initiating events for purposes of probabilistic safety assessment

8/3/2019 Defining Initiating Events for Purposes of Probabilistic Safety Assessment

1/152

IAEA-TECDOC-719

Defining initiating eventsfor purposes ofprobabilistic safety assessment

JAIASeptember 1993


2/152

The IAEA does not normally maintain stocks of reports in this series.However, microfiche copies of these reports can be obtained from

INIS ClearinghouseInternational Atomic Energy AgencyWagramerstrasse 5P.O. Box 100A-1400 Vienna, Austr ia

Orders should be accompanied by prepayment of Austrian Schillings 100,-in the form of a cheque or in the form of IAEA microfiche service couponswhich may be ordered separately from the INIS Clearinghouse.


3/152

P L E A S E B E A W A R E T H A TAL L O F T H E M I S S I N G P A G E S IN THIS D O C U M E N TW E R E O R IG IN A L L Y B L A N K


4/152

Copies of this lAEA-TECDOC may be obtained from:Safety Assessment SectionInternational Atomic Energy AgencyWagramerstrasse 5P.O. Box 100A-1400 Vienna, Austria

DEFINING INITIATING EVENTS FOR PURPOSES OFPROBABILISTIC SAFETY ASSESSMENT

IAEA, VIENNA, 1993IAEA-TECDOC-719

ISSN 1011-4289Printed by the IAEA in Austria

September 1993


5/152

FOREWORD

Safe operation of nuclear power plants is achieved by good design and prudentoperational practice. Both these aspects should be considered in the safety assessment of anuclear power plant. A plant specific probabilistic safety assessment (PS A) models a plantby addressing both design and operation and, in addition, takes account of operatingexperience.PSA can be used to quantify the safety of a nuclear power plant and it has become awidely used tool for the safety assessment of nuclear plants. Although its methodology isfairly standard, the complexity of the full scope plant specific PSA is such that guidance isstill needed to ensure its completeness. W ithin the framework of its programme on the

international p romotion of PSA, the IAEA ha s issued numerous documents on topics rangingfrom guidelines on how to perform P S A s to specific case studies and generic reliabilitydatabases.The PSA model is structured so as to permit consideration of a spectrum of possibledisturbances to a plant 's normal operation (in PSA terminology, initiating events), th eavailability of systems designed to cope with a particular disturbance and operator actions inthe course of the event. One of the areas where the level of completeness and the accuracyof the analysis could greatly influence a PSA model is the selection of the initiating events(IB s). The IAEA has developed this document in order to summarize and explain thedifferent approaches to the selection of IBs. It is based on examples taken from several P S Astudies.This document is primarily directed towards technical staff involved in the performanceor review of plant specific P S A s . It highlights different approaches and provides typicalexamples useful fo r defining the IBs. The document also includes the generic initiating eventdatabase, containing about 300 records taken from about 30 plant specific P S A s . In additionto its usefulness during the actual performance of a P S A , the generic IE database is of theutmost importance fo r peer reviews of PSAs, such as the IAEA's International Peer ReviewService ( I P E R S ) where reference to studies on similar N P P s i s needed.In the preparation of this document, the IAEA received support from several M emberStates in the form of material to be included. The main author of the document wasD . Ilberg from Israel, assisted b y B . Linquist from Sweden and J. Pereq from Israel.A. Bareith from Hungary wrote th e section on W W E R reactors. T he document wasreviewed by specialists from France, Germany, Hungary and the United Kingdom and byIAEA staff. T he IAEA project officer responsible fo r this report w a s B . Tomic from th eSafety Assessment Section of the Division of Nuclear Safety.


6/152

EDITORIAL N O T EIn preparing this document fo r press, s t a f f of the IAEA have made up the pages from theoriginal manuscript ( s ) . The views expressed do not necessarily reflect those of the governments of thenominating Member States or of the nominating organizations.The use o f part icular designations of countries or territories does not imply any judgement byth e publ isher, th e IAEA, as to the legal status of such coun tries or territories, of their au thorit ies andinstitutions or of the delimitation of their boundaries.The mention of names of specific companies or products (whether or not indicated as registered)does not imply any intention to infringe proprietary rights, nor should it be construed as anendorsement or recommendation on the par t of th e IAEA.


7/152

CONTENTS

1 . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71.1. Purpose of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71.1.1. Purpose of defining th e initiating events . . . . . . . . . . . . . . . . . . 71.1.2. Using this document for the evaluation of operational occurrences . . 81.2. Scope of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.3. Structure of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82. HISTORICAL B A C K G R O U N D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0

2.1. Development of data on transient initiating events . . . . . . . . . . . . . . . . . 1 02.2. Development of data on L O G A initiating events . . . . . . . . . . . . . . . . . . 1 32.3. Development of data on common cause initiators . . . . . . . . . . . . . . . . . 1 4

3. SELECTION OF I N I T I A T O R S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.1. Definition of initiating events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 83.2. Main categories of initiating events . . . . . . . . . . . . . . . . . . . . . . . . . . 1 83.3. Methodology for the identification of initiating events . . . . . . . . . . . . . . 1 93.3.1. Identification of LOCAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 93.3.2. Identification of transients and special initiating events . . . . . . . . . 203.4. Examples of initiating event lists . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.4.1. Initiating event lists for P W R s . . . . . . . . . . . . . . . . . . . . . . . . 27

3.4.2. Initiating event lists for B WRs . . . . . . . . . . . . . . . . . . . . . . . . 4 03.5. Completeness of IE lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 04 . G R O U P I N G O F INITIATING E V E N T S . . . . . . . . . . . . . . . . . . . . . . . . . 5 0

4.1. General principles for grouping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 04.2. Grouping o f initiating events f o r P W R s . . . . . . . . . . . . . . . . . . . . . . . 5 04.2.1. LOCAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504.2.2. Transients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 14.2.3. Special common cause initiators . . . . . . . . . . . . . . . . . . . . . . . 5 24.3. Grouping of initiating events for B W R s . . . . . . . . . . . . . . . . . . . . . . . 594.3.1. LOCAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594.3.2. Transients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 94.3.3. Special common cause initiators . . . . . . . . . . . . . . . . . . . . . . . 6 44 .4 . Impact of various transients on core damage f r e q u e n c y . . . . . . . . . . . . . . 6 4

5. DETERMINATION O F T H E F R E Q U E N C Y O F INITIATORS . . . . . . . . . . . 6 65.1. Approaches to quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 65.1.1. One and two stage Bayesian updating analyses . . . . . . . . . . . . . . 6 85.1.2. Mean frequencies of f r e q u e n t operational occurrences . . . . . . . . . . 6 85.1.3. Expert opinion on rare events . . . . . . . . . . . . . . . . . . . . . . . . 6 85.1.4. Frequency estimation by evaluation of failure rates andmission times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 8


8/152

5.1.5. Fault tree analysis f or special rare events . . . . . . . . . . . . . . . . . 6 95.1.6. U s e o f experience from other plants . . . . . . . . . . . . . . . . . . . . . 6 95.1.7. Special attributes of plant a n d location . . . . . . . . . . . . . . . . . . . 6 95.2. Some examples of IE frequencies in P S A . . . . . . . . . . . . . . . . . . . . . . 7 16 . EXAMPLES OF F R E Q U E N C Y D E T E R M IN A TIO N . . . . . . . . . . . . . . . . . . 74

6.1. Example of the use of the mean frequencies approach . . . . . . . . . . . . . . 7 46.2. Examples of determination of the frequency of common cause initiators ... 756.3. Examples of determination of the frequency of L O G A s . . . . . . . . . . . . . 7 76.3.1. M a in e Y a n k e e P S A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 76.3.2. Interfacing L O G A frequency based on pipe and valve rupture/leakage failure rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 76.4. Examples of determination of the frequency of transients . . . . . . . . . . . . 7 86.4.1. M a in e Y a n k e e P S A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 86 . 4 . 2 . G e r m a n Risk Study (GRS) P S A phase B . . . . . . . . . . . . . . . . . . 8 06 . 5 . Events in shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 07. IN IT IA T IN G E V E N T S FOR W W E R P L A N T S . . . . . . . . . . . . . . . . . . . . . 83

7.1. Selection o f W W E R initiating events . . . . . . . . . . . . . . . . . . . . . . . . . 8 47.1.1. Selection o f L O C A s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 17.1.2. Selection of transients and common cause initiators . . . . . . . . . . . 9 27.2. Grouping o f W W E R initiating events . . . . . . . . . . . . . . . . . . . . . . . . 9 67.2.1. Grouping o f L O C A s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 67 . 2 . 2 . Grouping of transients a nd common cause initiators . . . . . . . . . . . 9 97 . 3 . Determination of the frequency of initiators f or W W E R plants . . . . . . . . . 9 98 . H A Z A R D S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 5

8.1. M a i n categories of hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 58 . 2 . Evaluation o f hazard frequencies . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 58 . 2 . 1 . Seismic analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 58.2.2. Fire analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 78.2.3. Flood analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 79 . D E S C R I P T I O N O F T H E D A T A B A S E O F INITIATING E V E N T S . . . . . . . . . 1 1 2

9.1. General definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 29.2. Summary of database status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 3APPENDIX: PRINTOUT OF THE DATABASE ON GENERICIN IT IA T IN G E V E N T S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146L I S T O F A B B R E V I A T I O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 4 9


9/152

1. INTRODUCTION

S everal reports have been prepared [1] within the framework of the I AE A's programmeto provide Comprehensive Guidelines fo r Conducting Probabilistic Safety Assessment (PSA).The selection, grouping and frequency evaluation of initiating events (IBs) is one of the mostimportant tasks to be accomplished during a P S A Level 1 study, and this document isintended to provide additional guidance on this issue.The importance of the determination of IBs has been shown by performing uncertaintyand sensitivity analyses of past P S A s and by some peer reviews of P S A studies. Theinclusion of additional relevant IBs for completeness or revision of the estimates of theirfrequencies may change the results of a PS A s tudy.This document is intended to aid in the conduct and review of P S A s in M ember Statesby providing reference information that can help the specialist in defining IBs. It providesguidance by describing available methods and providing tables that compare approaches usedin various past PSAs and a database of IE lists and frequencies taken from many PSAs whichare available in the open literature.

