ML20033E825

Forwards Decription of Objectives of Peer Group & Fuel Cycle Data Studies & Preliminary Groupings of Westinghouse NSSS Units

ML20033E825
Person / Time
Issue date:	03/09/1990
From:	Hickman D NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:	Voytell K WESTINGHOUSE OPERATING PLANTS OWNERS GROUP
References
NUDOCS 9003140314
Download: ML20033E825 (11)

Text

.

t.

l

?.

u

, o xe N

J Mr. Ken Voytell p

Westinghouse Owners' Group Westinghouse Electric Corp.

P.O. Box 355 sPittsburgh, PA 15230

Dear Mr. Voytell:

-The enclosure provides a brief description of the objectives of the peer group and fuel cycle data base' studies.and our preliminary groupings of Westinghouse NSSS units..We have reserved Thursday, March 28, 1990, to meet with the WOG steering committee on this topic. Thank you for your assistance in this matter.

Sincerely,

/

Donald E. Hickman, Chief Performance Indicator Section Division of Safety Programs

' Office for Analysis and Evaluation of Operational Data Distribution:

TPAB RF TNovak MWilliams SPettijohn-DHickman

EDO, m"PDRd. 7 Utc :TPAB

TPAU -

.... :.....,. w...:...

NAME :Pif$fid0 Ogp:H -

DATE.= :J/R/901

3/.9 /90 OFFICIAL RECORD COPY l

9003140314 900309 PDR TOPRP EMVWEST C

PDC s

~.

l MAR 0 91990 Mr. Roger Newton-Wisconsin Electric Power Co.

231.W. Michigan Street-

- Milwaukee, W1 ~ 53201

Dear Mr. flewton:

. The enclosure provides a.brief description of the objectives of the peer group and fuel cycle data base studies and our preliminary groupings of Westinghouse NSSS units. We have reserved Thursday, March 28, 1990, to meet with the WOG steering committee on this topic.

Thank you for your assistance in this matter.

Sincerely, Donald E.

lickman, Chief Performance Indicator Section Division of Safety Programs Office for Analysis and Evaluation e

of Operational Data Distribution:

l TPAB RF, L

TNovak L

MWilliams SPettijohn DHickman E00 PDR l

l l

l 1

UFG :TFAB

TPAB

......:....h....:------......:---...--....:------------:-----....--:------------:----

' l'AME :P[(Tf

?gp:HI

' 3/j2/90-

DATE

)/9 /90 0FFICIAL RECORD COPY

' 1. '.,. -

o

' ~

===.

Background===

On March 15, 1989, the Nuclear Regulatory Comission approved the adoption of cause codes. extracted from Licensee Event Report (LER) data as a new performance indicator (PI).

The staff subsequently proposed displays for the cause code data that included a graph showing cause code data compared with the average cause codes for units having the same Nuclear Steam Supply System (NSSS) vendor (industry average).

The Commission, prior to approval of this display, directed the staff to address the comparability of LER data between licensees.

This project addresses the comissioners' concern regarding the desirability or.

I appropriateness of comparing cause code data derived from LERs between licensees on a global basis, recognizing that substantial differences may exist in plants even of the same NSSS design. The issue is to determine if groups of similar. plants (peer groups) can be established based upon sound technical considerations.

.The objective of this project is to construct groupings of plants based upon design similarities..Such groupings should support evaluations of individual plants relative to the k peer groups and of peer groups relative to each other.

Based on our statistical analysis considering the intended use of the peer group data, we.have determined that peer groups with six or more plants will' have relatively equal length confidence intervals.

Therefore, peer groups should have six or more plants. This assumes that all the plants in a peer

. group are consistent in terms of averages and standard deviations.

Inconsistencies among the plants in a peer group would dictate larger peer L

. group sizes..How much larger is dependent upon the inconsistencies.

Design Factors A detailed study of every design parameter for each operating reactor would conclude that-no two plants are identical, and therefore any effort to group L

them at this level would ultimately result in only one plant per peer group.

