meeting summary, held on 08/29/1991, the acrs subcommittee

% ,_i

by W. Kerr on 10/10/91

ACRS SUBCOMMITTEE MEETINGSUMMARY/MINUTES OF THE INSTRUMENTATION

AND CONTROL SYSTEMSAUGUST 29, 1991

BETHESDA, MARYLAND

INTRODUCTION

The ACRS Subcommittee on Instrumentation and Control Systems held

a meeting on August 29, 1991 in Bethesda, Maryland.

The entire subcommittee meeting was open to public attendance. The

documents submitted to the subcommittee Members for this meeting

are attached.

ATTENDEES NRCACRS

W. Kerr, ChairmanH. Lewis, MemberC. Wylie, MemberM. El-Zeftawy, Staff

OTHERS

K. Barrett, BechtelC. Tuley, KJ. Trotter, EPRIK. Klaube, NUSP. Ludgate, NUSC. Swanson, B&WG. Srikantiah, EPRIR. Krich, PE Co.G. Salamon, FP&LS. Palena, PE Co.J. Armstrong, PE Co.L. Hopkins, PE Co.G. Stewart, PE Co.

W.R.T.R.W.G.E.J.D.A.J.C.

Butler, NRRClark, NRRKenney, SRIVan Houton, SECYSwenson, NRRGalletti, NRRImbro, NRRJacobson, NRRNorkin, NRRGautam, NRRGallagher, NRRDoutt, NRR

I-ler I? -S-

9204130127 911010 -PDR ACRS2766 PDR I/I,

I

Inst & Cntl 2Mtg. (8/29/91)

Meeting Highlights. Actions. Agreements

1. Dr. Kerr, Subcommittee Chairman, convened the meeting and

stated that the purpose of the meeting is to discuss with theEPRI representatives, representatives of Philadelphia Electric

(PE) Company, and the NRC staff the following items,respectively.

* A set-point methodology developed by EPRI

* Transient Response Implementing Plan (TRIP) procedures

developed and implemented by PE Co.

* Electrical Distribution System Functional Inspections(EDSFIs) currently being performed by the staff.

2. Mr. Srikantiah, EPRI, briefed the subcommittee Members

regarding an EPRI procedure for determining the set-points forsafety systems. He described a generic process forestablishing appropriate and justifiable limits for the

reactor monitoring and protection systems set-points. This

process was designed to be applicable to utilities with either

PWRs or BWRs who plan to assume primary responsibility for

performing safety analysis and set-points determination.

Mr. Srikantiah stated that utilities are familiar with thegeneral aspects of set-point methodology, as well as theresults of its use in reactor operation. However, the methods

and assumptions used by vendors to determine set points are

not readily available for utility use. A plant's set points

depend on the operational goals of the licensee and are

derived from a consideration of safety analyses andperformance considerations. The performance evaluation takesaccount of the constraints on plant operation, given the

limitations imposed by the safety analysis. In turn, the

safety analysis verifies the acceptability of the assumed set

points. The safety analysis and the determination of set

points are interrelated processes.


An important step in the determination of acceptable set-

points is the treatment of the uncertainties. These include

uncertainties inherent in the actual equipment settings and

those in the inputs to the license basis analysis and plant

performance evaluations. The estimated values have an

associated uncertainty distribution.

EPRI's set-point process is related to the process used in

the final safety analysis report (FSAR) design analyses. The

FSAR focuses on plant response to design basis events (DBEs)

while EPRI's set-point process focuses on monitoring and

protection system set-points used in the analyses of the

limiting cases in order to assure that acceptable consequences

result.

Verification of the monitoring and protection system limits

is accomplished through an analytical demonstration of

acceptable consequences of DBEs. The analysis plays a central

role in the determination of set-points. The results of the

event analyses are compared to the event acceptance limits.

If the results are acceptable then the set-points are

considered acceptable. If the results are not acceptable,

then a margin recovery process is undertaken which may include

changing set-points, changing the operating envelope, changing

model inputs, changing analysis assumptions and/or methods,

or changing the plant design. An alternate approach developed

by EPRI is based on what is described as a more realistic

treatment of uncertainties. The approach involves a

statistical combination of uncertainties (SCU). EPRI has

developed guidance for determining the appropriate event

analysis model inputs from the plant design and determining

the operating envelope sources.


The following are EPRI's guidelines for LWR set-point

determination;

A. Complete the plant performance evaluation.

B. Identify the plant specific DBEs which must be analyzed

and the associated acceptance limits.

C. Identify the plant design model inputs and operating

envelope limits to be used in the event analyses.

a) Plant design model inputs

* From the plant design determine the numerical

ranges for the model inputs.

* Include in the ranges the appropriate

uncertainties.

b) Operating envelope limits

* From the monitoring system determine the

numerical ranges for the operating envelope

limits.

* Include in the ranges the appropriate

uncertainties.

* To these ranges apply any required maneuvering

to determine the analysis input ranges over

which the event analyses must be valid.

* Select axial power shapes.

D. Model the protection system in the event analyses.

* Identify trips to be used.

* Determine the protection system instrument set-

points.

* Apply appropriate uncertainty allowances to the

instrument set-points.

* Account for event effects through appropriate

allowances or explicit modeling to determine

the analytical set-point for the event

analyses.

E. Perform the event analyses insuring that limiting cases

have been analyzed for each event and all acceptance


limits have been satisfied.

F. Implement margin recovery process if required.

* Change the protection system instrument set-points.

* Change the operating envelope.

* Modify the event analyses models.

* Change the plant design.

* Apply statistical treatment of uncertainties.

Mr. Srikantiah stated that EPRI's statistical combination of

uncertainties methodology can be used to reduce excessive

conservatism inherent in the existing deterministic analysis

process. Use of EPRI's method is expected to decrease the

number of scrams and to eliminate some actuations of safety

and relief valves, thus increasing operating flexibility. Mr.

Srikantiah cited some applications of the EPRI's set-point

methodology, such as Oyster Creek high-pressure set-points,

Vermont Yankee reload analysis, Diablo Canyon safety valve

tolerance, TMI-1 flux/flow set-points, and Palo Verde safety

valve tolerance.

Dr. Kerr commented that EPRI's approach appears to be

plausible but complicated. Further several steps employ

"engineering judgement" making the results unpredictable.

EPRI representatives agreed that given the same data, two

independent reviewers could arrive at two different

conclusions.

Dr. Kerr stated that there is no evidence of any systematic

consideration of whether use of EPRI method would have any

effect on risk. He commented that it is possible if the

method decreases the number of unneeded scrams, and at the

same time does not increase the likelihood of exceeding a

safety limit, risk may be decreased by its use. He

recommended that prior to the NRC approval of this method for

;


widespread application, a systematic investigation be made of

its effect on risk. Dr. Lewis and Mr. Wylie agreed.

3. Mr. L. Hopkins, Limerick Units 1 & 2, presented an overview

of the TRIP program. He stated that following Three Mile

Island (TMI) accident, the NRC required licensees of operating

reactors to reanalyze transients and accidents and to upgrade

emergency operating procedures (EOPs). The NRC staff

presented guidelines to the licensees to upgrade EOPs and

described the use of a "Procedure Generation Package" (PGP)

to prepare EOPs. The PGP would include plant-specific

technical guidelines, a writers guide, a description of the

program to be used for the validation of EOPs, and a

description of the training program for the upgraded EOPs.

From this PGP, plant specific EOPs were to have been developed

that would provide the operator with the directions to

mitigate the consequences of a broad range of accidents and

multiple equipment failures. Some licensees such as Limerick

Generating Station (LGS) Units 1 and 2 and Peach Bottom have

developed and implemented their own EOPs, using guidance

formulated by the BWR Owners Group (BWROG), such as the

Transient Response Implementing Plan (TRIP). TRIP equals

EOPs. LGS is a BWR with a Mark III containment structure.

