ML20213A619

Summary of ACRS Subcommittee on Extreme External Phenomena 861120 Meeting in Washington,Dc

ML20213A619
Person / Time
Issue date:	12/09/1986
From:	Advisory Committee on Reactor Safeguards
To:	Advisory Committee on Reactor Safeguards
References
ACRS-2473, NUDOCS 8702030396
Download: ML20213A619 (16)

Text

(

h0$$~0 YS

. DATE ISSUED: 12/9/8

/4ggf ACRS MEETING MINUTES ON THE NOVEMBER 20, 1986 EXTREME EXTERNAL PHENOMENA MEETING WASHINGTON, DC The ACRS Subcommittee on Extreme External Phenomena met on November 20, 1986 at 1717 H Street, N.W., Washington, DC. The purpose of this meeting was to hear a status report on the work being carried out by Pacific Gas and Electric and the NRC Staff on the Diablo Canyon Long Term Seismic Program and the work that the NRC Staff is carrying on in the quantification of seismic margins program. The discussions began at 8:30 a.m. and were concluded at about 5:45 p.m., as discussions were held in open session. The principle attendees were as follows:

ACRS NRC C. P. Siess, Acting Subc. Chainnan H. Schierling J. Ebersole, Member C. Rossi D. Moeller, Member L. Reiter C. Wylie, Member G. Giese-Koch B. Page, Consultant R. Brown J. Maxwell, Consultant D. Guzy G. Thompson, Consultant N. Anderson M. Trifunac, Consultant R. Savio, Staff Pacific Gas and Electric D. Brand L. Cluff D. Hamilton W. Savage B. Tsai W. White R. Kennedy D. Bley F. Brady Highlights

1. The general requirements for conducting the work being carried out under the Diablo Canyon Long Term Seismic Program (LTSP) were established by a License Condition in the full-power license for 8702030396 881209 '

S- pon kMM 4_ . ').<W _

O EEP Meeting Minutes November 20, 1986 Diablo Canyon, Unit 1 (November 1984). The plan for the LTSP was reviewed by the NRC on July 31, 1985. An ACRS Subcommittee meeting was held on the work in March 1985 and an ACRS report issued on July 17, 1985 (See Attachment 1). Completion of the work is scheduled for mid-1988. The LTSP includes:

(a) An evaluation of the new information related to the seismic potential of the site which had not been considered in pre-1979 seismic evaluation used to establish the seismic design basis.

(b) The use of this new information to re-evaluate the potential earthquake magnitude and ground motion for the site.

(c) The assessment of the significance of any new conclusions using PRA and deterministic methods.

(d) The performance of a full scope (both internal and external events) Level 1 PRA.

Both PG&E and the NRC have assembled their own teams of expert consultants. The scope of the LTSP and the NRC review of this work are quite extensive. The NRC Staff has funded technical assistance for this review at $450K for FY 87 and has teen closely monitoring the LTSP work. The work is progressing on schedule and PG&E plans to begin to document completed elements of the work in the near future. The NRC Staff indicated that they are satisfied with the scope and quality of the work performed to date and will continue to work closely with PG&E. PG&E's schedule for completing the LTSP is shown on page 1 and 2 of Attachment B. The LTSP is currently in what has been designated as Phase III. This portion of the LTSP is described in some detail in the PG&E document "Long-Term Seismic Program - Development of Work for Phase III," dated January 1986.

l EEP Meeting Minutes November 20, 1986

2. The evaluation of the currently available knowledge of the geology, seismology, and geophysics relating to seismic potential of the Diablo Canyon site is a major element of the LTSP work. Potential seismic sources have been identified and the available information assessed. Where judged to be be necessary, additional surveys, trenching, and offshore profiling are being performed. The principle topics under evaluation are: (1) the Hosgri and San Simeon faults, (2) the Enda and San Miguelito faults and the San Luis-Pismo fold, (3) the West Huasna, Rinconada, and Nacimiento faults, (4) the Little Pine-Foxen Canyon fault and the onshore Santa Maria Basin, (5) the 1927 earthquake , (6) near-site faults, and (7) the tectonic model for the site region. The extent and configuration of the Hosgri and San Simeon faults and the associ-ated offshore fault system are major issues in this part of the LTSP. Arguments have been made in the past against linkage of the San Simeon and Hosgri faults, were based on the different configurations of the mapped portions of these two faults. Dr.