1.1. P U R P O S E OF THE R E P O R TT he determination of initiating events is an important part of a PS A. IBs directly affectthe core damage frequency in P S As. They are also a class of operating occurrences reportedat nuclear power plants worldwide. Processing of these occurrence reports on a plantspecific basis provides information on the actual operating experience of that plant. Thus,the information provided in this document allows comparisons of plant specific operatingexperience with ranges and frequencies of IB s considered in P S A studies of similar types ofplant.This report is intended to provide the reader with a comprehensive overview ofdifferent approaches to defining IB s which have been taken in a number of PSA studies. Itdoes not recommend the use of any specific approach. It is left to those performing the PS Ato select the approach which is the most appropriate for their particular application, and then,if necessary, to acquire more information from relevant literature listed in the references.

1.1.1. Purpose of defining the initiating eventsAn IE is a postulated event that could occur in a nuclear power plant. It is anoccurrence that creates a disturbance in a plant and has the potential to lead to core damage,

depending on the successful operation or failure of the various mitigating systems in theplant.The performance of P S A s to develop a comprehensive plant model requires ascomplete a list of IBs as possible. This list determines the points of departure of the accidentsequences that would be studied in the search for the dominant sequences that may lead tocore damage. Thus the frequency of IBs has a direct impact on the results fo r core damagefrequency, as well as on the spectrum of importance of individual components or actions.T he consequences of ill defined IBs are various. A missing I E i n a P S A means thatth e core damage frequency would be underestimated by the value of the IE frequency

multiplied by the conditional probability of safety system failure given th e occurrence of the


10/152

IE. A larger list of IBs than necessary (for example, due to inappropriate grouping) wouldresult in waste of resources because of the analyses of additional unnecessary accidentsequences. An IE list that is incomplete or is insufficiently precise in its frequencydetermination would generally result in an incorrect estimation of the core damagefrequencies.1.1.2. Using this document for the evaluation of operational occurrences

This document can assist in evaluating operational occurrences in three ways:(a) It describes approaches used for identifying occurrences and assigning them to knownor new basic initiating events, or broader categories of initiating events.(b) It describes methods for evaluating IE frequencies from experiences of operatingoccurrences in a specific plant over a defined period of tune and provides examples ofthe application of these methods. It can be used for selecting generic prior distributionfor Bayesian updating of plant specific experience.(c) It provides an extensive database of data from other power plants to help put the plantspecific experience in perspective.1.2. SCOPE OF THE R E P O R T

This report covers the topics of defining IBs for nuclear power plants w i t h P W R a n dB W R type reactors ( inc luding W W E R s). It does no t c ov er C A N D U type reactors or oldertypes or types of relatively limited distribution.T he "Procedures for Conducting PSAs of Nuclear P o w e r Plants" [1 ] classify the IBs

into internal IBs and hazards 1 (internal and external haza rds). Internal IBs are hardw arefailures in the plant or maloperation of plant hardware through human errors or due toman-machine interface problems.External hazards (often called external events) are events originating outside the NPPthat create extreme environments common to several plant systems. Internal hazards, which

are originated within the station boundaries, create similar extreme environments, and includeinternal flooding, fire and missiles. Loss of connection to the grid (complete or partial) isconsidered here as an internal IE. The scope of the document is confined to internal events.However, an introductory section on hazards is given in Section 8 .1.3. STRUCTURE O F T H E R E P O R T

T he report is organized as follows:Section 2 gives th e historical background for defining IB s by considering various

approaches that have been used in previous P S A s . It covers the methods that form th e basisfor approaches described in Sections 3-5.1 'Hazards' is the term used in this report for all types of 'external events' such as fires, floods (inside an d

outside th e plant), earthquakes, etc.


11/152

Section 3 provides a review of selected approaches in some recent P S As and de scribesthe main me thods used to ensure completeness in IE selection. A discussion of completenessof the IE list is given and tables of several lists that have been used in the past are provided.Section 4 similarly deals with the approaches to IE grouping. A review of severalgrouping structures is provided by comparing several recent P S A approaches for B W R s and

PWRs. Lists of IE groupings are combined from some of these P S A s to provide the PSAspecialist with 'maximal' and 'minimal ' lists to compare. The relat ive importance of someof the IE groups used in various PS As i s shown for B WRs and P W R s respectively.

Section 5 covers the determination of IE frequencies. T he various methods aredescribed and some examples are given. S everal sources of data for the evaluation of IEfrequencies are outlined.Section 6 provides examples of derivations of the frequency of initiators taken fromseveral PSAs. It includes examples of estimations of the frequency of:

- L O C A initiators;- transients;- special common cause initiators;ATW S initiators.An example of the treatment of initiators for non-power operating modes is alsoprovided in Section 6 .Section 7 is entirely devoted to W W E R reactors. These reactors are treated separatelybecause the level of development of definition, grouping and frequency determination is, forthe time being, somewhat lower than f or P W R reactors operated in other countries mentioned

in this report.A brief introduction to the treatment of internal and external hazards is provided inSection 8 .Section 9 covers th e initiating event database. It describes th e structure, th e recordformat and the data sources used for the database.The Appendix provides the printout of the database of initiating events compiled by theI A E A .


12/152

2. HIST ORICAL B A C K G R O U N DThe concept of initiating events (IBs) w as introduced in the Uni t ed States Nuc l earRegulatory Commission's Reactor Safety Study ( R S S ) in 1975 [2 ] together with the event treemet hodology. Twenty transient IE categories and four L O C A related categories were selectedfor B W R s. Tw enty-three transient and six L O C A r el ate d categories were selected fo r P W R s .

T he basis for the selection of the L O CA IB s was the p lant response (based on systems/trainswhich are required to work and/or estimated timing of the accident). Consistent with plantresponse three sizes of L O C A s were selected:- L a r g e L O C A : 6 inches (15.2 cm) up to double ended largest pipe diameter denoted"A";- S m a l l L O C A : 2-6 inches (5.1-15.2 cm) equivalent diameter denoted SI;- V e r y small L O C A : 1/2-2 inches (1.2-5.1 cm) equivalent diameter denoted S2.

In addition, reactor pressure vessel (RP V ) rupture was considered for both P W R s andBWRs.

F or P W R s tw o addi t i onal LO CA related IBs were selected:Interfacing L O C A ( k n o w n a s t h e R S S 'event V );- Steam generator (SG) tube rup ture .T he definition for the var i ous LO CA cat egor i es w a s used in the event tree analysiswithout further grouping. Not so for the transients. A general division into anticipated a ndunanticipated transients was made. Almost a ll anticipated transients were modelled as onegroup i n b o t h P W R a n d B W R e v e n t trees. Similarly, unanticipated transients were modelledas a second group. An additional special initiating event was considered, namely the loss of

off-site power for more than 30 minutes.2.1. D E V E L O P M E N T O F D A T A O N T R A N S I E N T I N IT IA T IN G E V E N T S

Following the use of operating experience in the R S S , th e Electric P o w e r ResearchInsti tute (EPRI) published in 1978 its first study of anticipated transients [3 ] referred to asN P - 8 0 1 . N P - 8 0 1 i n c l u d e s a compilation of operational occurrences from 1 2 B W R s and 30P W R s . F or B W R s it reported 459 events in 37 selected categories of different transient IB s(compared to 20 in the R S S ) and for P W R s it reported 1000 events categorized in 41 differentIBs (an expansion of the 23 found in the R S S ) . T he data in the N P - 8 0 1 c o v e r e d N P Pexperience up to 1978, which was the equivalent o f 4 1 a n d 13 1 plant-years for BWRs a ndPWRs respectively. T h e N P - 8 0 1 d a t a w a s used extensively in probabilistic risk assessment( P R A ) p e r fo rm e d in the period 1978-1983 when the update of the N P - 8 0 1 report, th eNP-2230 [4], became available.

T h e N P - 8 0 1 h a s many drawbacks that th e N P - 2 2 3 0 h a s treated partially and the laterupdate of transients event data b y E G & G [ 5 ] further improved:(a ) N P - 8 0 1 used an 'effective in-service date ' as supplied by the ut il it ies. N P -2230u ni f or m ly used th e first d a y o f commercia l operation as the starting point for reportingplant anticipated transients. B ecause of this change, 1 3 7 e v e n t s o f N P - 8 0 1 w e r e

excluded from th e N P - 2 2 3 0 update .1 0


13/152

(b) NP-801 reports 191 events within 37 plant-years that occurred in the years subsequentto the first year of plant operation (less than half the total number of events). NP-2230reports 647 events in 85.5 plant-years (70%of the total).It is clear that the N P - 8 0 1 i n c l u d e d very early periods of plant operation, i.e. fromcriticality to commercial operation, whereas the NP-2230 included events that occurred onlyaf ter commercial operation was initiated. In general this is about half a y ear later. Table 2.1

gives a comparison of the evaluation of selected initiator frequencies based on these two datasources [6].T he NP-2230 reported 903 events for BWRs a nd 2093 events for PWRs under th e sameIE categories as in NP-801. T he number of plants covered increased to 16 BWRs and 36

PWRs with 101.5pl ant- y ears and 21 3 plant-years for B W R s a n d P W R s respectively. Thisdata was used in PRAs after 1983.