Mr. J. Armstrong, Philadelphia Electric Company, indicated

that the NRC is currently performing inspections on different

plants to evaluate the EOPs at the licensee facilities. The

NRC staff considers the objectives to be met if review of the

following areas were found to be adequate: comparison of the

TRIP procedures with the LGS plant specific technical

guidelines (PSTG) and the BWR owners group emergency procedure

guidelines (EPG); review of the technical adequacy of the

deviations from the EPG; control room and plant walkdowns of


the TRIP procedures; real time evaluation of the TRIP

procedures on the plant simulator; evaluation of the licensee

program on continuing improvements of the TRIP procedures; and

performance of human factors analysis of the TRIP procedures.

Dr. Kerr discussed with PE representatives some of the

comments made in the staff's report transmitted by a letter

of July 3, 1990 to Mr. D. M. Smith (PE Co.). PE

representatives consider the comments made by the NRC staff

during the inspection as constructive and helpful.

Dr. Kerr commented that examination of some of the comments

and PE's responses might lead to the conclusions that PE

personnel consider staff suggestions tantamount to an order.

Dr. Kerr cited an example in which the staff commented that,

"the glare from the Plexiglass covering the flowcharts in the

control room is excessive." The PE personnel indicated that

they would have not been aware of this problem until the staff

called it to their attention and they corrected it

immediately.

4. Mr. E. Imbro, NRR, briefed the subcommittee Members regarding

the Electrical Distribution System Functional Inspections

(EDSFIs). He stated that during multidisciplinary

inspections, the NRC has identified many deficiencies related

to the electrical distribution system. To address these

deficiencies, the NRC has developed an inspection to

specifically evaluate the electrical distribution system. The

NRC completed 24 EDSFIs at the five NRC geographical regions.

During these inspections, the staff found several common

deficiencies in the licensees' programs and in the electrical

distribution systems as designed and configured at each plant.

These deficiencies included inadequate ac voltages at the 480

Vac and 120 Vac distribution levels, inadequate procedures to

Inst & Cntl 8Htg. (8/29/91)

test circuit breakers, and inadequate determinations and

evaluations of setpoints.

Mr. J. Jacobson, Mr. A. Gautam, and Mr. D. Norkin, NRR, stated

that for several nuclear power plants, the staff found that,

under certain conditions, the voltage available at the safety

buses would be inadequate to operate safety-related loads and

associated equipment. These conditions could occur when the

plant's electrical distribution systems were being supplied

from an offsite grid that had become degraded but that

continued to supply voltages that remained above the setpoints

at which the degraded grid relays would be activated.

The function of the degraded grid relays is to ensure that

adequate voltage is available to operate all Class 1E loads

at all distribution levels. In order to ensure that all

required Class 1E loads will remain operable during degraded

offsite voltage conditions, some licensees are currently

reanalyzing the basis for the degraded grid relay setpoints.

The staff identified repetitive deficiencies in licensees'

programs to test circuit breakers. These deficiencies

included inadequate procedures, inadequate test acceptance

criteria, inadequate test equipment, and inadequate control

of testing. At the Susquehanna plant, the staff found that

the licensee was testing dc molded case circuit breakers with

a procedure written for testing ac breakers. The licensee had

not established specific acceptance criteria for the dc

breakers.

Other findings identified during recent EDSFIs were related

to inadequate setpoint determinations. Some licensees have

operated equipment outside of acceptable limits because they

did not determine proper setpoints and did not evaluate and


account for instrument drift. The staff has identified these

circumstances primarily for instruments for which the

setpoints were determined by the architect/engineer or the

nuclear safety system supplier. Those setpoints not contained

in the plant technical specifications were also more

frequently found to be deficient. During recent EDSFIs, the

staff identified deficiencies in setpoints for diesel day tank

level indicators, diesel air start compressor controllers and

alarms, invertor low voltage shutdown circuitry, degraded grid

relays, and diesel overcurrent relays.

The staff has published Information Notice No. 91-29 on April

15, 1991. The staff believes that its inspections have

resulted in an increased sensitivity to the issue throughout

the industry. It is planned to continue performing EDSFIs

until all plants have been inspected by January 1993.

In response to Dr. Kerr's question regarding the risk

reduction that is likely to be achieved by this program, the

NRC staff had no answers and indicated that no effort has been

made to calculate or estimate the possible risk reduction.

Dr. Kerr stated that he would have liked to see evidence from

some source (may be NUREG-1150), that electrical systems in

operating plants represent a significant risk contributor, and

the result of the EDSFI program will produce some risk

reduction. The subcommittee Members commented that it should

be a Commission Policy to make decisions on starting new

programs based (at least partially) on the need and

expectation of risk reduction.

Dr. Kerr asked whether, now that a significant number of

plants have been examined and the results of those

examinations are available, it would be more effective to make

these results available and to ask the remaining plants to

conduct their own examinations as part of their IPEs. The

Inst & Cntl 10Mtg. (8/29/91)

staff appeared to believe that NRC management would not

approve of this approach, and the resulting examinations would

not be as thorough nor as effective as they would be if

performed by the staff.

In conclusion, the subcommittee Members agreed that the plant

electrical system is important, and that its reliability and

performance are important to both the prevention and the

mitigation of accidents.

FUTURE ACTION

The Subcommittee Chairman is planning to brief the full Committee

at the September 5-7, 1991 ACRS meeting regarding the above three

topics. Pending the outcome of such briefing, the Committee may

decide on a course of action..

Attachments:1. EPRI Setpoints Projects/Methodology

and Applications - by G. Srikantiah

2. Philadelphia Electric Co. (slides)

3. EDSFI briefing to ACRS

KOTEs Additional meeting details can be obtained from atranscript of this meeting available in the INRC PublicDocument Room, 2120 L Street, NW, Washington, DC 20006,(202) 634-3273, or can be purchased from Ann Riley andAssociates, Ltd., 1612 K Street, NW, Suite 300,Washington, DC 20006, (202) 293-3950.

EPRI/NPD

EPRI Setpoints Projects

Methodology and Applications

Presented by

.G mSrikantiah

August 1991

. -S & R/ Reactor Performance ProgramACRS 8/27/91 1

EPRI/NPD

EPRI Setpoint ProjectsMethodology and Applications

IIIi

OVERVIEW- Background

-- ;.Setpoin .t-Optimization ObjectivesExisting Methodology

- Safety Analysis Process

- Current Treatment of MeasurementUncertainties

- Development of Safety AnalysisInputs

- Potential Problems

EPRI SETPOINT METHODOLOGY

EPRI SETPOINT PROJECTSk I

S & R/ Reactor Performance ProgramACRS 8/27/91 2

EPRI/NPD

EPFRI Setpoint Projects

Background

* Setpoints are Instrument Settings atwhich System Automatic orOperator Action Must Occur toPreserve the Assumption in thePlant Safety Analysis

* Setpoints Must Satisfy Plant Safetyand Licensing Requirements

* Plant Capacity and OperatingFlexibility Should be Optimized

S & R/ Reactor Performance ProgramACRS 8/27!91 3

EPRI/NPD

EPRI Setpoint Projects

Setpoint Optimization Objectives

To Improve Plant Performance throughOptimized Instrument Settings whileSatisfying Safety and LicensingRequirements

*'To Gain Operational Margin

* To Improve Operational Flexibility

S & R/ Reactor Performance ProgramACRS 8127191 4

EPRI/NPD

EPRI Setpoint Projects

Existing Methodology

* Existing Methods are Based onDeterministic or License Basis AnalysisAssumptions, Including the MostAdverse Plant Conditions.

