Statistical Universal PowerReconstruction with Fixed MarginTechnical Specifications (SUPR-FMTS)Post-Submittal Meeting

Gayle ElliottJesse KlingensmithRich KochendarferDick Deveney

AR EVA

Presentation Overview

Oo Introduction GFE

Pp Background and Objectives RAK

Oo FMTS Background RAK

lo Changes to FMTS RAK

P SUPR JJK

lo Concluding Remarks JJK

Oo Summary and Next Steps GFE

AREVA Post Submittal Presentation - SUPR-FMTS - October 26. 2010 - D.2 AR EVA. .. . . . . . . . .. . . . . . . . . . . . . . .. . . . . ... I

Introduction

lo Discuss the SUPR and the FMTS methodologies and theirapplication to PWR-type reactors

No Obtain timely NRC feedback and interactions to support

regulatory review process

OP Respond to preliminary NRC feedback to cover:

1. The scope of the review as stated in Section 1.4. The TR addresses the technical issues associatedwith Super-FMTS, yet requests approval of the FMTS methodology with regard to Westinghouse andCE plants. Figure 1.1 contains insufficient information with regard to a comparison of the performanceof FMTS vs. Super-FMTS methodologies.

2. The NRC did not find discussion of the W, CE and B&W measurement systems. The interactionbetween FMTS and Super-FMTS appears to be central to the request for approval and there arecommon elements to the methodologies.

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.3 ARE EVA,

Background Discussion

Io PWRs presently measure RPD with one of severalinstrumentation systems

> Movable Detectors (TIP, etc.)

* Fixed Incore Detector (FIC)

* A combination of the two (Aeroballs and FICs)

Oo Used for monitoring operation of the core with periodic userequirements

Oo Preserves initial peaking assumptions for all events (LCOlimits) by means of monitoring global power

Axial Flux Difference (AFD), Quadrant Power Tilt (QPT), Rod PositionLimit (RPL)

-AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.4 AR EVA

SUPR-FMTS Objectives

Pp Support of multiple/diverse core power distributionmeasurement systems, e.g. TIP, FIC, Aeroball

c Kriging, RPD Check, Online Simulator

Oo Preserve safety analysis assumptions through directmonitoring of margin to power peaking limits rather thansecondary indicators (AFD, QPT, RI)

Q FMTS (Fixed Margin Technical Specifications, BAW-10158P-A)

}-

LZI

.I Extension of flux map surveillance intervals in plants withinterval measurements (non-continuous, e.g. TIP)

<* RPD Check, Online Simulator

c RPD Check use is optional. Need is determined by uncertainty analysisJ

Cn)LZ

U.o

=AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010- p.5

FMTS Background

A

Tech Spec MonitoringRequirements for FMTS

No Monitor Limiting Conditions of Operation (LCO)

0 LCO only -> no change to Limiting Safety System Settings (LSSS)

lo Monitor peaking rather than an overly restrictive combinationof RPL, AFD, and QPT parameters to preserve peaking limits

O Assure that the core is operating as designed

0 Preserve design basis for LSSS and backup LCO limits

Op Define actions when criteria are exceeded as determined bythe licensing basis

AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.7 ARkEV

Peaking Margin for Direct &Indirect Limits

In(lirect DireCt (FMTS)

Limitoperation margin for.steady, state effects t7

Worst:

.V(z). time inxenonrod posquadrai

-operating margin'fortransientý &steady state-effects,cycle

di stribution

;itionnt tilt

419r

4

SUPR-FTMSSystem:uncertaintymeasurement:or. inferred measurementat.all times

Power Peak Measurement-.(stea (lyý state- oilly)

.measurement'.uncertainty*

.4AAREVAAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.8

Monitoring RequirementsFMTS affects some, but not all, LCO

LCO Limits LSSS Limits

Monitored Directly Unchanged Unchanged

IJnsertion) insertion)

A•AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.9

Nuclear Instrumentation(RPD) Systems

O Categorize measurement systems according to:" Frequency" Data densityK> Directness

t Need to combine all measurementsK> Consider strengths & weaknesses of each measurement system

N Example - TIP v. FIC v. Thermocouple v. Core Simulator

Core Simulator

Futt reTIP

Aeroball

'a FIC• [T/C

Exe )relow.