T he EG&G study [5 ] updated further th e transient IE database f o r B W R a n d P W Rplants. It included for the same IE categories 2 51 B W R plant-years with 1832events and 423PWR plant-years with 3574 events. There most of the events come from the years fol lowingth e first two y ears of plant operation. This database was selectively used in the N U R E G - 1 1 5 0[7] type PRAs performed in the period 1986-1989.

T he f irst four columns of Table 2.1 show original BWR-PRA estimates, based o n N P -8 01 [3]. The next four col u mns represent results obtained from applying the samemethodol ogy to the more recent data source (NP-2230). The two last columns present resultsusing the u pdated source and the two stage Bayesian methodol ogy [8]. It can be seen thatmost of the increase in the B N L ini t ia tor frequencies derives from the updated experience ofB W R r e la te d events, rather than from the use of the Bayesian methodol ogy .

T he BWR-PRA [12] differentiated between the impact of failures during th e f irst y e a rof plant operation and failures occurring in later years. However, in the review [6] it wasargu ed that th e database used in N P -8 0 1 w a s n o t sufficiently refined for this purpose. Thelater update, given in NP-2230, showed that the impact of ignoring the first y e a r of plantoperating experience causes a reduction of about 20% in initiator frequencies (see last twocolumns of Table 2.1). The 'weighted average' approach utilized in the B W R - P R A weightedth e data from th e first year a s (1/35) and the data from subsequent ye a rs a s (34/35) (thisapproach does not consider ageing).

Table 2.1 shows that the nu mb er of shutdowns due to anticipated transients is higherthan that experienced in recent years . This is apparently because the NP-2230 databaseextends to 1981 only. The updated review of the experimental data [6] published in 1985may show a reduction in the frequencies of IBs. The trend of reduction in transient IEf re que n cy continues today in m a n y power plants, and therefore the latest data sources shouldb e utilized whenever possible.

The EPRI reports were made available as part of the E P R I programme on anticipatedtransients without scram (ATWS). For this purpose they include a section for lEs that occurat a power level above 25% of full power. This is done because transients that occur at alower power level do not challenge the reactor shutdown system. This is further discussed inan example in Section 5.1 which treats frequency determination for ATWS.

1 1


14/152

T A B L E 2.1.C O M P A R I S O N O F S E L E C T E D IN IT IAT OR F R E Q U E N C I E S A N D S O U R C E S O F D I F F E R E N C E S

S N P S - P R A [12]E P R I - N P - 8 0 1 D a t a

Transient

Loss ofC ond enserV a c u u m (2,4,8)Turbine TripM S I V Closure (5 )Loss of FW (22)LOOP (31)I O R V (11)C R W ( 27 ,2 8 )TOTAL

1stY e a r

1.6

16.92.20.60.40.70.1

22.5

Subseq .Y e a r s

0.38

4.140.190.160.110.080.035.09

AllY e a r sAverage0.67

7.30.670.270.160.200.049.3

S N P S - P R AW e ighte dAverage*0.41

4.460.2401.80.08 +0.090.035.49

1stY e a r

1.0

13.41.670.270.130.530.1317.1

Subseq .Y e a r s

0.38

6.390.270.110.120.150.107.52

B N L Review: [6 ]EPRI-NP-2230 D a t a

A ilY e a r sAverage0.47

7.390.470.130.120.210.118 .9

W eig htedAverage

0.40

6.590.310.120.08 +0.160.107.76

B N L Review: [6 ]Two-StageBayesianSubseq .Y e a r s

0.40

6.850.290.110.120.190.118.07

AllYears**

0.50

7.890.570.130.15+ +0.250.129.65

N u m b e r s in parentheses correspond to E P R I NP-801 [3] categories+ Based on SN PS g rid data * U s e d in the PRA+ + Based on N S A C - 8 0 report [10] ** U s e d in th e BN L review


15/152

T he validity of the N P - 2 2 3 0 [4 ] data w as reviewed in two cases:(a ) In the Oconee PRA [9] the data o f N P - 2 2 3 0 w a s checked against the licensee eventreport (LER)data of the D u k e P o w e r Co., the owner of the plant. It was found thatin most of the cases a good agreement exists. O n l y in the division betwe en 'partial lossof M F W a n d 't u rb i n e trips', significant differences were found. M a n y more events

were categorized as turbine trips in the O c o n e e PRA than as the partial loss of MFW.This w as based on the in-plant data records, which was considered more accurate thanth e E P R I data.

(b) A thorough comparison was performed in the E G & G study [5] of the N P - 2 2 3 0 databaseand the 'NRC G r a y B o o k ' [11] database for 11 plants which were selected for thecomparison. For each plant selected, the events that occurred during the third andeighth year of operation were carefully compa red. It wa s found that 66 (27%)eventswere categorized differently based on the event description in each of the two sourcescompared. However, about one third of the discrepancies were because the Gray Bookevent description contained less information than th e N P - 2 2 3 0 description. T he finalconclusion was that on the whole the NP-2230 data was found to be valid and isindicative of US commercial N P P experience. This is because the deviations weresmall and, in general, did not cross 'borders' of the broad groups of transients used inth e P R A studies. Another important conclusion wa s that a sufficient amount of detailsin th e event descriptions, provided by the plants, is crucial for a correct ca tegorization.

2.2. DEVELOPMENT OF DATA ON L O C A INITIATING EVENTSUnli ke transients, the categorizat ion of L O C A IE categories has not much changed sinceth e original R S S defini tions. The main changes in de finitions were th e inclusion, in most of

the newer PSAs, of the SG tube rupture (rather than the SG rupture in the RSS which wasnot further anal y s ed there) and a group of very small LOCAs at various locations (rather thancontrol rod drive (CRD) pump leakage in the RSS). Table 2.2 compares several frequenciesused in P S A for the same L O C A IE category. It should be noted that break size definitionsof various size LOCAs are not uniform in PSAs and d i f f er ent f or P W R s a nd BWRs inparticular.

The Reactor S af e t y S tu dy (RSS)estimated LOCA frequencies by inference from genericdata from pipe breaks in the non-nuclear industries. This is the basis for the mean val uesshown in Table 2.2 for the RSS. T he reactor pressure vessel (RPV) rupture probability w asalso based on non-nuclear vessel experience. W hile the frequ ency of the latter did not changemuc h in newer PRAs (most of them still use t he RS S value), t he LO CA frequenci es havebeen reevaluated in the newer PRAs (e.g. th e M i dland P R A [13]).

Oconee [9] and S eab rook [14] P R A s u s e d experiential data for the evaluation of partof th e L O C A frequencies rather than th e pipe break data used in the RSS. T h e O c o n e e P R Aconsidered the following events in a population of 35 plants:- L a r g e L O C A (A): No event occurred;S m a l l L O C A (S): One event that occurred at Zion Unit 1 in 1975;- SG rupture (R): Three events of SG tube ruptures with leakage ratesgreater than 100 gpm occurred: Surry Unit 2 (Nov.1972) , Point Beach Unit 1 (Feb. 1975), and Prairie

Island Unit 1 (Oct. 1979) .1 3


16/152

A two-stage Bayes ian analysis was applied to the above generic data andto the Oconee plant specific experience which reflects none of the aboveevents in any of the three units on-site. A revi ew of t he O conee P RA [15]added another relevant event:V e r y s m a l l L O C A (VS): O ne event that occurred at H.R. Robinson U nit 2( M a y 1975) .This has added a frequency of 3 x 1 0 ~ 3 (see Table 2.2). T h e B / W owner group[17] based their estimate of 'VS' on the precursor study [55] which introduced th e 'Robinsonevent' mentioned above.T o summarize, th e development of frequency evaluat ion for RPV and large L O C A h as

not changed significantly since the RSS was made. On the other hand, v e ry small LOCAsand interfacing L O C A s received additional attention and some new studies have been madewhich are discussed further in Sections 5.1.4 and 6.3 which provide examples of the treatmentof the frequency of L O C A type IBs.2.3. D E V E L O P M E N T O F D AT A O N C O M M O N C AU S E I N I T I A T O R S

T he R S S h as already treated some IBs as special common cause initiators (CCIs). Tw oexamples are the loss of off-site power for more than 30 minutes and the 'V event. TheP S A s t h a t followed the RSS have added more CCIs, and in general the C C I s are of a plantspecific nature. S o m e of the CCIs treated in P S A s are:- Loss of instrument air;- L o s s of DCpower bus;- L o s s of service water or component cooling water system;- L o s s of ACpower bus(es);- Steam line break;- Reactor water level instrument line failure.