L Design parameters used to define peer groups for this study were of a more general or broad-based level so that several plants would satisfy the group criteria. Examples of broad-based design characteristics common to multiple plants include:

(1) NSS vendor-The NSSSs for all operating units are supplied by either Babcock and Wilcox, Combustion Engineering, General Electric or Westinghouse.

Each vendor has employed varying design concepts resulting in four distinct types of nuclear plants. Each has its own operational characteristics.

3.

.~

a p[

-a-

}

l

- (2) Product line-For a particular NSSS vendor, different generic designs (orproductlinesorvintages)haveevolvedovertheyears.

These product lines represent changes in major design parameters such as rated thermal output, number of coolant loops, containment design, and control or protection system technology. The changes from one product line to the next were generally of a significant enough magnitude that an effect on operational characteristics would -

be expected.

1 (3) Age-Another measure of plant design relates to the age of the plant as measured in years of commercial operation, year ordered, date of construction permit, date of operating license, year of intended operation, date of initial critcality, etc. This measure, also closely tied with product line or vintage, is not as technically based as those mentioned previously; however, it is an indicator of overall plant complexity which can have a significant impact on plant operational characteristics.

gerGroups(Preliminary)

.All of the nuclear units in the United States which are currently licensed or expected to operate have been-assigned to one of eight preliminary peer groups. The preliminary groups were constructed after careful consideration of design related information, minimum acceptable peer group sizes from a statistical perspective, and the unique features of several plants. The Westinghouse grouping is discussed in this section. The design information which formed the basis for many of the peer group assignments is given in Appendix A.

Westinghouse. Peer Groups There are 52 Westinghouse NSSS units located at 33 different sites. These units either entered comercial operation from July 9,1960 through May 20, 1989 or are expected to operate-(e.g., new units). One of'the major design characteristics of the Westinghouse NSSS that evolved over these years was the number of primary coolant loops. This single feature offers a valid consideration for peer group definition.

Subdividing the Westinghouse units by the number of loops results in three groups consisting of two-loop, three-loop and four-loop plants.

Each of the groups is discussed in the following sections.

Two-Loop NSSS.

Grouping the Westinghouse units by the number of loops, with certain exceptions, results in the two-loop Westinghouse group shown in Table 4.1.

There are three plants included in this group that do not have two coolant loops, Yankee Rowo, San Onofre 1, and Haddam Neck. These were included in this group because they more closely resemble the two-loop plants u

from an age,' size and corresponding complexity standpoint than they resemble the other three and four-loop Westinghouse plants.

y [.

.A L

3 Three-Loop WSSS. Grouping the Westinghouse units by the number of loops, with certain exceptions, results in the three-loop Westinghouse group shown in Table 4.2.

As noted above, San Onofre 1 was included in the two-loop peer category because its age, size and corresponding complexity more closely

~

resemble the two-loop-plants than the three-loop plants, four-Loop NSSS.

Grouping the Westinghouse units by the number of loops, with certain exceptions, results in the four-loop Westinghouse group shown in Table 4.3.

As noted above, Yaunkee Rowe and Haddam Neck were included in the two-loop peer category because their age, size and corresponding complexity more

-closely resemble the two-loop plants than the four-loop plants.

Interview Objectives i

L l

-The objectives of the interview are to (1) determine whether organizations j

have already addressed the issue of comparing plants based upon design factors and (2) obtain their input on the proposed groupings based primarily on design j

factors.

More specifically, (1) Would it be reasonable to assume that all Westinghouse plants would l

l.-

perform similarly, if all other factors were the same for each plant?

(2) Would product evolution, certain design features, complexity, etc.

j render such a grouping unreasonable?

- If so, would structuring of peer groups based on design class or vintage satisfactorily address these concerns?

- If not, then what peer groupings would be most appropriate?

i.-

(3)- For plants within the same peer group, what design differences exist l2 that might be expected to have a significant effect upon plant performance?