* Highly Conservative Methodology-RestrictsOperational Flexibility andMay Result in Derates

* Major Improvements Achievable byQuantifying and CombiningUncertainties in Key Variables


EPRI/NPD

CURRENT SAFETY ANALYSIS PROCESS

(Deterministic Methodology)

EVENT LIMITS

ALLOWABLE RANGE

ANALYSIS RESULTS

r

INSTRUMENT TRIPS(ANALYTICAL LIMIT)

1 PLANT TRAUSIEWRESPONSE

MODEL INPUTS/OPERATING LIMITS


,-° EPRI/NPD

CUFMEASU

THRE. TYPICAL APPROACH.S

IRENT TREATMENT OFREMENT UNCERTAINTIES

1. HISTORICAL

PROCESSVAR;AB':

SAFETY ANALYSISRESULTS

NOMINALSET PO!NT

ANALYTICAL EVENTLIMIT LIMIT

2. ENGINEERING JUDGMENT

PROCESSVARIABLE

ALLOWANCE PORINSTRUMENT DRIFI 1

ALLOWANCE FOR INSTRUMENSYSTEM UNCERTAINTIES ANDCALIBRATION UNCERTAINTY

SAFETY ANALYSISRESULTS I

I I I I

NOMINALSETPOINT

ALLOWABLESETPOINT

ANALYTICALLIMIT

EVENTLIMIT

3. SIMPLIFIED STATISTICAL

PROCESSVARIABLE

MAXIMUM INSTRUMENIDRIFT (2d SAFETY ANALYSIS RESULTS

NOMINALSETPOINT

I I I

ALLOWABLE ANALYTICAL EVENTSETPOINT LIMIT LIMIT


EPRI/NPD

POTENTIAL PROBLEMS

EVENTLiMITS

ALLOWABLE RANGEANALYSISRESULTS

PLANT TRANSIENTRESPONSE

INSTRUMENT'TRIPS

MlODEL INPUTS& OPERATINGLIMITS

Deterministic AcceptableRESULTS ACCEPTABLE

Statistical AnalysisMay Be Solution

ANALYSISRESULTS

MVNTLUITS

PLANT TRANSIENTRESPONSE

INSTRUMENTTRIPS

MODEL INPUTS& OPERATINGLIMITS

RESULTS UNACCEPTABLE

ANALYSISRESULTS

EVENTLIMITS(

ANALYSISRESULTS

MODEL INPUTS& OPERATINGLIMITS

MODEL INPUTS& OPERATINGLIMITSINSTRUMENTTRIPS

ANALYSIS RESULTSUNACCEPTABLE

= S & R/

INCREASED OPERATING SETPOINT WVERLAMARGIN

Reactor Performance ProgramACRS 8/27/91 8

EPRI/NPD


OVERVIEW

EPRI SETPOINT METHODOLOGY

* Safety Analysis and Setpoint Process

* Safety Analysis and MeasurementUncertainties

a Setpoint Modification Processa Methodology Options

-Project Plan Development

C Statistical Analysis Methodology

* Statistical Combination of Uncertainties* Network Considerations* Combined MethodologiesEPRI SETPOINT PROJECTS


EPRI/PPD

GENERALIZED SAFETY ANALYSISAND SETPOINT PROCESS

< _ S & R/ Reactor Performance ProgramACRS 8/27/91 10

EPRI/NPD

SAFE_SFTPOII

TY ANALYSIS RELATED4TS AND UNCERTAINTIES

Actual BarrierIntegrity

Defined BarrierIntegrity

Event AcceptanceLimits

Event AnalysisResults

AnalyticalInstrumentSetpoints

Not 4iADDITIONAL MARGIN Ouzntffitble

Marin 4

'4 ~~~~~~~~~~MARGINCORRELATION OFUNCERTAINTIES Quantifiable SAFETY

CORRELATION INPUT MarginUNCERTAINTIES 1-

ALLOWABLE RANGE

MODEL UNCERTAINTIES 4Event Analyses/ EVENTPlant Dynamic Responses ANALYSIS

MODEL INPUTUNCERTAINTIES

'4^CALIBRATIONUNCERTAINTYMEASUREMENTSYSTEMUNCERTAINTIES

INSTRUMENT DRIFT

Measurement

UncertaintiesINSTRUMENT

SETPOINTPROCESS

AllowableSetpoints

NominalSetpoints

OPERATING MARGINOperatingEnvelope

Limits

*PLANNED MANEUVERING 4 4ALLOWANCE Operating OPERATING- - - - - - - - - - - - - --- - Envelope ENVELOPEao a a * ,. _- _ _ lesZICAVI NIAIL Allowanlces FHVkwLo

Steady State OPERATING ALLOWANCEOperation '

~ S & R/ Reactor Performance ProgramACRS 8/27/91 11

EPRI/NPDSETPOINT MODIFICATION PROCESS

-

CURRENT PLANTSAFETY ANALYSIS

PLANT DESIGNINFORMATION

RASP GENERICSETPOINT

METHODOLOGY

_ - _

I

. i 4 IPLANT SPECIFIC

SETPOIN1METHODOLOGY

I- I

4 _DETERMINE EVENT

ACCEPTANCELIMITS

IDENTIFY EVENTSTOBE

ANALYZED

PERFORM BASELINEEVENT ANALYSIS

. _

I

i 4 IF��

QUANTIFYUNCERTAINTIES

-__-4DETERMINE ANDVERIFY THE NEW

SETPOINTS_

u ,._

PREPAREDOCUMENTATION J


EPRI/NPD

Description of Methodology Options

Deterministic* Conservative selection of analysis inputs* Analytical instrument setpoints at 95% probability* No additional model uncertainty required* Results very conservative

U Used in current safety analyses

Complete SCU* All analysis inputs statistically treated* Results determined at the 95% probability/95%

confidence level* Model uncertainty considered in the analysis* Minimum amount of conservatism in results* Only used to:determine ,DNBR/M1CPR safety limits

Mixed SCU* Limited number of analysis inputs statistically treated* Remainder of inputs same as deterministic process* Model uncertainty considered in the analysis* Results determined at the 95% probability level for the

statistically treated parameters* Intermediate amount of conservatism results

Increasing usage in the safety analysis process


EPRI/NPD

SEETPOINT MODIFICATIONJECT PLAN DEVELOPMENTPRO.

I

I

a

I

I

I

i

I

I

0

0

. . . . . .. . .. .. .. .. . . . . . . . .. ..... ................. . . . . .. . . . . . .. . .

Plannina

PROJECTPLAN

UNDERSTAND CAPABILITIESAND TRADEOFFS

AVAILABLEMETHODOLOGY

RESOLUTION OFPROBLEM

.x TREATMENT OFUNCERTAINTIES

ASSESSMENTOF IMPACT

SELECT OPTIMUM APPROACHCOSTiMARGIN TRADEOFF

............................ …I. . . . . . . . . . . . . . . . . . . . . . .

0

4

II

DETERMINISTICANALYSISPROCESS II

STATISTICALCOMBINATION OFUNCERTAINTIES

Available Methodologv

04

0

0

0

0

0

0

I

I. . . . . .--------------------- -- -------

ENHANCED UNDERSTANDINGAS PROJECT PROGRESSES

F Recycle asNecessary 1

S & R/ Reactor Performance ProgramACRS 8127/91 14

EPRI/NPD

{ cQnf -4�stical CombinationWJ LGLI L 1

of Uncertainties

* Methodology

* Data Manipulation

* Relationship to Screening Process

* Analysis Input Screening

* Response Surface Development

* STARS Methodology

* Benefits


<-Ak EPRI/NPD

-44"STATISTICAL COMBINATION OF UNCERTAINTYMETHODOLOGY

IDENTIFY EVENT LIMITS

DETERMINE TREATMENT OF INPUT PARAMETERS -SCREENING PROCESS

1. Identify Analysis Input Parameters to Be Treated Determinlistically

2. Identify Analysis Input Parameters to Be Treated Statistically

3. Characterize Statistical Inputs

DEVELOP UNCERTAINTY DISTRIBUTION FOR THE EVENTBEING ANALYZED - SCU PROCESS

1. Develop ResponseSurface

2. Simulate Event to Obtain Combined Effect of UncertaintiesIncluding the Uncertainties in the Response Surface Fit

DETERMINE EVENT PROBABILITY AS A FUNCTION OF THEEVENT LIMITS

COMPARISON TO THE EVENT ACCEPTANCE LIMITS

S & R/ Reactor Performance Program °ACRS 8/271/91 16

EPRI/NPD

SETFDATA MANIPULATION FOR

POINT MODIFICATION ANALYSES

Deterministic Statistical Combinationof Uncertainties

DATA FOR SPECIFICCOMPONENTS TO

BE CHANGED

DATA FORPOTENTIALLY. SENSITIVEPARAMETERS

DETERMINEUNCERTAINTIES

CONSERVATIVELYINCLUDE IN

ANALYSIS PROCESS

PERFORMANALYSES

. .4DETERMINE

UNCERTAINTIES

SCREEN

DEVELOPRESPONSESURFACE

I _.