-tr

low high

frequency AAREVA Post Submittal Presentation - SUPR-FMTS - October 26. 2010- p.10 •AR E•VA..................................... ... ) I

FMTS Modifications

,AtAIR.EV"A

FMTS FlowchartFigure. 1.2-1 FMTS. Monitoring .Procedure

Core

3D -PowerDistribution

Signal n -Power ,Margin Ratio and

Condition'ing :Conversion Calculation Alarms

raw Limitingsigaw s.Peakingsignals. Criteria

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.12 AREVA

Comparison of FMTS andSUPR-FMTS

Amfiw

- -an5MMMlI-A

laW Ir CE

AAREVAAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.13

Proposed Clarification

1. The scope of the review as stated in Section 1.4. The TR addresses thetechnical issues associated with Super-FMTS, yet requests approval of the FMTSmethodology with regard to Westinghouse and CE plants. Figure 1.1 containsinsufficient information with regard to a comparison of the performance of FMTSvs. Super-FMTS methodologies.

2. The NRC did not find discussion of the W, CE and B&W measurementsystems. The interaction between FMTS and Super-FMTS appears to be centralto the request for approval and there are common elements to themethodologies.

Topical Text, from Section 1.4:Application of the margin monitoring portion of the FMTS methodology toWestinghouse (W), EPR, and Combustion Engineering (CE) plants, as well as thepreviously approved Babcock and Wilcox (B&W4) plants [6]. This extension is based onthe previous approval for Babcock and Wilcox plants, the improvement in bothdirectness and frequency of power distribution monitoring, and the improvedquantification of measurement system uncertainty through the use of the Monte Carlosimulation methodology for determining system uncertainty.

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26,2010 - p.14 AR EVA

Characteristic UncertaintyDiscussion

Po Quantify Uncertainty in:

K* Measurement systems (r Po7tP t Disribntion

tcrs Mcasurcment* Plant parameters 3

SReconstruction methodsetes

AAiR•EVAAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.15

Determination of Total SystemUncertainty

Oo Sample uncertaintiesusing Monte Carlosimulator

O Calculate the effect ofuncertainty during "hardto measure" events

Oo Quantify theuncertainty,

total/system

[2

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p. 16

Application to Meet Tech Spec (Requirements

SUPR section produces RPDmeas and process uncertainty

FMTS section compares adjusted RPD to limit criteria with margincalculation

Limits based on available margin

Action required when negative margin condition occurs -> moverods, and/or lower power to restore margin

Monitor operation relative to design to ensure the basis for LSSS &backup LCO limits are preserved.

AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010- p.17 AR EVA

I I

I I

m

0

Cn

-!n

a)0

Cy)CD

N3

C,)

00

CD)

0

CD

0

21Note; LCO fuel design criteria are

a. During a LOCA, peak cladding temperature must not exceed 22000 F.b. During a loss of flow accident., there must be at least 95% probability at the

95% confidence level (the 95/95 DNB. criterion) that the hot fuel. rod in thecore does not experience a DNB condition.

q. During an ejected rod accident, the fission energy input to the fuel must notexceed 280 cal/gm.

d. The control rods must be capable of shutting down the reactor with a mini-mum required SDM with the highest worth control rod stuck fully withdrawn..

Margin Calculations with FMTS

Calculate margin as:

MIVIt,k. = (1 -

RPDi,jk" Rioca" * FppFlex} 100Li,j,k

Where F.,c includes the total system uncertainty and the adjustedprocess variance.

E

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.19

SUPR - Statistical UniversalPower Reconstruction

AA!R EVA

Power ReconstructionMethodology (SUPR) Overview

0 Reconstruct power with localized kriging models

Oo Dynamically calculate the process variance of the local model

Oo When using infrequent power measurements - RPD Checkroutines are used to:

* Calculate Assembly Exit Thermocouple and ExCore detector responses

*• Compare calculated to measured responses

* Impose variance penalties if necessary

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.21 A4R.EVA.