I i

m-

, 4, [*.]_,

c i'..

Table 4.1 Westinghouse Unhs wkh Two Loop NSSS s-Commercial Achitect Turbine Coolant Operatine Said D2rd$ $$- M loops 1!rdry Ope <ste 10R!0tir

.Qfni!A19r

?

Ginne 244 1520

'470' 2

09/19/68 07/01R0 GIL '.

WE Haddam Neck 213

.1825 382 4

06/30/67 01/01/84

$4W WE Newesnee 308 1660

$38 2

12/21/73 08/18/74 FLOUR WE

' Point Beach 1 284 1618 497 2

10/06R0 12/21R0 SECH WE-Point 8each 2 -

301 1818 497 2

03/08/73 10/01R2 BECH WE Prairie Island 1 282 1650 630 2

04/06/74 12/18#3 ' FLOUR WE Prairie Island 2

' 308 1650 630 2

10/20#4 12/21/14 FLOUR WE

' San Onotre 1 208 1347 438 3

03/27/67 01/01/08 BECH WE Yankee Rowe 38 000 176 4

07/08/80 07/01/51 S&W WE T8ble 4.2 Westinghous'8 Unhs whh Th7ee Loop NS$$

Coolant Operating Commercial Achitect Turtsne 333 Docket 83 Myf 1992)

Uconse Operation Enoineer Gene'ator Seaver Valley 1 3**

2652 635 3

07/02RS '10/01#8

$&W WE Seaver Valle-l 412 2600 433 3

08/14/87 11/17/47 S&W WE

- Fartey 1 348 2662 628 3

08/26/77 12/01/77 SSI/BECH WE i

fortey 2 384 2852 829 3

03/31/81 07/30/01

$$1/BECH WE Harris 1 400 2T75 ~ 900 3

01/12/87 06/02/87 EBASCO WE North Mna 1 MS 2276 807 3

04/01R$

06/08#8 S&W WE

North Mna 2 338 2276 907 3

08/21/80 12/14/80 S&W WE l

Robinson 2 281 2300 700 3

09/23R0 03/0701 EBASCO WE 3

Summet 1 395 2775 900 3

11/12/42 01/01/84 GIL GE Sutry 1 280 2441 788 3

06/25/72 12/22/72 S&W WE Sorry 2 281 2441 788 3

01/29/73 06/01R3

$4W WE Turkey Point 3 250 2200 893 3

07/18R2 12/14R2 BECH:

WE Turboy Point 4-251 2200 693 3

04/10R3 09/07/73 8ECH WE T8ble 4.3 Westinghouse Units with Four-Loop NSSS Coolant operating commercial kohitect.

Turtene Esm Derde M.MWe 1eens Lunw

_Qperation Eaniaeer Generator r

Braidwood 1 454 '

3425 1120 4

07/02/87 07/29/84 S&L WE Staidwood 2 457 3425 1120 4

06/20/04 10/17/88 S&L WE eron 1 454 3411 1120 4

02/14/45 09/14/85

$4L WE Spon 2 455 3425 '1120 4

01/30/87 08/21/87 S&L WE

' Callaway 1 483' 3411 1171 4

10/18/04 12/19/84 SECH GE Catawba.1 413 3411 1145 4

01/17/06 08/29/85 OPC OE Catawba 2' 414 3411 1145 4

06/15/08 08/19/88 OPC GE l'

Comanche Peak 1 44S 3411 1150 4

//

//

G&H AC Cook 1 318 3250 1030 4

10/2SO4 08/27R5 - AEPSC GE Cook 2 318 3411 1100 4

12/23/77 07/0103 AEPSC-BBC l

Doblo Canyon 1 275 3338 1088 4

11/02/84 06/07/85 PGE WE Dablo Canyon 2 323 3411 1119 4

08/26/85 03/13/06 PGE WE

'~

bd;an Point 2 247 2758 873 4

09/2803 08/01R4 UE&C WE Indian Point 3 208 3025 965 4

04/05#8 08/30R8 UE&C WE Mc Guire 1 304 3411 1180 4

07/08/8t 12/01/01 DPC WE Me Ovire a 370 3411 1180 4

06/27/83 03/01/84 DPC WE l.