This section describes the treatment of one specific C C I common to all PSAs, namelyth e loss of off-site power initiator. O t h e r approaches to CCI evaluation are covered inSections 5.1.5 and 6.2.Loss of off-site p o w e r ( L O O P ) experiential data have been reviewed in four studiessince 1980:

(1 ) Scholl [24] reviewed the data received from licensees following a June 1980 N R Crequest to submit licensee experience with L O O P events. This review includes a listof 109 occurrences o f L O O P e v e n t s .(2) The results of a L O O P study were summarized in E P R I -N P - 2 3 0 1 [25] which uses datacollected from 4 7 nuclear power plant sites. The report presents frequency and durationof L O O P s based on 45 occurrences through April 1981, representing 375 plant-yearsof experience.(3 ) A N S A C /O R N L s tu dy w a s reported in N S A C -8 0 [10] which covered 52 nuclear powerplant sites, for the period prior t o December 1983. It summarizes 4 7 L O O P events in

530 plant-years.14


17/152

T A B L E 2.2. A C O M P A R IS O N O F L O C A F R E Q U E N C I E S IN V A R I O U S P R A s [16]LOCA TypeInitiator (*)

PRAARKANSASIREP f!8lMIDLANDPRA f!31B/W OwnerGroup T171OGONEEPRA f9lNUREG-4550PWR [401NUREG-4550BWR 1361LIMERICKPRA 1201SHOREHAMPRA F121BWR-6 1211SEABROOKPRA f!4lRSS-PWR

F 2 1RSS-BWRf 2 lPALUEL 1300PSA T231German RiskStudv F221

Very Small SmallLOCA LOCA (**)(


18/152

(4 ) An U S N R C s t u d y [26] for the resolution of the ' Station Blackout ' issue was reportedin N U R E G - 1 0 3 2 . The study covered 5 2 N P P s (all th e U S N P P sites of D e c e m b e r 1983excluding three with on e off-site power connection). It summarizes 55 events in 533plant-years.T he review of the data sources has resulted in several findings:

(a) The S choll database is rather conservative and needs additional evaluations prior to itsutil ization in PRAs.(b) The N P - 2 3 0 1 d a ta b a se is more realistic. A few events are apparently missing from thissource. Its recovery probability information is relatively conservative for use in P R A s .( c) T h e N S A C - 8 0 database appears to be suitable for realistic P R A analyses. Itrecommends exclusion of several total L O O P occurrences during shutdown which it

judges to be 'impossible' during operation. It can however be assumed that these areinadvertent human errors that should be included i n L O O P frequency evaluation forcompleteness. T he later N U R E G - 1 0 3 2 considered them in its statistics.The above three studies reported the L O O P events by plant and per geographicalregions having similar weather conditions and an interconnection agreement with respect tokeeping a reliable electric supply in that region. Another approach was proposed by a laterstudy:

(d ) T h e N U R E G - 1 0 3 2 d a ta [26] are based on almost the same database a s t h e N S A C - 8 0 .It includes th e 'shutdown' events as well. T he main improvement of this study is thatit provides a breakdown of all the L O O P events into well defined causes which allowstailoring of the L O O P frequency of a new plant according to its design and also allowsfor evaluating th e improvement that m a y b e expected by a design change in an olderplant (see Section 5.1.7 for more details on this approach).All these data sources and a pproaches have been used by P S As in the USA. T he L O O Pis considered a CCI when the conditions of no electric power are continuing for a long timeperiod (n o recovery of off-site power). W h e n off-site power is recovered within a short time,it is considered a transient similar to loss of condenser, because condenser cooling would belost in this case. Therefore, also data on recovery tim e s o f L O O P a r e part of the informationneeded fo r quantification of i ts CCI frequency. T he data for recovery times i n t he U S A havealso been developed in parallel to the L O O P e v e n t data discussed above. N U R E G / C R - 5 0 3 2[27] is the more recent one for the USA.One of the first works providing L O O P recovery data was the E P R I study [25]. T heN S A C / O R N L study [10] provided more precise recovery information on events i n t he US Aunti l D ec em ber 1983. T he N U R E G - 1 0 3 2 [26] study divided the N S A C / O R N L [10]d a t a ( w i t hsmall modifications) into three subgroups.

- S e v e r e w e a t h e r t y p e L O O P ;Grid related L O O P ;- Pl ant centred (hardware failure r e l a t e d ) L O O P .

16


19/152

F or each of these subgroups th e study provides a frequency of occurrence versusduration plot. T he p la n t c e n tr e d LOOP occurrences are the most frequent cases but most ofthem are recovered within half a n hour [27]. Therefore, they may be treated like internalevents. The grid and weather related L O O P have lower frequency and larger duration andare therefore treated more like th e special C C I group of IBs.

1 7


20/152

3. SELECTION OF INITIATORS

3.1. D E F I N I T I O N O F I N I T I A T I N G E V E N T SAn initiating event (IE) is one that creates a disturbance in the plant and has thepotential to lead to core damage , considering successful operation or not of the various

mitigating systems in the plant. T he following definition of I E has been taken from R e f s [1 ]and [28]."A n initiating event is an incident that requires an automatic or operator initiatedaction to bring the plant into a safe and steady-state condition, where in theabsence of such a ction the core dam age states of concern ca n result in severe coredamage. Initiating events are usually categorized in divisions of internal andexternal initiators reflecting the origin of the events".This section deals only with internal events. Internal hazards (such as internal f loods)and external hazards (such as seismic events) are covered in Section 7. The initiating events

considered here are only those normally resulting in an automatic or manual scram andoccurring above a certain power level (usually 5 to 25%).3.2. M A IN C A TE G O R IE S O F I N I T I A T I N G E V E N T S

T he internal initiating events may be looked upon as consisting of three m ain ca tegories:(A ) L O C A s ;(B ) Transients;(C) S pecial common cause initiating events (common cause initiators (CCIs)).

S o m e of the recent P S A s (for example the Pa lue l PSA [23]) devoted substantial effortto th e events during plant shutdown. A brief introduction to these specific events is providedin Section 6.5.(A) LOCAs

T he loss of coolant accident ( L O C A ) initiators include prima ry system breaks result ingin loss of primary coolant. Pipe breaks and ruptures of dif ferent sizes, inadvertent openingand fai lures to re-close (stuck open) of val ves are being considered in this category.(B ) Transients

T he transient initiating events are those which introduce the disturbance in normal plantoperation, without loss of primary coolant and which require an automatic or ma n ua lshutdown of the reactor. Typical examples of transient initiators include disturbance infeedwater f low, turbine/condenser, reactivity control, reactor recirculation, etc. Certaindisturbances in some of the support systems will also fall into this category.(C ) Special common cause initiating events

Special initiating events are events which, in addition to requiring reactor shutdown,simultaneously disable one or more of the mitigating systems required to control the plantstatus following th e initiator. Typically, they are unique to the plant being analysed.1 8


21/152

Initiating events which are common to most plants and have a typical CCI character areevents such as:- L o s s of off-site power;- L o s s of DCpower;- Spurious containment isolation;- Loss of instrument air;- L o s s of component cooling.

Even t hough some of t he LO CAs may have CCI character (damage to equipment dueto pipe whip, environmental influence (e.g. high temperature or humidity) on safety systems),LO CAs are not normally looked upon as CCIs.3.3 M E T H O D O L O G Y F O R T H E I D E N T I F I C A T I O N O F I N I T I A T I N G E V E N T S3.3.1. Identification of L O C As

Loss of coolant accidents (LO CAs) are usually identified by studying the primary andsecondary system equipment, piping and valve arrangement. Due to the different sizes ofpiping and valves as well as different mitigating systems available, a n umbe r of L O C A s br e a ksizes are normally assumed. T he specific boundaries of the L O C A event categories for eachplant are established on the bases of overall mitigating system performance requirements (o rsafety system success criteria) for various cases of LO CAs break s i zes .The mitigating systems usually provide protection for different sizes of steam versusliquid breaks. Therefore, different boundaries may be used for small and medium L O C A s ofl iquid and steam l ine breaks (for B W R s ) .In general th e categories used f or L O C A initiators i n P S A reflect small variation amongpast PSAs. There are in use between three to five basic size categories taken from th efol lowing list:

- V e r y very small L O C A (e.g. less than 1 inch equivalent inside diameter);- V e r y small L O C A (e.g. 1-2 inches equivalent inside diameter);- Smal l L O C A ( e . g . 2-3 inches equivalent inside diameter);- M e d iu m L O C A (e.g. 3-6 inches equivalent inside diameter);- L a rg e L O C A (e.g. greater than 6 inches equivalent inside diameter) .Differentiat ion between hot leg and cold le g bre a k locations are generally ignored in

P S A studies.In addition, the following L O C A initiators are also being considered in most P S A s :

- R eactor pressure vessel rupture;- L O C A outside containment (loss of coolant accident via the interface of low pressuresystem with R C S . In this case, in addition to unrecoverable loss of coolant, thecontainment is bypassed);- Steam generator t ube rupture (P W Rs only).In several cases of steam generator tube rupture only one tube rupture is considered.

E v e n though this is a very small L O C A the plant response is in general different from th e1 9


22/152

very s m a l l L O C A case (due to filling of affected S G a n d eventually overpressurizing it) and,in addit ion, a path to bypass containment is created in this case, which makes this initiatoru niq u e .S o m e P S A s consider specifically th e following initiators:

- Reactor coolant p u m p seal leakage or failure (in P W R s ) ;- Control rod drive system leakage or failure (in B W R s ) .

In ot her P S As they are a part of the 'very small L O C A ' initiator presented above.3.3.2. Identification of transients and special initiating events

In order to obtain a comprehensive list of transients and special IBs, a number ofmethods a nd approaches have be e n use d in P S A s . A n o v e rv i e w of the m e t h o d s utilized isprovided in Tables 3.1 ( B W R ) a n d 3 .2 ( P W R ) . T h e t a b le s include th e following m a i nmethods:(A) E nginee ring eva luation or technical study of plant;( B) R e fe re n ce t o p re v i o us P S A s ;(C ) EPRI list o f I B s (such a s E P R I - N P - 2 2 3 0 or N U R E G / C R 3 8 6 2 ) ;(D ) Logic al cl ass if ica t ion: M L D , energy balance, barrier analysis;(E ) P l a n t e n e rg y balance fault tree;(F ) Analysis of oper at i ng experience for actual plant;(G ) F a i l u r e m o d e a nd ef fect analysis;(H ) O t h e r m e t h o d s.