Process Is Similar, but Number ofParameters and Their Treatment

Is More Extensive for SCU Process


EPRI/NPD

REL.ATIONSHIP OF SCREENING PROCESSTO SCU METHODOLOGY

STATISTICALDATA

I

DETERMINEtM EAN AND

-STANDARDDEVIATION

1 _~~~~~~~~~~~~~Fy

I SCREENING.-PROCESS

_ _ _

RESPONSESURFACE

DEVELOPMENT

STATISTICALDISTRIBUTIONNORMALUNIFORMOTHER

RESPONSESURFACE

FITTINGCOEFFICIENTS

MONTECARLO

ANALYSIS

I EVENTACCEPTANCE

LIMITS


EPRI/NPD

ANALYSIS INPUT SCREENING PROCESS

r DesiredSetpointChanges

I1 Plant Model withAll Parameters at

License BasisValues - - �EVENTS

.-REQUIRINGANALYSIS --- A-4

VERIFY SETPOINT-RELATIONSHIPS'RETAItNED FOR

DESIRED CHANGES-

ELIMINATENOT APPLICABLE

PARAMETERS FOREVENTS TO BE

ANALYZED

4SCREEN FOR POTENTIALLYSIGNIFICANT PARAMETERS

SMALL UNCERTAINTIES ENGINEERING JUDGMENT- Physical Properties 1 Difficult- Geometric Inputs * Expensive

V

IDENTIFY NOMINALVALUES AND

VUNCERTAINTIESOF PARAMETERS

POTENTIALLYSENSITIVE

:''PARAMETERS.DENTIFIED

[ ~~SENSITIVITY STUDIES

Run Baseline Run Parameter Run ExtremeCases for Variation Cases Sensitive Cases

Each Event for Each Event as Necessary

DEVELOP CRITERIAPARAMETERS TO FOR SELECTION

BE USED IN F OR SELECTIONOF SENSITIVE ITHE SCU PROCESS PARAMETERS


EPRI/NPD

STATISTICAL L COMBINATION OF UNCERTAINTIES PROCESSESPONSE SURFACE DEVELOPMENT" I

or

STATISTICALLYTREATED INPUT

PARAMETERS

Ir

xi

i(x2)

X2

IAt(xi)

xi

Analysis Input Patameters Probability Density Functions_ -

I II.5

DETERMINISTICINPUT PARAMETERS H EVENT

ANALYSISMETHODOLOGY H RESPONSE

SURFACE DESIGN

(Type of Fit andCase Definition)

. .

:m;i-:4I-A L-

NN N

y~x)= BO+XB9x, +FxRixfoull i-I

rn

i(r)N N

+ X' TBqxjxi.1=1 i-"ij t..

- ~NC2at = NC K1 Y (Yk - YRSk 2

Lt- NC -K - 1 k-1 S

For OCCO

NC*2N+2N+1

K wN +) IN *Z2

;2 MONTE CARLOSIMULATION

_ .

Cases for a 3 Parameter OCCO (N-3)and Normally Distributed inputs

Par-± rl Par E2aLrAtj.-20 IA-20; p-2ag.+2a IA-2a IL-2cis-2c0 psa is-20JA+2a p.-2o i- 2oIL-2a p-2la is+2p+2a gL-2c IL+2p-2c i.+2a p+20p.+2a ipti2c p+20

IL-2Aa is )Aist2Aa A HL

,A js-2Aa p

H 9A p-AH H $L-2Aaas a

EVENTLIMITI EVENT 1~~~~~~~.0

i

, ._ A

S & R/

Figure of Merit

Reactor Performance Program 0ACRS 8/27/91 20

EPRI/NPD

C"TARS METHODOLOGY

-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ICODEINPUTS K

RESULTS OF STATISTICALSYSTEM PROPERTIES

TRANSIENT OF INPUTI CALCULATIONS PARAMETERSI III- - - - r - - - - - - - - -

III

RESPCS.URF

.CONSTR

(Least SEvalui

)NSEACEUCTION

iquaresation)

PolynomialCurve Fit

iTICAL,YSIS '

lo Sampleeier.Set)

Probability Distributionfor System Analysis I

STARS I

CODE…I-- - - - - - - - - - - - - -.

II

STATISANAL

(Monte Caiof Param;

V_

PROBABILITYOF EXCEEDING

EVENT LIMIT


EPRI/NPD

BEN

0.06

0.05 _

IEFITS OF STATISTICAL COMBINATIONOF UNCERTAINTIES METHODOLOGY

-OM0

0.04

0.03

0.02

0.01

0.00

FIGURE OF MERITProbability Density Function

75

0~

w

-J

7-)

1.0

0.9 -

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

I~~~~~~~~~~

-~ ~ ~ ~ ~ ~ ~ ~ ~ ~ .4

/ ~~~~~~~~~~~~I/C Cd

/~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

IncreasedMargin

I I

I DetermninisticIAnalysis Results

Acceptabl(at 95% Pr(

Event? Value I Limit,bability I

I I . I

FIGURE OF MERIT

Cumulative Distribution Function


CDD5I /K1 fJr

Network Analysis

* Network Considerations

Typical Component Networks

v"Development of-Analysis -Inputs

* PLANETS Methodology

* Typical Individual ComponentPerformance

* Examples of Network Performance

* Event Analysis Inputs


EPRI/NPD

Network Considerations* Component networks are frequently

encountered in nuclear power plantdesign

- Single failure considerations- Avoidance of spurious trips or

system startups- Improved system reliability

* Safety Analyses Typically TreatComponent Networks as IndividualComponents

- Deterministic methodologyphilosophy

- Computer code model limitations* Component Networks

- Superior to individual components- Methodology available to quantify

benefitsI I

S & R/ Reactor Performance Program °ACRS 8127191 24

EPRI/NPD

TYPICAL COMPONENT NETWORKS

CHANNE-ATRIP

SYSTEM-- -- ACTUATiON

1 of 1 Logic

CHANNELATRIP

I - SYSTEMCHANNELB ACTUATIONIRiP

2 of 2 Logic

CHANNE_.ATRiP

CHANNEL3,RIP

CHANNELAIRIp

CHANN=E3T RIP

CHANNELCTRIP

CHANNEL3TRIP

CHANNELATRIP

CHANNEL3TRIP

CHANNELCTRIP

CHANNELNTRIP

CHANNELATRIP

CHANNELETRIP

CHANNELCTRIP

CHANNELNTRIP

SYST--tlZ oa - ACTUA-rON

1 of 2 Logic

SYSTEt.1ACTUATiON

CHANNELATRIP

CHANNELB SYSTEMTRIP ACTUATION

CHANNELCTRIP

2 of 3 Logic

CHANNELATRIP

CHANNELBTRIP

CHANNELCTRIP

CHANNELE0TRIP

El- SYSTEMD-~ACTUATION

1 of 2 X 2 Logic

2 of 4 Logic

1 of N Logic

CHANNELATRIP

CHANNELBTRIP _ Y

XINCHANNEL CTRIP

CHANNELN C

CHANNELA NTRIP A

CHANNELATRIP

CHANNELCTRIP V

CHANNELN

Xof N byYofMLogic

'YSTEMACTUATION

V nf NI I 0%lF^ -G .. ,-Vsp

S & R/ ReactorACAS 8/27/91 25

J

Performance Program $D

EPRI/NPD

Development Of Analysis InputsFor Component Networks

* Analyses Typically Simulate OneComponent - Even for ComponentNetwo rks

-* Component Performance can beStatistically Based (95% Probability)

* Network Performance is Superior toIndividual Components

* Equivalent Network Performance canbe Defined

* Network Inputs can be DevelopedConsistent with Deterministic orStatistical Combination ofUncertainties Methodology