Page raised the question of a yet undercovered fault of the same configuration as the San Simeon, and possibly linked to the San Simeon, existing between the Hosgri and the plant site. PG&E's offshore mapping should help to resolve this issue.

3. The ground motion associated with potential seismic sources will be reevaluated using state-of-the-art ground motion models, recent earthquake data, and earthquake records that have been obtained at the site. Wave propagation eff+: cts will be included in the model-ing of ground motion. Soil-structure interactions have been evalu-ated using three-dimensional continuum half-space (CLASSI) and three-dimensional finite element (SASSI) models. The analyses performed for the Diablo Canyon site yield almost identical results.

0 EEP Meeting Minutes November 20, 1986 Foundation embedment was found to reduced significantly the build-ing response. Structure to structure interaction had only a small effect on the building response.

4. A fragility analysis is being performed (by R. Kennedy, Structural Mechanics Consulting) as part of the input to the seismic PRA. The methodology being used is similar to what has been used in previous commercial seismic PRA's with an expanded effort directed to identifying potential risk-important plant-specific contributors and developing an improved quantification of fragility for these contributors. The contributors that appear to dominate the risk are listed in Figure 3 in Attachment B.

5. The PRA for Diablo Canyon is being performed by Pickard, Lowe and Garrick (PLG). A preliminary PRA has been performed in Phase II and the results used to prioritize the detailed work being performed in Phase III. Transients, loss of offsite power, loss of component cooling water, small LOCA's, and loss of the 480V switchgear ventilation were identified as potential dominant risk contributors. A summary is given on Figure 4 in Attachment 8.

6. D. Guzy and N. Anderson discussed the scope of and the regulatory uses of the NRC Seismic Design Margins Program. The program was initiated in 1984 with the organization of some existing and some new research programs into a single plan coordinated by a NRR/RES working force. The program is for the most part directed at~

Eastern U.S. seismic design issues. NRC has been utilizing an expert panel to review the progress of the work and contractors to perform the actual work. The personnel employed for these func-tions are listed on page 5 of Attachment B. A methodology (de-scribed in NUREG/CR-4334, August 1985 and NUREG/CR-4482, March 1986) has been developed which uses what has been learned in the current group of seismic PRA's (7 published, 6 unpublished), actual 1

EEP Meeting Minutes November 20, 1986 earthquake experience with industrial facilities, and the work of .

the Seismic Qualification Utilities Group (USI A-46) to identify potential seismic vulnerabilities. The potential vulnerabilities are then subjected to a more detailed analysis. The procedure is outlined on page 6 of Attachment B. The procedure is currently being used on the Maine Yankee plant via a cooperative NRC-Utility effort. EPRI is also developing a methodology that is similar in many ways to what NRC is using. Negotiations are currently under-way to organize a joint NRC-EPRI application of the two method-ologies to a BWR. The intent would be to compare the methodologies and to develop a consensus NRC-EPRI procedure for performing this type of analysis.

7. The ACRS consultants were asked to submit written coments by January 20, 1987. The Subcommittee expects to report to the Full Committee and make recommendations, as appropriate, in the February-March 1987 time frame.

NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 4 1717 H Street, N.W., Washington, DC, or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Washington,

, DC 20001, (202) 347-3700.

f e

9 5

4 4

ATTACHMENT A D

O Mtuq'o UNITED STATES y n NUCLEAR REGULATORY COMMISSION

• i V .I ADVISORY COMMITTEE ON REACTOR SAFEGUARDS

$ f WASH WGTON,0. C. 20555

% .....* 3 ,

July 17,1985 1 l

Honorable Nunzio J. Palladino Chairman U. S. Nuclear Regulatory Comission Washington, D.C. 20555

Dear Dr. Palladino:

E p

SUBJECT:

ACRS COMMENTS ON THE LONG TERM SEISMIC' PROGRAM PLAN FOR THE DIABLO CANYON POWER PLANT During its 303rd meeting, July 11-13, 1985, the Advisory Connittee on Reactor Safeguards completed its review of the Long Term Seismic Program (LTSP) Plan, dated January 1985, submitted by the Pacific Gas and Electric Company (Licensee) for the Diablo Canyon Power Plant. This plan was considered by a Subcomittee at meetings on March 21, 1985, in Culver City, California, and on July 10, 1985, in Washington, D. C.

During our review, we had the benefit of discussions with represen-( tatives of the NRC Staff and the Licensee.

the documents referenced.