These met hods a re described below.( A) E n g i n e e r i n g ev alu at i on or technical study of plant

In this approach, th e plant systems (operat ional as w ell a s safety) a nd m a j orcomponents a re systematical ly reviewed to see whether any of the fa i lure m o d e s (e.g. fa i lureto operate , spurious opera t ion, disrupt ion, collapse) could lead direct ly , or in combination withother fai lures, to significant disturbances of plant operat ion, r eq u i r i ng operat ion of mitigat ingsystems. Part ial fai lures of systems should also be considered since, al though they aregenerally less severe than complete failure, they are of higher frequency and are often lessreadily detected. T h i s is, in principle, s imilar to a classical eng i neer i ng a p p ro a ch t a ke n duringth e pl ant ' s design aimed towards providing safety systems to cope with the spectrum of designbasis accidents. For PSA purposes, the initiators which are beyond design basis ( for e xa mp le ,due to their low frequency of occurrence) are also taken into account.

In o rd e r to aid the engineering evaluat ion i t may be helpful to structurize th e init ia t ingevents by m e a n s of categorizat ion trees , a s h a s be e n d o n e i n t h e Swe d i sh projectS U P E R - A S A R [29]. T he basis for this structurization is the availability of main functionssuch as the primary system integrity, off-site power, feedwater and condensation in the maincondenser.

This approach was at least partially utilized in m a n y P S A s , and it is an indispensablemethod in identifying special , c o m m o n cause IB s wh i ch a re u ni q u e to a p a r t icul a r plant .

20


23/152

T A B L E 3 . 1 A P P R O A C H E S U S E D F O R T H E S E L E C T I O N O F lE s IN V A R I O U S P S A s ( B W R s )Method for Engineering Refe rence EPRI ListNPP Identifying Evaluation to previous of IBsPSA lEs: or Technical PSAs (such asStudy of Plant EPRI-NP-2230)

LogicalClassification:MLD, EnergyAnalysis ofOperatingExperienceBalance, Barrier for ActualAnalysis Plant

OtherIdentifi-cationMethods

Peach Bottom, [19]Unit 2(-89)Grand Gulf, [36]Unit K-89)Ringhals 1 [29]Barsebaeck 1 [29]Oskarshamn 1 [29]Oskarshamn 2 [29]Forsmark 3 [29]Forsmark 1/2 [29]Shoreham [12]Browns Ferry [30]K-82)Limerick 1 [20]Big Rock Point[46]* For CCI only

GE-NEDO 2A708AASEP ListGE-NEDO 24708AASEP List

The Swedish Energy Conn.F3 - Study

FSAR


24/152

toto T A B L E 3.2. A P P R O A C H E S U S E D F O R T H E S E L E C T I O N OF IBs IN V A R I O U S P S A s (PWR s)Method for Engineering ReferenceNPP Identifying Evaluation to previousPSA IBs: or Technical PSAsStudy of Plant

EPRI Listof IBs(such asEPRI-NP-2230)

LogicalClassification:MLD, EnergyBalance, BarrierAnalysis

Analysis ofOperatingExperiencefor ActualPlant

OtherIdentificationMethods

Ringhals 2 [29] +Oconee NSAC [9] +Seabrook (-83)[14]Midland [13]Surry, Unit 1 (-86) [40]

Galvert Cliffs (-84) [41]Arkansas 1(-82)[18]Sizewell B(-82) +*[50]

FMEAFSAR

List of SubtleInteractions (SNL)ASEP ListFMEAFMEAFSARRESAR

German Risk [35]Study (Phase A)German Risk [22]Study (Phase B)Zion [44]

Maine Yankee [45]Yankee Rowe [31]Connecticut Yankee(56)

FSAR & IndustryExperienceFMEA

FMEA

For CCI


25/152

(B ) R e f e r e n c e to previous P S A sA large n u m b e r of P S A s as well as P S A revi ews are available today. It is very usefulto refer to the lists of IBs presented in these PSAs, especially those for similar reactors.Reference to available IE lists m ay serve as the starting point in compiling a list of plantspecific IBs. This approach was utilized in many P S A s .

(C ) EPRI l i s t of IBsThe lists of IE categories for B W R s and PW R s provided in E P R I - N P - 2 2 3 0 [4] andN U R E G / C R - 3 8 6 2 [5] was used as the starting point for IE selection in a large number ofP S A s . T he lists are derived by analyzing operating experience of a few hundred reactor-yearsin the U S A . Therefore, th e lists can be considered as one of the best sources fo r providinga generic IE list for P S A of a new plant of similar design and reasonably similar operatingpractice.

(D ) M a s te r logic diagramThe so-called ma ster logic diagram ( M L D ) is similar to a fault tree. It presents a m o d e l

of a plant in terms of individual events and their combinations. A typical top event may bethe 'significant release of radioactive material ' . It develops into a plant level logic structurewhose basic input events are the initiating events. O ne example o f a n M L D is shown inFig. 3.1 taken from R ef [7]. The initiating events are the lowest level of the tree.The particular advantage of the M L D method is that the issue of completeness is putinto a more tangible perspective compared to other methods. H owe ver, in the cumbersometask of listing the specific causes of initiating events, the possibility of incompleteness stillexists. Examples o f the use o f M L D f or selection of IBs can be found in S eabrook and

Y a n k e e Rowe PSAs [14, 31].(E ) Pl ant energy balance fault tree ( E B F T )

A transient condition in a nuclear reactor implies an imbalance in the state ofequilibrium in the transfer of thermal energy from the reactor core to the environment.Therefore, a fault tree analysis of the plant energy balance (sometimes referred to as heatbalance) is a valuable tool for deriving initiating events. T he energy imbalance fault treeanalysis m a y b e carried out for any degradation in thermal equil ibrium including those fromfull power, partial power, hot s tandby a nd cold shutdown. The typical top event is 'imbalancein energy transfer causing a plant 's IE to occur'. One example of an energy balance fault treeis shown in Fig. 3.2. As for the M L D , the basic input events derived from the analysis arethe initiating events at the equipment level.

Relative t o t he M L D i t i s unnecessary to include additional system failures in this typeof faul t tree. There is no need to introduce associated simplifications in the logic and, hence,the deductive power of fault tree analysis need not to be diminished.(F ) Analysis of operating experience for actual plant

In this approach the operational history of the plant in question and of similar plantsis reviewed to search for any events which should be added to the list of IBs. This approach

23


26/152

tL A A G CLOO

3MfOiuM1003SHALLLOCA

4LfAKAGCtoSCCOftOAAVCOOLAKT

fLOB OFPHIMA8YCOO LAN!FLOW

LOS5OFFtfOFLOW

7lOS OFSTEAMFLOW

mtifiNM.INITIATOAS

IHtTl TORS

lHi*NALtINITIATORS

ihllAOKS

tMKKNAVINITIATORS

*

IKIT1ATOH&

IhTtRhA.IHTlATOftS

*

IKtTtATORS

mTEAHALINITIATORS

INtTlATOHJ

INYCnNALINITIATORS

INITIATORS

1NT E PIN XVINT TORS

iNTI TORS

INTERNALINITIATORS

INITIATORS

INTERNALINITIATORS

*

INITIATOR4*

INTERNALINITIATORS*

INITIATORS

INTERNALINITIATORS

INITI TORS

INITIATORS

INITI T O R S

INITIATORS

INITIATORS

INITIATORS*

INITIATORS

INITIATORS

INITIATORS

FIG. 3.1.Example of a master logic diagram (MLD).


27/152

L E V E L 1

L E V E L 21ENERGYIMBALANCEBETWEENCORE RCS

*

ENERGYIMBALANCEBETWEENRCS &SCS

ENERGYIMBALANCEBETWEENSCS&PLANT UTPUT

L E V E L S CHAN G E I N COREHEAT GEN ERATIONCHAN G E I N RCSHEAT REM OV ALF R O M R C S

L E V E L 41

I N CREASE I NCORE HEATGENERATION

1DECREASE INC O R E HEATGENERATION

L E V E L 5

Reactor TripL E V E L 6 Manual ShutdownControl Rod Drop

IINCREASE NRCS HEATREMOVALFROM CORE

1DECREASEDRCS EATREMOVALFROM CORE

INWARDM O V E M E N T O FCON T ROL ROD SD ECREASE I NN EU T RONMODERATION

INCREASE INCOOLAN T BORONCONCENTRATION

RCS - Reactor Coolant SystemS C S - Secondary Coolant System

N one CV CS Malfunction

toL/i FIG. 3.2. Energy balance fault tree.


28/152

is to be looked upon as supplementary and would not, of course, be expected to reveal lowfrequency events but could show common cause initiating events of higher frequency. Suc han approach is used also in precursor studies [32] to identify the IBs and to estimate th ecorresponding frequency.(G) Fai lure mode and effect analysis

Failure m o d e a n d ef fect analysis ( F M E A ) is a powerful technique used extensively b yth e aerospace industry to reveal component failures that ca n impose critical effects on systemperformance. Dealing at the component level it becomes very detailed and laborious.Therefore , it is not commonl y used in PSA.S o m e P S A s have used this methodology to identify IB s within the plant control system sthat have th e potential to fail a mitigating system due to dependencies [31].

(H) O t h e r methods

T he most commonly used methods or approaches to selecting IB s were covered in theprevious subsections. In special cases some other plant specific methods w e r e used. O ne ofthese i s t h e N U R E G - 1 1 5 0 a p p r o a c h that has created a generic list of IBs in a special studyand this IE list (ASEP l ist) was used as a generic list of IBs in all "1150" P S A s [33].Another special approach is the Sw e dish S U P E R - A S A R [29] project approach whichis presented in Section 3.4.2.