EPRI/NI mmr -

ru UPLANETS METHODOLOGY

r --------------------------- _ _ _ ICODE INPUTS

GROUP DISTRIBUTIONDEFINITIONS OF INITIATING

;Norninal Sepooint PARAMETERSI ana Numoer o0 (By Component

Componens Group

----- r - -- -- -- -- -- -- -- -- - -- r -

, Il

NETWORKl ~ANALYSIS

(Evaluate Distributionfor Component Logic)

ProbabilityDistribution forLogic Networkas a Function

PLANETS of InitiatingCODE Parameter

I~~~ I

PROBABILITYTHAT THE

COMPONENTNETWORK WILL

INITIATE

=S & R/ Reactor Performance ProgramACRS 8/27/91 27

EPRI/NPDTYPICj AL INDIVIDUAL COMPONENT PERFORMANCE

0.7

0.6

o-

coco0:

C.5

0.4

0.3

0.2

0.1

0.0.4 -3 -2 -1 0 t

STANDARD DEVIATION

Probabilitv Densitv Function

.93 +4

1.0

m-i

ICn

0a-w

-J

0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0-4 -3 -2 -1 0 +1

STANDARD DEVIATION+2 +3 +4

Cumulative Distribution Function


EPRI/NPD

EX.AkMPLES OF NETWORK PERFORMANCE

m-

coJo

0

0.

a-wco

0-

C.

3ll

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

-4 -3 -2 -1 0 -,-1 +2STANDARD DEVIATION

Netwvork Probabilit% Densitv Function

+3 +4

-4

ACRS 8/27/91 29

-3 -2 -1 0 +1 +2 +3STANDARD DEVIATION

Network Cumulative Distribution Function

_ S & R/ Reactor Performance Program

+4

EPRI/NPD

DEVELOPMENT OF EVENT ANALYSIS INPUTSFOR SIMULATING COMPONENT NETWORKS

(Deterministic Methodology)

1.0 -

0.9

1-J

O0c:

P:-j

C2

3

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0-4 -3 -2 -1 0 +1 +2 +3 +4

STANDARD DEVIATIONNetwork Cumulative Distribution Function


EPRI/NPD

DEVELOPMENT OF EVENT ANALYSIS INPUTSFOR SIMULATING COMPONENT NETWORKS

(Statistical Combination of Uncertainties Methodology)

1.0

0.9

-

-J

F0

co

I--

M-

0.8

0.7

0.6

0.5

0.4

0.3

0.2

1 of 1I rinirt

2 of 4Logic

Equivalent Points/3 Parameter OCCD)

1 of I 2of 4

J.±9.2.43a ±+1.14ap- 2.00a .±+0.89a

}~~ ~ ~~~~~~ p -0.29jpi-2.00a p.-1.52ap.-2.43a 9-1.80a0.1

0.0-4 -3 -2 -1 0 +1 +2 +3 +4

I ~~~~~~STANDARD DEVIATIONNetwork Cumulative Distribution Function


EPRI/NPD

Combined Methodologies

* Overview

, Network Performance Overlap

* Overlap Example

* Code Relationships

* Evaluation of Statistical Limits


EPRI/NPD

COMBINED METHODOLOGIES OVERVIEW

COMPONENT COMPONENT PROBABIITY ROBABILTTYNETWORK 1 NETWORK 2 DISTRUBUTION DISTRIBUTION

PROBABILITY ROBABILITY OR F ShyDISTRIBUTION DISTUTION ANALYSIS UMIT

I r I~~~~~~ji~~~~~~~~~-

OVERLAP ANALYSISF___________. _________________________. __ ___,I I I ~~~~~~~~~~~~~MOONSCODE'

It $1~~~~~~~~~~~~~I

PROBABILITY OF PROBABILITY OF

NETWORK ORPERFORMANCE ORE LEINEVNT

OVERLAP UI


EPRI/NPD

Network Performance Overlap

System actuation overlaps can occur

- Nominal setpoints in close proximity

- Large measurement uncertainties

- Associated with deterministic treatmentof analytical limit

Network performance -may be acceptable

- Low probability of overlap acceptablefor many applications

- Statistical assessment of overlapfeasible


EPRI/NPDNEETWORK PERFORMANCE OVERLAP

(\j

E

C

ej C U.0 .1)Z L%) ?ACL CC :.,t MQ zM 1; qr.z a

I

E

C

Nominal Selpoint 1- ..

-4 -3 -2 -1 0 -+1

STANDARD DEVIATIONDeterministic Setpoints

Nominal Setpoint 2

- +2 +3 +

0.8

0.7

0.6

-j

4:

0

0.5

0.4

0.3

0.2 -

0 .0 LC4 -3

ACRS 8/27191 35

-2 -1 0 +1 +2 +3STANDARD DEVIATION

Network Probability Density FunctionS & R/ Reactor Performance Program

EPRI/NPD

COMBINED METHODOLOGIES RELATIONSHIPS

r---------------------------------------…-________iI CODE INPUTSI

GROUP DISTRIBUTI1DEFINITIONS OF INITIATIN

(Nominal Setpoint PRESSURESana N.mber of (By ComooneCo.moonents) Group)

-- ---_--------- --________ _

l~~~G RESULTS OF STATI!C SYSTEM PROPMnt TRANSIENT OF II

CALCULATIONS PARAM

_~~~~~~~~~

STICAL IERTIES IINPUT IIETERS

- -I- ---

I-

II -NET1I ANA

(Evaluatefor Compc

: PLANETS£ CODE

.-- - -

MORKLYSIS

Distributiononent Logic)

IIIIIIIIIIIIIIIII

.ItIIII

.i

I-I

IIIII

RESPCSURF

COMM~f

(Least S. Evalw

STATISANAL

(Monte Carof Param

ProbabilityDistributionfor EventLimit

_-______________---r

)NSE tACEUCTION _

;quaresation)

PolynomialCurve Fit

;TICAL.YSIS

rlo Sampleeter Set)

Probability Distribution Ifor System Analysis

STARSCODE

__ __ _ __ _I

[LAP.YSIS

mit andknalysisIntegral) MOONS

CODE '

WO,

-Y

I- - - -

II

OF -"N II

PROBABILITYTHAT AN EVENT -4EXCEEDS LIMITS I

I

4.....~~~~~I

OVEFANAL

(Event LiSystem dOverlap I

I I


EPRI/NPD

[

0.06

0.05

0.04 -

< 0.030X 0.02

0.01

0.00

WALUATION OF STATISTICAL LIMITS

FIGURE OF MERIT

Probability Density Function


EPRI/NPD

(1 EPRI Setpoint ProjectsMethodology and Applications

OVERVIEW

EPRI Setpoint Methodology

.E4EPRI Setpoint Projects

Oyster Creek -Hich Pressure Setpoints* Application Overview* High Pressure Setpoint Relationships* Current and Proposed Setpoints* Analysis Considerations'Licensing Basis Events Selection Process* Plant Performance Events Selection

Process* Event Assessment* Analysis Results* Approach to Safety Valve Opening

Probability* Conclusion


EPRI/NPD


Vermont Yankee Reload Analysis* Application Overview

Approach and Results

Diablo CanYon-Safety Valve Tolerance* Application Overview* Approach and Results

TMI- 1 Flux/Flow Setnoints- Application Overview* -Current Flux/Flow Setpoints* Proposed Flux/Flow Setpoints* Code Sequence* Project Status

Palo Verde Safetv Valve Tolerance* Application Overview* Project Approach


EPRI/NPD

Oyster Creek High PressureSetpoints Overview

* High pressure setpoints did notprovide sufficient margin forinstrument drift and calibrationtolerances

* Safety valve setpoint drift resulted infrequent and costly testing of valves


EPRI/ I IMI r4 1 L

OYSTER CREEK HIGH PRESSURE SETPOINT/TECHNICAL SPECIFICATION RELATIONSHIPS

1 090 _ EMRV (,CE) Tech. Spec.r-----i 1 _

<- - 080 - - - - - - - - - --- EMRV (B,C,E) Setpoint2 5 1080

-EMRV (AD) Tech. Spec.