The Model Illustrated

AA R!EVAAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p. 2 2

Background on KrigingProcess

lo Kriging

Named after Danie G. Krig, South African mining engineer who, in1951, developed empirical models of ore grade distribution based onsampled points

o Formalized by Georges Matherton

* Founder (1968) "Centre de Geostatistique et de MorphologicMath~matique" at Paris School of Mines in Fontainbleau

* Considered to be the "father" of spatial statistics

0o Example applications

* Interpolating grades of ore between measured points

* Intelligent combination of sparse, accurate measurements withplentiful, less accurate measurements (rain gauge and radarmeasurements of rainfall)

AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010- p.23 AR 1EVA...... .. . ... . ... s I

Interpolation with KrigingModels

Oo Form of the statistical model:

y(x) = g(x) + E(x)where

g(x) is the trend model

e(x) is the error, a random deviation from the trend

lo Assuming an unbiased model and introducing the covariancefunction, the model becomes:

Y(x0) = g(xo) + r(xo)TR'!(y _ g)

,AA'R 1EVAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.24

Interpolation with KrigingModels, II

Io Expanding the kriging model:

R(x0,xj)

R(x0 ,,XN)

R R(xi, x1)R x XN

R(xl,. xN,)

.R(XN,rx1) 1R(X1,XN) I

VA-AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.25

Variance Calculation

Oo The kriging variance is:o2(Xo) -Var [•(Xo) - y(Xo)]

lo Approximated as:

P"AAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.26 A!RE.V:A•

RPD Check - CalculateResponses

lo NOTE: Only used for systems with significant time intervals

between measurements

O Generate responses based on reconstructed power

O, Assembly exit thermocouple temperatures

Calibrate mass flow at measurement time

Calculated Tout using enthalpy balance

0 ExCore detector power signals

Adjoint weights correlate power to current

Calibrated values convert current to voltage

l Evaluate all responses w.r.t. kriging variance

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26. 2010 - p.27 A~REVA..... ............. . a I

RPD Check - CompareResponses

A R !E•VA•AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.28

RPD Check - VarianceAdjustment

R, •V VA,AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.29

RPD Check - VarianceAdjustment

ARAREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.30

Concluding Remarks

AAiR!E VA

Conclusions

SUPR-FMTS utilizes the present instrumentation systems, theiruncertainty analyses, along with the Technical Specifications (withminor modifications) and core monitoring approach.

SUPR provides a method to consistently use diverse core powerdistribution measurement systems to maintain the applicability ofthe safety analysis and take advantage of the strengths of eachdiverse measurement system.

FMTS monitors the core power distribution relative to theCondition II and LOCA peaking limits based on the three-dimensional measured power distribution.

SUPR-FMTS takes advantage of thermocouple and excore detectorsignals for plants with infrequent measurement systems (TIP) topossibly extend flux mapping intervals.

AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.32 AIR EV

Summary and Next Steps

OP Commence formal review and RAI process

O Approval of Statistical Universal Power Reconstruction

with Fixed Margin Technical Specifications (SUPR-FMTS)

Topical Report

AARE VA

AREVA Post Submittal Presentation - SUPR-FMTS - October 26, 2010 - p.33

List of Acronyms andAbbreviations

AFD - Axial Flux DifferenceCOLR - Core Operating Limits Report

ExCore - Excore neutron detector

FIC - Fixed In-Core detectorFMTS - Fixed Margin Technical Specifications

IC-DNB - Initial Condition, Departure from Nucleate Boiling

LCO - Limiting Condition of OperationLOCA - Loss Of Coolant AccidentLSSS - Limiting Safety System Setting

NRC - Nuclear Regulatory CommissionPWR - Pressurized Water Reactor

QPT - Quadrant Power Tilt

RPL - Rod Position LimitRPD - Relative Power Density

SUPR - Statistical Universal Power ReconstructionTIC - ThermocoupleTIP - Traveling In-core Probe

Tech Spec - Technical Specifications

AAREVA Post Submittal Presentation - SUPR-FMTS - October 26. 2010- D.34 A!R E-VA

w I