3.4. EXAMPLES OF INITIATING EVENT LISTSSection 3.3.2 presented several methods for identifying transient a nd special IBs. T w o

of the methods referred to give lists of IBs from other studies. This Section providesexamples o f t h e E P R I lists (see Section 3.3.2.C) and three specific lists that were used inprevious P S A s (two for P W R s and one for B W R s ) :F o r P W R s :

(A) The EPRI list (considers transients only);(B) The O conee list;(C) The G e r m a n Ri sk S t udy (P hase B ) list.F or B W R s :

(A) The E P R I l is t (considers transients only);(B ) The Grand Gulf list.A s stated before, th e E P R I lists for P W R s a nd B W R s a re we l l known and are referredto by a number of PSAs, especial ly US studies.T he P S A s for Oc onee and Grand Gulf provide IE lists that are partly based o n th e E P R Ilist, but in addition considerable work has been done using a few of the other met hodsdiscussed in Section 3.3.2 in order to find any additional IE which should be considered.

26


29/152

3.4.1. Initiating event lists for P W R s(A) The EPRI list

The EPRI list of t ransient initiating events [4] is an expansion of the list prese nted inth e RSS. T he expansion is based o n U S reactor operating experience data. T he data fo rP W R s includes 2023 events occurring over 21 3 plant-yea rs at 36 plants. The E P R I P W R listis given in Tabl e 3.3. T he n umbe r associated with each initiator is referred to in Section 4 .(B) The Oconee list

T he selection of the lEs used in the Oc onee PSA [9] was performed in several steps.Initially, candidate IBs were proposed, and engineering evaluation performed for each IE .The IE list generated from those two activities was compared to IE lists available from E P R I ,R S S , and the R S S M AP study for t he Oc onee N P P [34]. T h e L O C A events and theirrespective sizes were selected on the basis of mitigating system requirements. Finally, plantsystems were reviewed to determine relevant special common cause initiators.The resulting list of initiating events for Oconee is given in Tables 3.4-3.6. Table 3.6shows the categorizations of the Oconee initiators and includes comments from the initiatorselection process, which should clarify how a particular initiator has been treated. Table 3.6indicates th e final category (based on mitigation requirements) under which a part icularinitiator ha s been considered in an initiating event broad group that w as further treated in anevent t ree a nalysis.

(C) The G erman Risk Study listT h e G e r m a n Risk Study (GRS) transient IBs [22,35] are treated in two frequency

classes: anticipated transients that are derived from operating experience and unlikelytransients. T he unlikely transients mainly include cases of steam line related events whichhave large effects on the plant systems. In comparison to these effects, th e reactivity typetransients inflict very small impact on the plant systems, and, therefore, they are notconsidered in phase B of the GRS. The list of transient initiators in the GRS are shown inTable 3.7. It should be noted that in many cases the anticipated transients considered in theGRS lead to the opening of the pressurizer relief valve and,in the case of its failure to close,a L O C A event occurs. These cases of L O C A are treated w ithin th e L O C A initiating events.

L O C A t y p e IBs are covered in the GRS in great detail as shown in Table 3.8. Itincludes L O C A s o f v a r i o u s sizes in the primary cooling system, in the pressurizer, L O C A sin th e annulus between the circular containment liner and containment shielding (because theycan impact on important additional systems by flooding) and steam generator tube ruptures.

F or cases of failure of the reactor scram, A T W S events are also treated in the GRS.

Text cont. on p. 40.27


30/152

T A B L E 3.3. L I S T O F P W R T R A N S I E N T I N I T IA T IN G E V E N T S

1. L o s s o f R C S flow (one loop)2. Uncontrolled rod withdrawal3. Probl ems with control-rod drive mechanism and/or rod drop4 . Leakage from control rods5. L e a k a g e in primary system6 . Low pressurizer pressure7. Pressurizer leakage8 . Hi gh pressurizer pressure9 . Inadvertent safety injection signal1 0. Containment pressure problems1 1 . C V C S malfunction-boron dilution1 2. Pressure, temperature, power imbalance-rod-position error1 3. Startup of inactive coolant pump1 4 . Total loss o f R C S flow1 5. L o s s or reduction in feedwater f low (one loop)1 6 . Total loss of feedwater flow (all loops)17. F ull or partial closure o f M S I V (one loop)1 8 . Closure of all MSIVs1 9 . Increase in feedwater flow (one loop)2 0. Increase in feedwater flow (all loops)2 1 . Feedwater flow instability-operator error22 . Feedwater flow instability-miscellaneous mechanical causes23. L o s s of condensate pumps (one loop)24. Loss of condensate p u m p s (all loops)25. L o s s of condenser vacuum26. Steam-generator leakage2 7 . Condenser leakage28 . M iscel laneous leakage in secondary system29 . Sudden opening of steam relief valves30. L o s s of circulating water31 . L o s s of component cooling32. Loss of service-water system33. Turbine trip, throttle valve closure, EHC prob l ems34 . Generator trip or generator-caused faults35. Loss of all off-site power36 . Pressurizer spray failure37. Loss of power to necessary plant systems38 . Spurious trips-cause unknown39 . Automatic trip-no transient condition4 0. M a nu a l trip-no transient condition4 1 . Fir e within plant

28


31/152

T A B L E 3.4. L O C A I N I T I A T O R S I N C L U D E D I N T H E O C O N E E P R A

E vent D escription

1 . S m a l l-b re a k L O C A

2 . L a r g e L O C A

3. Interfacing-system L O C A

4 . R P V r u p t u r e

5 . Steam generator tuberupture

A break or leak 1/2 to 4 inches in effective diameter.These are spontaneous events: i n d u c e d L O C A s weretreated directly.A break or rupture greater than 4 inches in effectivediameter except those noted below.A large loss of coolant through the valves acting as aboundary between high and low RCS pressure.A loss of reactor-vessel integrity precluding the ability tomaintain coolant inventory.A rupture of a steam generator tuberesulting i n an RCS leak greater than 1 00 gpm.

T A B L E 3.5. S P E C IA L I N IT IA T O R S I N C L U D E D I N T H E O C O N E E P R A

E v e n t Description

1 . L o s s of instrument-airsystem

2. Loss of service-watersystem

3. Loss of integrated andauxiliary control systems

4 . L o s s of DC power system

T he system may cause a reactor trip and afailure of instrumentation and equipment that may beneeded for a successful response to the trip.T he system may fail by a pipe break or pumpfailure and in addition prevent other safety systemoperation that depend on service water cooling supply.T he Integrated Central System (ICS) iscontrolling feedwater, pressurizer heaters etc., and maycause a transient with loss of a protecting system.B us 3TC failure during power operation may result infailure to supply power to a number of pumps in train Asupplying several mitigating systems.

29


32/152

T A B L E 3.6. C A T E G O R I Z A T I O N O F T H E O C O N E E IN IT IA T IN G E V E N T S I N T OT R A N S I E N T S AN D L O C A s

Potentialinitiating event Treatment in refinementprocess Final category

1 . R od drop

2. Inadvertentro d withdrawal

3. R od ejection

4 . Inadvertentboration ordilution

If shutdown is required, th e con-ditions after th e trip are thesame as those for the reactor tripcategory.It was determined that the reasonfo r ro d withdrawal would notaffect th e other reactor systems.IC S failures are treated elsewhere.Conditions after th e trip are thesame as for the reactor-trip cate-gory since the additional reactivitybefore th e trip would not signifi-cantly change boundary conditions.Detailed analysis not performed.Probabilistic consideration isincluded in the frequency of non-mitigatable vessel ruptures.Boration would force th e reactortoward shutdown. Credibledilutions would result in reacti-vity effects judged to be insigni-ficant with respect to mitigation.Dilutions resulting in substantialreactivity were judged to be ofsmall probability due to the timerequired and the amount of nonbor-ated water required. C o r e melt dueto a dilution accident w as judged tobe dominated by more frequent transients.

Reactor/turbine trip


E v e n t frequencyjudged to be verylow. Consequenceslimited to brief R C Soverpressure.Vessel rupture


5. React or trip Reactor/turbine trip

30


33/152

T A B L E 3.6. (cont.)


6 . Cold-wateraddition

7 . R C P t r i p

8 . R C P seizure

9 . Flow-channelblockage

1 0. Loss of mainfeedwater

1 1 . Excessfeedwater

The conditions at shutdown areno t significantly differentfrom those of a reactor trip.Four-pump trip assumed to bedominated by the loss of off-site power or the loss of ser-vice water.On ly four-pump seizure w as judgedto be significant with respect toreactor conditions. Simultaneousfour-pump seizure was judged to beof low probability and thereforewould not result in dominantaccident sequences.The potential for blockage w as exa-mined and found to be small due tothe presence of internals at the bottomof the vessel which wouldprevent large objects from block-in g flow channels. Smaller block-ages were j u dged to not affectdominant core-melt risks.Due to its gross effects on heatremoval through th e secondaryside, this w as treated as aninitiator.Excessive feedwater that resultsin overcooling transients weretreated as a transient category.Othe r excessive feedwater eventsthat result in a loss of mainfeedwater as the cause of tripwere included in the loss ofmain feedwater category.


L o s s of off-sitepower, loss ofservice wat er

Loss of mainfeedwater

Excessivefeedwater

31


34/152



1 2. Loss ofcondenservacuum

1 3. Inadequatemainfeedwater

1 4 . Feedwater orcondensateline breaks

1 5. Steam linebreaks

Treated separately due to itslong-term affects on main-feedwater availability andbecause it changes the avail-ability of the condenser as aheat sink whereas the loss ofmain feedwater does not.Some trips result from feedwaterinstabilities that do not de-grade th e reliability of feed-water after th e trip. These areincluded in the reactor-tripcategory. Another category wasdefined, partial loss of mainfeedwater , which describes thesituation when the operabilityof the main feedwater system isdegraded but not lost (e.g.,one-pump trip or condensate-pump trip).Treated as an initiating eventdue to the effects on theavailability of condensate tothe main and emergency feed-water systems.Kept as a separate category dueto its effects on both the pri-mary and secondary systems in-cluding overcooling and lossof steam-dump capability. Sub-categorieswere used to differentiate the responsesto breaks inside or outside thereactor building.