6411 stone Point 3 423' 3411 1154 4

01/31/08 04/23/06 S&W GE i

Salem 1 272 3338 1090 4

12/01R8 06/30/77 PEG WE Salem 2 311 3411 1115 4

06/20/81 10/13/81 PEG WE l

Seabrook 1 443 3411 1200' 4

//

//

UEAC GE i

Sequoyah 1 327 3411 1148-4 09/17/80 07/01/01 TVA WE Sequoyah 2 328 -

3411 1148 -

4 09/15/81 08/01/82 TVA WE South Texas 1

- 498 3000 1250 4

03/22/08 08/25/04 S&R WE South Tense 2 498 3000 1250 4

12/14/88

//

8&R WE Trojan 344 3411 1130 4

11/21R$

06/20R8 -BECH OE Vogtle 1 424 3411 1122 4

03/18/87 06/01/87 SSI/BECH OE Vogtle 2 428 3411 1101 4

02/08/04 06/20/88 SSI/BECH GE Wolf Creek 1 482 3411 1170 4

06/04/85 09/03/85 BECH GE Zion 1 295 3250 1040 4

10/10/73 12/31R3 S&L WE Zion 2 304 3250 1040 4

11/14R3 09/17/74 S&L WE l

. [*y CONSENSUS FHASE DRAWINGS FOR THE OUTAGE DATA BASE STUDJ The first part of the fuel cycle study was-to estimate the characteristics that each of NRC's performance and maintenance effectiveness indicators will exhibit during each phase of a typical fuel cycle.

The estimates were based on engineering staff experience in generating the indicator data and on prior i

personal experience in commercial power plant operations.

The results of the task provided a basis for focusing the subsequent, much more detailed analysis' of indicator trends during each fuel cycle phase.

The' task was initiated by developing a hypothesized phase drawing (see Figure 1) that shows the plant power level during the entire fuel cycle.

Once the phase drawing was completed, each performance and maintenance effectiveness indicator was evaluated to identify its expected behavior during each phase of the fuel cycle.

For example, the SCRAM rate is expected to be high during the startup phase following refueling, but low during full power, steady state operation.

L This letter report documents the results of the phase drawing development and the evaluation of the phase-dependent behavior of each performance and maintenance effectiveness indicator.

Note that the results of this task are subjective but do not need to be totally accurate since they are meant to be a 1

focusing device.

Also, the exact results will be determined as a part of the next task through a series of analyses using the actual indicator data l.

histories.

I Figure 1 presents the. fuel cycle phase drawing.

The fuel cycle begins at the end of one refueling outage continues through a mid-cycle maintenance outage and terminates at the end of the next refueling outage.

To further clarify L

the cycle phase diagram, the time line has been separated into 16 phases as l

indicated on the horizontal line.

The following will define each phase.

Phase I This phase begins at the start of a recovery from a refueling outage and ends when the reactor power attains criticality.

(Duration = hours to weeks) l l

1 l

,. ?, _

,.y '

]

Phase 11 This phase begins when the reactor power reaches criticality and continues until reactor power reaches 15% of full rate:

reactor power. (Duration hours to days)

Phase !!!

This phase begins when the reactor power reaches 15% and.

continues until the plant is stable at 100% reactor power.

(Duration = hours to days)

Phase IV This phase begins when the plant is stable at 100% reactor power and continues-until the plant.is ready for a power descent for the mid cycle outage. (Duration months) l Phase V This phase begins when the plant starts a power descent for the mid-cycle outage and continues until the plant reaches 15%

reactor power level.

(Duration days to weeks.

There could L

be a wide variation due to varying levels of planned maintenance activity prior to the power descent.)