_7 a' t l.

iiii

POENTIALRL

REGION

1020

ISU = Instrument System UncertaintiesD = Instrument Setpoint Drift


EPRI/NPD

OYSTER CREEK HIGH PRESSURE SETPOINTS

1250 -

1240 -

1230 -

1220 - BANK

1210 -B BANK

1200 - i S.VS.V. -ANK

BANK Prga #4

1190 # 71 BANK

1180 - BANK#2

s.v.-~1170 BANK

#1

1100

1090 rI ~~~~~~EMRV

1080 GRP. #2

1070 -VR

1060 - GRP. #1

I I I ~~~~~~~~~C/RPT1050 ~~IC/RPT EMRV

GRP. #1 CA

1040 SCRAM

Current Setpoints Proposed Setpoints

S & RI Reactor Performance Program 00ACRS 8/27/91 42

EPRI/NPD

Analysis Considerations

License Basis Requirements

- Evaluation of recirculation pump trip andalternate rod insertion for high pressuresetpoi-nts for anticipated transientswithout scram

- Opening no more than one relief valvewithout scram

- Analysis to establish minimum criticalpower ratio operating limit

- Demonstration peak system pressureacceptable for proposed setpoints


in n r IKI r% I*Op rI11111ru X

Analysis Considerations

Plant Performance Considerations

- Low probability of relief valve openingprior to isolation condenser initiation

- Low probability of alternate rodinsertion prior to isolation condenserinitiation

- Low probability or recirculation pumptrip prior -to scram

- Low probability of relief valve openingprior to scram

- Low probability of a safety valveopening as a result of an anticipatedoperational occurrence

S & R/ Reactor Performance Program °ACRS 8127/91 44

EPRI/NPD

LICENSE IBASIS EVENT SELECTION PROCESS

_

[ GENERIC LICENSEBASIS EVENTS

TO BE CONSIDEREDI

ELIMINATE ALL0 EVENTS TMAT ARE

NOT APPLICABLETO OYSTER CREEK

U.~

IDENTIFY ALL EVENTSTHAT ARE NOT LIMITING

FOR OYSTER CREEK

.4nLICENSE BASISEVENTS TO BE

ANALYZED

IDENTIFY ALL EVENTSREQUIRING FURTHER

EVALUATION INCLUDINGTHE POTENTIAL IMPACTOF SETPOINT CHANGESK 2

S & R/ Reactor Performance Program °ACRS 8/27/91 45

i

EPRI/NPD

PLANT PERFORMANCE EVENT SELECTION PROCESS

-

GENERIC PLANTPERFORMANCE EVENTS

TO BE CONSIDEREDIN

4ELIMINATE ALL

EVENTS THAT ARENOT APPLICABLE

TO OYSTER CREEK

ELIMINATE ALL LOWPROBABILITY EVENTSFOR OYSTER CREEK

I

IDENTIFY ALL EVENTSTHAT ARE NOT LIMITING

FOR OYSTER CREEK

.~~~

U PLANT PERFORMANCEEVENTS TO BE

ANALYZED

-

.

IDENTIFY ALL EVENTSREQUIRING FURTHER

CtlA I I IATldQJ ItfflI I lhIJV:l C 1L0S I Iva I %OI & I%

THE POTENTIAL IMPACTOF SETPOINT CHANGES


EPRI/NPDAssessment of BWR Events thatReach High Pressure Setpoints

AssessmentTransients Plant Performance

Inacvertent High Pressure MakeupSystem Start Not Applicable

Pressure Regulator Fail!-re-Closed Low ProbabilityGenerator Load Rejection with Bypass Not LimitingGenerator Load Rejection witnout Bypass Low ProbabilityTurbine Trip with Bypass AnalyzeTurbine Trip without Bypass Not LimitingMain Steamline Isolation Valve Closure AnalyzeLoss of Condenser Vacuum Not LimitingRecirculation Pump Trip Not LimitingRecirc. Flow Controller

Failure-Decreasing Flow Not LimitingInadvertent Opening of a

Safety or Relief Valve Not LimitingPressure Regulator Failure-Open Not LimitingLoss of Feedwater Flow Not LimitingLoss of Offsite Power Not LimitingLoss of Shutdown Cooling Not LimitingFeedwater Controller Failure Low Probability

AssessmentLicse Basis

Not ApplicableNot LimitingNot LimitingNot LimitingNot LimitingAnalyzeNot LimitingNot LimitingNot Limiting

Not Limiting

Not LimitingNot LimitingNot LimitingNot LimitingNot LimitingAnalyze

Accidents

Control Rod Drop AccidentMain Steamrine BreakSmall Loss of Coolant AccidentRecirculation Pump Seizure

LowLowLowLow

ProbabilityProbabilityProbabilityProbability

Not LimitingNot LimitingNot LimitingNot Limiting

SpecIal Events

Shutdown from Outside the Control RoomOverpressure Protection AnalysisAnticipated Transients without Scram

Low ProbabilityLow ProbabilityLow Probability

Not LimitingAnalyzeNot Limiting


EPRI/NPD

Analysis Results

License Basis (License Basis Methodoloav)|- Recirculation Pump Trip and Alternate Rod Insertion

Performance Acceptable- No Overlap of Relief Valve and Scram Setpoints- Current Minimum Critical Power Ratio Operating Limit

Acceptable- Peak System Pressure Below Safety Limit

Plant Performance (PLANETS Analysis)- Probability of Relief Valve Initiation Prior to Isolation

Condenser < 7X10-4 per Event- Probability of ARI Initiation .Prior to Scram < 10-6 per

Event- Probability of Recirculation Pump Trip Prior to Scram <

5X10-2 per Event- Probability of Isolation Condenser Initiation Prior to

Scram < 5X10-2 per Event- Demonstration of Acceptably Low Probability of Safety

Valve Opening for Anticipated OperationalOccurrences Most Difficult Task


EPRI/NPDANALYSIS PROCESS FOR SAFETY VALVE OPENING PROBABILITY

CODE INPUTS…__________I

-GERIOUP DISTRIBUTIONDEFINITIONSOF OPENING

(Nominal Se:ooin: PRESSURESa-c Numnoer of eSafe:y VaesBy Va ve Gr :

_. .. _

--- r-- -- -- -- -- -- -- -- - -- I -I I

I II

RESULTS OFSYSTEM

TRANSIENTCALCULATIONS

N f.

STATISTICAL IPROPERTIES

OF INPUT IPARAMETERS I

…~~~~~~~~~~~~~~~~~~~< _~~~~~~~~~~~~~~~~~

_____~~ -r - - -

I

RESPONSE* SURFACE

-CONSTRUCTION

. (Least SquaresEvaluation)SAFETY VALVE

ANALYSIS

(Evaluate PressureDistribution for First

Opening Valve)

PolynomialCurve Fit

IIIIIIIIIIIIIIIIIIIIIIIIII

ProbabilityDistributionfor Pressureof First OpeningSalety Valve

- - - - - - - - - --_ _

STATISTICALANALYSIS

(Monte Carlo Sampleof Parameter Set)

PLANETSCODE

-

Probability Distributionfor Peak SystemPressure

STARSCODE

-_- _____- __- __- __- __-

I;

II

IIII

I- - -

PROBABILITY ITHAT A TRANSIENT

__ I... .I

4 4 t 1OVER LAPANALYSIS_ . o_ . , .lI

I OPENS A ;tzatety valve ana II SAFETY VALVE I System Pressure I

I Overlap Integral) MOONS II CODE I


EPRI/NPD

Conclusions

* All License Basis Requirements Were Satisfied forthe Proposed Changes to the High PressureSetpoints for the Demonstration.

* No Other Changes to Any Other Current Plant.Operating Limits or Technical Specifications AreRequired.

* The Plant Performance Requirements Were Satisfiedfor the Proposed Changes to the High PressureSetpoints for the Demonstration.

* The EPRI RASP Setpoint Guidelines Can Be Used inEstablishing Complex and Interrelated Setpoints.

* Statistical Combination of Uncertainties MethodologyCan Be Used to Reduce the Excessive ConservatismInherent in the Deterministic Analysis Process.The Statistical Combination of UncertaintyMethodology is Relatively Simple to Apply.