Loss of condenservacuum

Reactor/turbinetrip and partialloss of mainfe e d wa t e r

Feedwater linebreak

Steam line break

32


35/152


Potentialinitiating event Treatment in refinementprocess Fi nal category

1 6 . Turbine andcontrol valvemalfunctions1 7. Turbine-bypassvalve inad-vertent

opening1 8 . Turbine

malfunction1 9 . L o s s of con-denser cir-

culatingwater20. S m a l l R C Spipe breaks

2 1 . L a r g e R C Spipe breaks

N o significant effects on re-sponding plant systems.

T he inadvertent opening was in-cluded in the steamline breakcategory.

T he only turbine malfunctionsingled out for special studyw as turbine-missile generationEffects of the transient werefound to be included in theloss of condenser vacuum.

Grouped in one category based onan analysis of the responserequired. Breaks less than4 inches in diameter were in-cluded in the small-LOCA-categoryAll breaks greater than 4-inchdiameter were grouped into onecategory based on required res-ponse and success criteria.Special locations fo r these L O C A swere reviewed, but the effectswere judged not to be importantwhen compared to the importantmitigation system failure modes.


Steam line break


Loss of condenservacuum

S mall L O C A

L a r g e L O C A

33


36/152

TABLE 3.6.(cont.)


22. InadvertentP O R V orsafety-valveopening

23. Reactor cool-ant p u m p sealfailure

24 . Control ro ddrive sealleakage25. Interfacingsystem lossof coolant26 . Reactor vessel

rupture

27. Steam genera-tor tubeleak/rupture

Included in the s m a l l L O C A cate-gory. It should be noted thatPOR V o r sa f e ty-va lve L O C A s resultingfrom other transients are modelledexplicitly in the event tree.Included in the small L O C A cate-gory if the seal failure is aspontaneous initiating event.Seal failures resulting frominadequate protection after atrip are modelled explicitly.This event would have no signifi-cant differences from th esmall LOCA initia ting event.Treated in a special study due toth e specificity of the eventand its effectsTreated as a separate initiatingevent. The separate category isdefined to include ruptures thatcannot be mitigatedThe event is defined as its owncategory due to unique operatoractions and differentradiological considerationsfrom s m a l l L O C A s

S mall L O C A

S m a l l L O C A

Sma l l L O C A

Interfacing systemsL O C A

Reactor vesselrupture

Ste a m generatortube rupture

28. Chargingexceedsletdown

Inadvertent H P I operation w asjudge d to be the most restric-tive case and is treated as aninitiating-event category.O t h e r possibilities a retreated as subsets.

Spurious engineeredsafeguards signal

34


37/152

TABLE 3.6. (cont.)


29 . L e t d o w nexceedscharging30. Inadvertenthigh pressureinjection

31. F ai lure on orof f ofpressurizerheaters

32. Failure on oroff ofpressurizerspray

33. Loss of off-site p o w e r

This transient was judged to bea subset of the s m a l l L O C Acategory.Treated as a separate initiat-ing events category due to dif- engineeredferent boundary conditionscreated by the event.A review of the ICSidentified th e potential for fail-ure causing heaters to fail onand the P O R V to fail closed.This event w as treated as aseparate initiating event.Sprays failing of f would result ina high pressure trip, b ut condi-tions af ter trip should not begreatly dif ferent from those ofa reactor trip. Sprays failingon would have effects similarto those of event 26.Treated as a category due to itseffects on nearly all event-treetop events. The event was sub-divided into switchyard faultsa nd grid loss due to a differ-ence in the availability of theK e o w e e overhead for the twocases and because of the differ-ent recovery potentials.

S m a l l L O C A

Spurioussafeguards signal

Spurious lowpressurizerpressure signal


L o s s of substationswitchyard andloss of grid orfe e d e rs

35


38/152



34 . Loss of powerto necessarysystems

35. Loss of powerto controlsystems36 . Loss of

service water

Bec ause of the large number of po-tential initiating events in thiscategory, a special study w asconducted to determine th e mostsignificant events with respectto their effects on plant sys-tems (see Table 3.5).These events were examined assubsets of event 29 and event 1of Table 3.5.This category w a s analysed forspecific effects. Loss of servicewat er

37 . Loss ofcomponentcooling

38. Loss of instru-ment a ir

The service-water system atOc onee provides th e most impor-tant needs. Component coolingrequires service water for theheat exchangers. The loss ofcomponent cooling results fromthe loss of service water andwas treated in that category.Loss of the instrument air systemaf fects a number of top events a irsignificantly and also changesth e availability of control-room indications to the operator.The air system was reviewed andth e loss of system pressure wasj u d g e d to be bounding case overless severe malfunctions.

Loss of servicewat er

L o s s of instrument

36


39/152

TABLE 3.6. (cont.)


39. Integratedcontrolsystem power

4 0. Fi res affect-ing necessarysystems4 1 . Internal flood-in g affecting

necessarysystems

42.

43 .

GeneratorfaultsG r i d distur-bances

4 4 . Administrativeshutdowns

45. MSIV closure(1 or all)4 6 . A T W S

The ICS analysed in great detailboth as a support system and anevent initiator.S ee Section 6 on internal/externalhazards.

T he potential for internal plantflooding w as recognized and re-viewed. See Section 6 on internal/external hazards.D o e s no t affect systems requiredfo r adequate response.Gr id disturbances that result ingenerator trip do not changeplant response. Othe r gridfaults were assigned to thegrid-failure initiating event.These shutdowns were included inthe reactor-trip frequency ifduring th e shutdown a trip wasrequired. Shutdowns that pro-ceed orderly and do not requirea trip are considered successfulresponses.Not relevant to the Oconee plants.

Treated separately (see Section 6.1).

ICS initiators




37


40/152

TABLE 3.7.TRANSIENT INITIATING EVENTS AND THEIR ESTIMATEDF R E Q U E N C Y O F O C C U R R E N C E IN T HE G E R M AN R I S K S T U D Y (PWR)

Transient initiating events GRS phase B GRS phase A[22] [35

(D ) Operational transients (anticipated transients):1 4 . L o s s of emergency power 013 0.11 5. L o s s of main feedwater withoutloss of heat sink 0150.81 6 . L o s s of main feedwater andloss of heat sink 0290.31 7 . Loss of heat sink withmain feedwater on-line 036 -(E ) Steam line break transients (unlikely transients):1 8 . Large leakage in steam lineinside containment 1.6 x 10~4 -1 9 . L a r g e leakage in steam lineoutside containment 4.8 x 10~4 -20. Intermediate leakage in steam lineinside containment 2.7 x 10~5 -21 . Intermediate leakage in steam lineoutside containment 1.1 x 10"4(F ) A T W S22. A T W S by loss of main feedwater 4.7 x 10"6 4 x 10"5

23. A T W S by loss of emergency power 3.4 x 10"624 . A T W S by loss of circulating water 7.5 x 1 0~ 6 2.5 x 10"525. A T W S by other transients 2.3 x 10~5

38


41/152

T A B L E 3.8. L O C A I N I T I A T I N G E V E N T S AN D T H E IR E S T IM AT E D F R E Q U E N C Y O FO C C U R R E N C E I N T H E G E R M A N R I S K S T U D Y (PWR)

L O C A initiating events G R S phase B G R S phase A[22] [35

(A) L eakage in the primary cooling system piping:la . L a r g e L O C A (>10")


42/152

3.4.2. Initiating event lists for BWRs(A ) T he EPRI l i s t

Since there is a considerable difference b e t w e e n I B s i n P W R a n d B W R plants, E P R Ideveloped a separate IE list based on operating experience of plants with B W R s. The datafor B W R s i n clu d e 9 0 3 events occurring over 101 reactor years. The EPRI BWR l is t i s g ivenin Table 3.9. As for the P W R list, the number associated with each initiator is given inSection 4 .(B) The Grand Gulf list [36]

This list was obtained by making use of the IE lists in about 1 0 available PS As for U Sreactors. Generally, the initiators identified in these studies incorporate all events that hadoccurred i n U S nuclear power plants by 1980. A summary of these actual transient eventsis reported in EPRI-NP-801 [3] and i t s update EPRI-NP-2230 [4].T he above information was supplemented with actual plant trip data for Grand Gulf fora period of about 2 years of operation.A review of the Grand Gulf design for special initiators was also undertaken.The loss of coolant events were categorized into break size ranges, based on successcriteria for the safety injection systems. The resulting list of initiating events for Grand G u l fis given in Tables 3.10and 3.11.