1

' Phase VI This phase begins on a power descent for a planned mid cycle l '

outage when the reactor power reaches 15% and continues until the reactor is shutdown.

(Ouracion hours to days)

Phase VII This phase begins on a power descent for a planned mid cycle outage when the reactor is operating at a subcritica' oower level and continues until the plant is operating in an outage mode.

(Duration hours to days)

Phase VIII This phase begins when the plant reaches a shutdown condition and continues until the plant-is ready for a startup at the ~

completion of the planned outage.

(Duration - days to months)

Phase IX This phase begins when the plant is ready for a plant startup following a planned mid-cycle outage and continues until the reactor power reaches criticality.

(Duration - hours to days) 3

' f l _

,p-yn u

' Phase X

'This phase'begins when the reactor reaches criticalt'y and

. continues until the plant reaches 15% reactor power.

(Dura *. :-

= hours to days).

Phase XI This phase begins when the plant reaches 15% reactor power and

[

continues until the plant is stable at 100% reactor power.

i (Duration hours to days)

Phase XII This phase begins when the plant is stable at 100% reactor-power and continues until the plant is ready for a power descent for the refueling outage.

(Duration = months)

LPhase XIll This phase begins when the plant starts a power descent for a refueling outage and continues until the plant reaches 15%

reactor power.

(Duration = hours to days.

There could be a wide variation due to varying level of planned maintenance activity ; ior to the power descent.)

Phase XIV

.This phase begins when the plant reaches 15% reactor power l

while it is descending to a shutdown condition for a refueling outage and continues until the plant reaches a subcritical reactor power level.

(Duration hours to days)

L H

Phase XV This phase begins when the plant reaches a subcritical reactor

[

power level and continues until the plant reaches a refueling j

mode condition.-

(Duration hours to days) l

' Phase XIV This phase begins-when the plant is operating in a refueling condi. tion and continues until the plant has completed refueling and is prepared to enter a power operation mode.

1,

.During the-development of the phase diagram three assumptions were used.

First, to indicate plant condition during a fuel cycle, the tracking of reactor power was selected as the most accurate and definitive method.

This assumption was made based on the knowledge that most of the performance and maintenance effectiveness indicators are sensitive to reactor power due to the 4

Quantity and types of plant associated work performed at different power levels. As an example, more maintehince it performed during an outage when the plant is operating at low power level as compared to a minimal amount of work being performed at high power levels when plant operation can be impacted. When more mai;itenance is perfonned, several indicators may incur counts.

Second, short term outages such as those resulting from scrams or power reductions coula affect some of the indicators; however, the impact would not shown on the phase diagram.

This assumption was made based on the fact that these events are unplanned and therefore, deviations from the normal mode of operation.

Since these types of events are not planned, the impact on the indicators are not easily predicted.

Given that the reactions are unpredictable and the fact that the plant would be operating in a condition other than normal, the assumption was made to not include short term outage during the fuel cycle phase development.

This assumption was made for the preliminary phase diagram development; however, as part of this study, it will be evaluated further to determine whether this assumption was correct, it will be evaluated by studying the power reductions and unplanned outage events against the indicators to determine how much impact the these events actually have.

Particular attention will be paid to the scram and engineered safety feature actuation events.

Third, for this study, the number of fuel cycles a that plant has completed would not have an impact on the behavior trends of the performance and taaintenance effectiveness indicctors.

This assumption will be tested during the subseauent' tasks, i

+*4.

t

a 4 s.

Startesp f rom Shutdoorn for Startup from Speetdowyn for Refeseling Maintenance thetage Maintenance Octay pefesellvig Outage Ered ot mef ese1ing Isorsei Operatiore maintenance t>ntspe poorse1 operation sefweling outage IOM I

s 3

5 6

a t -

N i

i I

e i

i I

i e

i i

l l

I I

I FItases II III IV V

V3 Vll V8tl II III IIII IIV IV 8

N NI I Stri t. (7t'e1.

I t9we I.

PH45( DIALAAft

....