* Analysis Capability for Complex Valve Networks Is anImportant Contribution Made by this Demonstration.

* The Setpoint Demonstration Program Identified aNumber of Important Lessons Learned.The Experience Gained Th'ough the Performance ofthis Demonstration Should Be Invaluable to OtherSetpoint Applications.

S & R/ Reactor Performance ProgramACRS 8/27!91 50

crlnfl% I t rll00- rml,^v~ru V

Vermont Yankee Reload AnalysisOverview

* Substantial Economic Benefits for HighEnergy Reload Core Design

* Planned High Energy ReloadConstrained by Minimum Critical PowerRation Operating Limit


EPRI/NPD

Vermont Yankee Reload AnalysisApproach and Results

* Reduced Critical Power Ratio MarginRequirements by.use of 1-D KineticsAnalysis and EPRI Guidelines

* Scram Speed Determined to be MostSensitive Analysis Parameter

.*. .Applied EPRI's Statistical Combinationof Uncertainties Methodology andCombined Scram Speed and ModelUncertainties to Justify SimplifiedAnalysis Approach

* NRC Approval - October 1989


0

__- [=n01/K1r3r,% -

Diablo Canyon Safety Valve ToleranceOverview

.ObjLectives

- Relaxation of Lift Tolerance on. Pressurizer and Steam Generator

Safety Valves

- Justify Increase of ToDIerance to ±3%

Benefits

- Fewer Safety Valve Surveillance TestFailures

- Reduction in Plant Down Time for ValveTesting and Maintenance

- IncreasEL in Operational Flexibility


EPRI/NPD

Diablo Canyon Safety Valve ToleranceApproach and Results

* Applied EPRI Reactor Analysis SupportPackage Guidelines to PerformAnalyses for Setpoint Modifi-cation

* Used PLANETS Codelto DetermineSafety Valve Opening Probability

* Analyses Demonstrated that VesselOverpressure Protection Requirements

-Satisfied for increased Safety ValveSetpoint Tolerance

* Statistical Analysis Demonstrated thatIncreased Tolerance Covered byPrevious Conservative Analyses

S & R/ Reactor Performance ProgramACRS 8/27,'91 54

EPRI/NPD

TM-1 Flux/Flow Setpoints

Overview

* Increase in Flux/Flow OperatingMargin Desired

* Reduction in Potential for SpuriousScrams


EPRI/NPD

CURRENT FLUX FLOW SETPOINTS FOR TMI-1

---------

1i20

-30. 5. 1 0 B (24.5.

At-50.0.70.0)

(-50.0.42.6)

(-50.0,15.1)

/ /

04 AAcceptable

i Operation 1IC 1

I II 1

1 ;-3 .80 60 (24.5.80.6)1

I Acceptable3&4 Pump

I Operation I

I I

-601 -30 C.53-l (24.5.953.1)1

108)

01II

II

IIIIII

Acceptable2 3. & 4 Pump

Operation C

40.'

L-

4I. I

) ~~~~~~~I

I I

I I I~

(45.0.70.0)

(45.0X42.6)

(45.0,15.1)

1. 6

so 60

I I-,

-60 -50 -40 -30 -20 -10 0 10 20

Axial Power Imbalance (%)30 40


EPRI/NPD

ll PROPOSED FLUXFLOW SETPOINTS FOR TMI-1

.~~~ -0 - - - - - - - - - - - - - - - - - - - - - - - - - - - -I AcceptaDle4 Pump

Operation

-50.0.85.01--…-…-…-…-…-I Accepiaole ~~~~~~Supponted by

3 I& 4 PUMp Analyses inOperation this Report

I 60

IIIIII11II

. IIIIIIIIII

I

(45.0.110)

(45.0.85.0)

(45.0,55.0)

II

(*50 .0.55.0)L …-----------------4

Acceptable2.3. & 4 Pump

OperationII 40-

I

II

1 20 -

I I I II

I l

IF

-

-

a

C0.-

J

a_

…~~~~~~~~~~~~~~~~~~I

D I

D I

D I

i I I. I I

-60 -50 -40 -30 -20 -10 0 10Axial Power Imbalance

20(%)

30 40 50 60

Requires Maneuvering-…-…-…-…-…- - fnrp rr.nmma

Justification


EPRI/NPD

TM I-1 CODE SEQUENCE FOR FLUXI FLOW ANALYSES

i ,.~~~~~~~..... 0.....I II I

M ANEAN'Ei'J'ER NG A,'ALY C-- S T. ,- v -Ad ------ - '-SIMULATE * 3P Ad - - -

-

a - aa a,

II0

I -. 1. - --- . I -

I II If I* - - - -e - -- ~~ 0

I~~~~~~~~CASMO

I II I

I I

-0A..

TWO-GROUP, I

-DNUCLEAFCONSTANTS

Analysis Process to Justify Setpointsfor 0 Axial Imbalance

Maneuvenng Analysis Required to- . . . Complete Justification of Flux/Flow

Setpoint Change

S & R/ Reactor Performance Program001

ACRS 8/27/91 58

EPRI/NPD

iI

TMI-1 Flux/Flow Setpoints

Project Status

Phase 1 _Comnete

* Safety Analysis for 0 Axial ImbalanceAcceptable

* Documentation Being Prepared

Phase 2- Beina Initiated

* Maneuvering Analysis Necessary

= S & R/ Reactor Performance ProgramACRS 8/27/91 59

EPRI/NPD

(f Dnlt%Verde Safety Valve Tolerance II "AW

Overview

* Test Data on Pressurizer and MainSte-am Safety Valves Indicate.Drift

* Technical Specification"Tolerance of±1 % too Restrictive

* Tolerance Relaxation to ±3% Desired

*'Requires "Demonstration of Adequacyof Overpressure Protection

* Requires Demonstration of LowProbability of Safety Valve OpeningPrior to Scram


EPRI/NPD

Palo Verde Safety Valve Tolerance

Project Approach

* Use EPRI Statistical Methodology to-Justify Larger Uncertainty in Valve"Opening Pressure

* Use PLANETS Code to Simulate theBehavior of the Valve Network

. Quantify the Overlap in the :ProbabilityDistributions of the Safety ValveOpening and Scram Occurrence

* Use License Basis AnalysisMethodology to Justify AcceptableOverpressure Protection

S & R/ Reactor Porformance ProgramACRS 8127/91 61

A~TTAHcmEN 2

- F998 JIMA22.P 10826 B196-XE

N

'N

Nac ,...

.. .:, .

F998 JIMA2.P 10826 19"E

MA-191NN

fff-la Ea 1#1 are

111RO

01

L srAl-

FM9f JIMA3.P 108261 196-XE

LUN[EPNE81K JWHlb PI P[A( n O i3Qr-f (ON

~~~~~~~~~~~~~~........ ........ .. :L_

F998 JIMA4.P 1082S 8196-XE I

D> Tca Mselv Fmc arir .TPA

D> To=2 $ G@ t Pro--dums

Df TRhP Bs

D> PSTS

DE PawsD>val

9i )

dpi

fbLCDi;

'MD

(E)

f2-

.g

AS

EbbC

9b-t)I

�LQ-

A A

V V

04

v

Ic--va,

RU

5:~3

s 3 ...

.,

F998 JIMA7.P 10826b l9r-XE

TRU B-AE

1> O~ns a&,P ]°uST %,n %a

F998 JIMAB.P- 10826LI1I96-XIE

LOS ThOP8

NAWM mmmm m Uffli"m AAMWr----Wa ml Ur--WA- oil

1-h! � I m gm' --o"I

.1 �IMOIR& r I A-

I% MmEft molow m I ... mmummMr_--", Iff-r-4�a_ INS_w= mpfffflug

&M M-M

F998 JIMA9.P 10826 B196-XE

.pt

D> uaa b (dewslp $b q$wruE$ *~oM gm

F998 JIMA 1I.P 10826 B196-XE

> Las,$PTa

to Epa > Los PaTS

D> LL$ PSTHa >TRIPS

1> LOS FPSTOH enwbolsH

F998JIMA16.F 10826 8196-XE

Rae

LONEROCK PLAN $SPECOFO7ECHN CAL SURDELONES= P J\~ .O.Q.

SPILuL

SOECURE. p*o b diwmeow Aft

, F998JIMA19. 10826 8196-XE

LN EAnROC

Ar~mm I& Mv. 14-l

EPO Toc pt7Xa 3FOORJLNU~

~~~~~~ W $D1J

LGS OST STEP

SP/0alJ

D Xt l~~vrb$u" oft f

- Shyn Ad r0t ¢$ha

WM M*"5 Imo oavm mmby~~~- i $1 an o m7d bteo

F998 JIMA20.P 10826 U19&-XE

9pays 7 TH3OP CN En3sO©Q O E©E35

LGW/TRWP .E . VW P. a

LWI-MIPMP

.... w|zlwluwlel~,,,,l....l~ll,,,,l,,l .. J.arpI-1 .