3 . 5 . C O M P L E T E N E S S O F I E L I S T ST he first step in order to get a list of initiating events as complete as possible is to useseveral of the methods presented in Section 3.3. As an obvious minimum , one should includeth e following methods:

- Engineering evaluation;- Reference to previous PSAs and other available IE lists (e.g. E P R I list);- Reference to operating experience (if any relevant is available);- Logical evaluation ( M L D or E B F T ) .Each approach examines potential IB s from a different perspective yielding a highdegree of confidence that all risk significant IB s have been identified.T he more frequent IBs should, in general, be based on collected operating experience.This group of IBs is therefore considered to be relatively well covered.The problem of completeness arises mainly with respect to the low frequency IBs.L owfrequency IBs may be considered in three areas: pipe breaks or ruptures, other componentfailures and rare human errors. Pipe breaks and ruptures are considered to be relatively wellcovered if a proper engineering evaluation of the plant is performed. An important point isto examine piping outside th e containment which ha s connections to the primary system, inorder to reveal possible interfacing L O C A initiators. In that respect, the reliability of isolation

valves has to be adequately considered. Text cont. on p. 49.40


43/152

TABLE 3.9. LIST O F B W R T R A N S IE N T IN I TIA TIN G E V E N T S

1 . Electric load rejection2. Electric load rejection with turbine bypass valve failure3. Turbine trip4 . Turbine tr ip with turbine bypass valve failure5. Main steam isolation valve (MSIV) closure6 . Inadvertent closure of one M S I V7. Partial M S I V closure8 . Loss of normal condenser vacuum9 . Pressure regulator fails open1 0. Pressure regulator fails closed1 1 . Inadvertent opening of a safety/relief valve (stuck)1 2. Turbine bypass fails open1 3. Turbine bypass or control valves cause increase in pressure (closed)1 4 . Recirculation control failureincreasing flow1 5 . Recirculation control failuredecreasing flow1 6 . Trip of one recirculation p u m p1 7 . Trip of all recirculation pumps18 . Abnorma l startup of idle recirculation pump1 9 . Recirculat ion p u m p seizure20. Feedwaterincreasing flow at power21. Loss of feedwater heater2 2. Loss of all feedwater flow23. T rip of one feedwater pump (or condensate p ump )24 . F eedwat er low flow25. Low feedwater flow during startup or shutdown2 6 . High feedwater flow during startup or shutdown27. R od wi thdrawal a t power28. High f lux due to rod withdrawal at startup29 . Inadvertent insertion of control rod or rods30. Detected fault in reactor protection system31 . L o s s of off-site power32. Loss of auxiliary power (loss of auxiliary transformer)33. Inadvertent startup o f H P C I / H P C S34 . Scram due to plant occurrences35 . Spurious trip via instrumentation, RPS fault36 . M a nu a l scram; no out-of-tolerance condition37. Cause unknown

41


44/152

T A B L E 3.10. L O C A s A N D S P E C IA L I N I T I A T O R S I N C L U D E D IN T H E G R A N D G U L FP R A [36]

L O C A SIZEA (large L O C A )

51 (intermediate LOCA)

52 (small L O C A )

53

RV

Steam >0.4 sq.ft. (370 cm2)Li qui d >0.4 sq.ft.Steam 0.13-0.4 sq.ft.Liq u id 0.007-0.4 sq.ft.Steam


45/152

T A B L E 3.11. T R A N S I E N T I N I T I A T O R S I N T H E G R A N D G U L F P R AI N I T I A T I N GE V E N T G R O U P

E P R I T R A N S I E N TC A T E G O R Y R A T I O N A L E F O R I N C L U S I O N

T l( T r a n s i e n t s t h at

c a u s e L O S P )3 1 A l o s s o f t h e o f f s i t e g r i d w i l l r e s u l t I n a r e a c t o r s c r a m a n d l o s s

o f n o r m a l A C p o w e r T h e o n - s i t e e m e r g e n c y d i e s e l g e n e r a t o r s a r er e q u i r e d t o s t a r t t o s u p p l y A C l o a d s .

T2( T r a n s i e n t s w i t h

l o s s o f PCS*)

L o s s o f t h e n o r m a l a n d p r e f e r r e d s t a t i o n t r a n s f o r m e r s r e s u l t i n ar e a c t o r s c r a m a n d l o s s o f n o r m a l A C p o w e r T h e o n - s i t e e m e r g e n c yc i i e s e l g e n e r a t o r s a r e r e q u i r e d t o s t a r t a n d s u p p l y A C l o a d s .A g e n e r a t o r l o a d r e j e c t i o n r e s u l t s I n f a s t c l o s u r e o f t h p t u r b i n econtrol v a l v e s w h i c h i n t u r n scrams t h e r e a c t o r A s u b s e q u e n tf a i l u r e o f t h e t u r b i n e b y p a s s v a l v e s t o o p e n r e s u l t s i n c o m p l e t eI s o l a t i o n from t h e condenser.A t u r b i n e t r i p r e s u l t s i n c l o s u r e o f t h e m a i n t u r b i n e s t o p v a l v e swhich i n t u r n s c r a m s t h e r e a c t o r . A s u b s e q u e n t f a i l u r e o f t h et u r b i n e bypass v a l v e s t o o p e n r e s u l t s i n c o m p l e t e i s o l a t i o n f r o mt h e c o n d e n s e r .M S I V c l o s u r e r e s u l t s i n a r e a c t o r s c r a m a n d c o m p l e t e I s o l a t i o nf r o m t h e c o n d e n s e r .C l o s u r e o f o n e M S I V m a y r e s u l t I n a h i g h s t e a m l i n e f l o w s i g n a lt h a t I s o l a t e s t h e m a i n s t e a m l i n e s f r o m t h e c o n d e n s e r b y dosingt h e r e m a i n i n g M S I V s . M S I V c l o s u r e s c r a m s t h e r e a c t o r

L o s s o f PCS, t h a t i s , r e a c t o r I s o l a t i o n f r o m m a i n c o n d e n s e r


46/152

TABLE 3.11. (cont.)

I N I T I A T I N GE V E N T GROUP

E P R I T R A N S I E N TC A T E G O R Y R A T I O N A L E F O R I N C L U S I O N

T2( T r a n s i e n t s w i t h

l o s s o f PCS*)( C o n c l u d e d )

P a r t i a l c l o s u r e o f o n e M S I V m a y r e s u l t i n a h i g h s t e a m l i n e f l o ws - g n a l t h a t i s o l a t e s t h e m a i n s t e a m l i n e s f r o m t h e c o n d e n s e r b yc l o s i n g t h e r e m a i n i n g M S I V s . M S I V c l o s u r e s c r a m s t h e r e a c t o r .A l o s s o f c o n d e n s e r v a c u u m c a u s e s a c l o s u r e o f t h e m a i n t u r b i n es t o p v a l v e s . M S I V s . a n d t u r b i n e b y p a s s v a l v e s . T h e t u r b i n e t r i pi n i t i a t e s r e a c t o r s c r a m .A p r e s s u r e r e g u l a t o r f a i l u r e i n t h e open p o s i t i o n w i l l c a u s e t h em a i n t u r b i n e c o n t r o l v a l v e s a n d b y p a s s v a l v e s t o c o m p l e t e l y o p e nr e s u l t i n g i n a l o w t u r b i n e i n l e t p r e s s u r e i s o l a t i o n o f t h e M S I V s .M S I V c l o s u r e i n i t i a t e s r e a c t o r s c r a m .

1 0 A p r e s s u r e r e g u l a t o r f a i l u r e i n t h e c l o s e d d i r e c t i o n w i l l r e s u l ti n c l o s u r e o f t h e m a i n t u r b i n e c o n t r o l v a l v e s a n d i n h i b i t o p e n i n go f t h e t u r b i n e b y p a s s v a l v e s . A h i g h n e u t r o n f l u x s i g n a l w i l lscram t h e r e a c t o r .

1 2 A n i n a d v e r t e n t o r e x c e s s i v e o p e n i n g o f a t u r b i n e b y p a s s v a l v e w i l ldecrease t h e m a i n s t e a m l i n e pressure r e s u l t i n g i n a l o w t u r b i n ei n l e t p r e s s u r e c l o s u r e o f t h e M S I V s . M S I V c l o s u r e i n i t i a t e s r e a c t o rs c r a m .

1 3 F a i l u r e o f t h e t u r b i n e c o n t r o l v a l v e s a n d b y p a s s v a l v e s i n t h ec l o s e d p o s i t i o n w i l l i s o l a t e t h e r e a c t o r f r o m t h e m a i n c o n d e n s e r .C l o s u r e o f t h e t u r b i n e s t o p v a l v e s i n i t i a t e r e a c t o r s c r a m .

37 A l l t r a n s i e n t s w i t h u n k n o w n causes a r e a s s u m e d t o r e s u l t i ni s o l a t i o n o f t h e r e a c to r v e s s e l f r o m t h e m a i n c o n d e n s e r .

"Loss of PCS. that is, reactor isolation from main condenser.


47/152

TABLE 3 . 1 1 . (cont.)

I N I T I A T I N GE V E N T G R O U P

E P R I T R A N S I E N TCATEGORY R A T I O N A L E F O R I N C L U S I O N

T 3 A( T r a n s i e n t w i t hP C S a v a i l a b l e )

A g e n e r a t o r l o a d r e j e c t i o n c a u s e s a f a s t c l o s u r e o f t h e t u r b i n ec o n t r o l v a l v e s t h at i n t u r n c a u s e s a r e a c t o r s c r a m . T h e M S I V sw i l l r e m a i n open a n d t h e t u r b i n e b y p a s s v a l v e s w i l l a l l o w s t e a mf l o w t o t h e c o n d e n s e r .A t u r b i n e t r i p r e s u l t s i n c l o s u r e o f t h e m a i n t u r b i n e s t o p v a l v e sw h i c h i n t u r n causes a r e a c t o r s c r a m . T h e H S I V s w i l l r e m a i n o p e na n d t h e t u r b i n e b y p a s s v a l v e s w i l l a l l o w s t e a m f l o w t o t h ec o n d e n s e r .A f a i l u r e o f t h e r e c i r c u l a t i o n f l o w c o n t r o l w h i c h i n c r e a s e sr e c l r c u l a t i o n f l o w r e s u l t s i n a h i g h n e u t r o n f l u x s c r a m o f t h ereactor. T h e t u r b i n e c o n t r o l v a l v e s w i l l c l o s e u p o n d e c r e a s i n gt u r b i n e pressure. T h e M S I V s r e m a i n open a n d t h e t u r b i n e bypassv a l v e s w i l l a l l o w s t e a m f l o w t o t h e c o n d e n s e r

1 5

1 6

A f a i l u r e o f t h e r e c l r c u l a t i o n f l o w c o n t r o l w h i c h decreasesr e c l

defining initiating events for purposes of probabilistic safety assessment

Documents