It

T~btho kw aammm hM tSI- b wwsf mmwmmf-

F99OU JIMA I lk' 10826 DI196.XE

LQR TIMPS

I

I v 4 I

Is an fmoom

F998 JIMA12.P 10826 B196-XE

D~~~~~L 1Y ChitO

*-a

-~~M np

D>ow m ~ c o ms-mTl-(i -wp au

D>E u poe-m m,

F998 JIMA13. 10826 B15iXE

LaS ThOPn

mom"m � ft I& M"Mt. Allft M&', 'ItIm W& I&N,--j19I.- rl affpmm mlF- %,-ffl I

ANW&

MmnmhhL AM 1� .Aloft

I

e, a

' 0g

.e'w

eCO)

S0

z

9=

lg

C

(0

0f

IM

A A

i

=itno

i

14)I

A

*-'N

0~

s&

9o--

Ad 1

K-. Xp

ATTAhCMENT 3

EDSFI BRIEFING TO ACRSAUGUST 29. 1991

SCOPE OF EDSFI

- ADEQUACYSYSTEM

- EVALUATESUPPORT

OF THE ELECTRICAL DISTRIBUTION

ENGINEERINGCAPABILITY

AND TECHNICAL

* TEMPORARY INSTRUCTION IN USE BY REGIONS

* TO ACHIEVE CONSISTENCY OF TI IMPLEMENTATION

- FIVE PILOT INSPECTIONS CONDUCTED BY- TRAINING PROVIDED AT TTC- DRIS PARTICIPATION IN REGIONAL EDSls

DRIS

* STATUS OF TI IMPLEMENTATION

- 24 INSPECTIONS COMPLETED- TARGET COMPLETION OF TI -

AS O0 AUGUSTJANUARY 1993

30, 1991

PAGE 1

i

EDSFI BRIEFING TO ACRSAUGUST 29, 1991

* REGIONAL COUNTERPART MEETINGS HELD APRIL 24 ANDSEPTEMBER 5, 1991

* REGULATORY INFORMATION CONFERENCE EDSFI SESSIONMAY 7

* INFORMATION NOTICE 91-29 ISSUED

- ADDRESSES INITIAL FINDINGS FROM EDSFIs

* DEGRADED GRID VOLTAGE* INADEQUATE SETPOINTSE INADEQUATE TESTING OF

RELAY SETPOINTS

CIRCUIT BREAKERS

PAGE 2

...

r

ELECTRICAL DISTNRIBUTION FUNLTINAL INSPRECURRING PROBLE-MAllEAS

* DESIGN DEFICIENCIES

* PROCEDURAL WEAKNESSES

* CONFIGURATION MANAGEMENT

3

l

r

ELECTRICAL DISITRIBUTIN FUNCTIONAL INSPRE RRINa PRBLEM-AREA

* ELECTRICAL DESIGN IEFICIENCIES

- FAULT PROTECTION* UNDERSIZED AC 4* LACK OF SHORT

& DC CIRCUIT BREAKERSCIRCUIT ANALYSES FOR DC CIRCUITS

- VOLTAGE REGULATION* GRID UNDER VOLTAGE CONDITIONSa DIESEL GENERATOR TRANSIENT LOAbING* LOAD SEQUENCING

CONDITIONS

- BUS TRANSFER

- BREAKER COORDINAtION

- RELAY SETTINGS

- PHYSICAL SEPARATION

- RACEWAY FILL

4

i

.

ELECTRICAL DISTRIBUTION FUNCTIONAL INSPRECRRING PROBLEM ARE

MECHANICAL DESIGN DEFICIENCIES

- TEMPERATURE RELATED PROBLEMS* CONTROL OF MINIMUM BATTERY ROOM TEMP. EQUIPMENT QUALIFICATION

MINIMUM COOLING WATER TEMPERATUREMAXIMUM AMBIENT TEMPERATURE

- DIESEL GENERATOR FUEL OIL SYSTEM* CAPACITY OF FUEL OIL STORAGE TANK* FUEL SUPPLY AVAILABLE IN DAY TANK* TORNADO PROTECTION

- SEISMIC QUALIFICATION* FUEL OIL TRANSFER SYSTEM* DG LUBE OIL AND JACKET WATER HEATERS* FIRE PROTECTION tYSTEM IN DG ROOM|* RELIEF VALVES ON bG AIR START ACCUMULATORS

i

I

ELECTRICAL DIS~TRIBUTION FUNCTIONAL INSPRECURRIFING PROBLEM ARIEAS

* PROCEDURAL INADEQUACIES

- LACK OF ACCEPTANCE CRITERIA* MAINTENANCE PROCEDURES* BATTERY TESTING

- ERRORS OR OMISSIONS* INCONSISTENT WITH VENDOR RECOMMENDATIONS* EOP ELECTRICAL LOAD TABULATION

APPROPRIATE PROCEDURES DO NOT EXIST. FUSE CONTROL* MAINTENANCE

RELAY CALIBRATION

-FAILURE TO IMPLEMENT EXISTING PROCEDURES

i

o- Om

r

-~~~~~~~~~~~~CELECTRICAL DISTRIBUTION FUNcRIONAL INSPR.Eg PRIN PR BLEM AREA5

* CONFIGURATION MANAGEMENT

- LACK OF DESIGN BASES* BREAKER/RELAY SETTINGSa TRANSIENT ANALYSIS OF VOLTAGE

DURING LOAD SEQUENCINGPROFILE

- PLANT DOES NOTa BREAKER/RELAY

CONFORM TO DESIGN DOCUMENTSSETTINGS

7

j

I.

EDSFI BRIEFING TO ACRSAUGUST 29.1991

* TWO LEVELS OF UNDERVOLTAGE RELAYS

* PSB-1 REQUIREMENTS

- THE SELECTION OF UNDERVOLTAGE AND TIME DELAYSETPOINTS SHALL BE DETERMINED FROM AN ANALYSISOF THE VOLTAGE REQUIREMENtS O0 THE CLASS 1ELOADS AT ALL ONSITE SYSTEM DIST1IBUTION LEVELS

* GRID STABILITY

* 7 OF 21SECOND

PLANTS INSPECTED HAVE HADLEVEL DEGRADED GRID RELAY

INADEQUATESETPOINTS

PAGE 8

i


* PROGRAM RESULTS

- LICENSEE'S CONDUCTING THEIR OWN EDSFIs- INCREASED ATTENTION TOWARDS EVALUATING

AND IMPROVING THE DESIGN BASIS OF THE EDS- IMPROVED THE OVERALL FUNCTIONAL CAPABILITY

OF THE EDS

PAG E 9

r


* CONCLUSIONS

- EDSFI FOUND DEFICIENCIES THAT WERE:

* INTRODUCED* PRESENT IN

BYTHE

PLANT MODIFICATIONSORIGINAL DESIGN

- INDICATES NEED FOR BETTER

* ENGINEERING AND TECHNICAL SUPPORT* LICENSEE SELF-'ASSESSMENT PROGRAMS* UNDERSTANDING OF DESIGN BASES* AVAILABILITY OF bESIGN DOCUMENTATION

PAGE 10

meeting summary, held on 08/29/1991, the acrs subcommittee

Documents