We also had the benefit of The LTSP Plan was prepared by the Licensee in response to a condition in

the Operating License for Unit 1 of the Diablo Canyon Power Plant. This condition requires the Licensee to develop and implement a program to reevaluate the seismic design bases for the plant.

The NRC Staff's evaluation of the LTSP Plan was provided to us in a letter from T. M. Novak to R. F. Fraley dated July 2,1985. The Staff has concluded that the program plan, as amended during discussions with the Licensee during the past five months, is responsive to the license condition for a reevaluation of the seismic design bases. We agree with this conclusion.

During our review, we discussed at some length the appropriate scope for the Probabilistic Risk Analysis (PRA) that the Licensee has proposed to use, in part, to assess the significance of the conclusions that will be drawn from the reassessment. The Licensee has proposed a level 1 PRA; this will yield core melt frequencies and plant damage states, but will not yield containment failure probabilities, source tems for radio-active material released into the containment or to the environment, and its effects on the surrounding population. The NRC Staff considers a Level 1 PRA acceptable. We agree, nw ,

t }- p .

i n; -

Honorable Nunzio J. Palladino July 17,1985

(.

Additional coments by ACRS members William Kerr, David Okrent, and -

I David Ward are presented below.

sincerely, '

[f David A. Ward

- Chaiman a

Additional Comments by ACRS Members William Kerr, David Okrent, and

David Ward We recomend that a Level 2 PRA, rather than a Level 1 PRA, be per-

!. formed, if a PRA is to be perfonned. Otherwise, neither the Licensee i nor the NRC Staff will have a satisfactory basis for estimating the likelihood of various containment failure modes and the associated relationship between radioactive release beyond containment and fre-quency of occurrence. The reasons for this deficiency with only a Level 1 PRA include the following: -

. No evaluation will be available of the likelihood of containment i

failure and mode of failure as a function of pressure and tempera-ture. Potential low pressure capacity points at penetrations will not be uncovered.

. Diablo Canyon is sufficiently different from Zion and Indian Point, or other PWRs for which a relatively complete and recent full scope PRA existt, that there is no good surrogate for detailed evaluation of the eifects of partial or full failure of various engineered safeguards including containment spray, emergency power, service:

i , water, and :;hutdown heat removal, t

! . There will be no evaluation of the potential for a severe earth-

quake to degrade containment perfomance capability, as distinct from causing direct failure.

. A severe earthquake can be the source of far more complex trans-1ents and accidents than are usually considered in a full-scope i PRA. Not only multiple failures, but a major loss of infomation i together with spurious infomation in the control room, coupled

- with a highly unusual and stressful situation, make invalid prior estimates of operator actions which could lead to a change in containment failure likelihood and mode; this would have to be reassessed.

!f\

References:

' 1. Pacific Gas and Electric Company, "Long Term Seismic Program Program Plan," dated January 1985 l

---

.--,--,-.,,---,----,w

4 Honorable Nunzio J. Palladino July 17,1985

( s .

2. Lett'er froh Thomas M. Novak, Assistant Director for Licensing.

Division of Licensing to Raymond F. Fraley, Executive Director.

ACRS, dated July 2, 1985 transmitting NRC Staff Evaluation of the PG8E Long Term Seismic Program Plan l

4 r

t 6

1 e e

6 4

ATTACHMENT B

DIABLO CANYON LONG TERM SEISMIC PROGRAM ,

SUMM ARY SCHEDULE-PH ASE Ill 1988 1987 '1988

• a e G 9 e f R R I

I

. 11/1' 8/1s _ pgggg GEOL /SEIS/GEOPHY ,. - neronT

=

El l5 siter1 10I1 El 5 ,,,. fu, ,

j "^'

CROUND MOTIONS , ,, , __

,,,o,7 E BE E E E EE BE g 55 -

E E E

E EE Js.iiss EE 1 o f i. ,

SEISMIC HAZARDS gE E EE E E E g E E E l3 E E e EE YE E*w y A 8 ers

,] y,,3t SSI REPORT E E E E 15 10/1 FRAGILITY  %,

ll E

E 5 .

PRA, -

] n'e"o")G,,.

, ll i,1 i) 1

~

2 O '

M A

R .

'G O .

R .

7 9 P E 6 e C 8 .

I M

9 1 . k f

S .

f f /

R h/

I E

S E B f f .

5 /s M M .

o f.

f 8

//

/

m/

. s E

R E V

/J f f

8

/ /

T O N / f f /

G N - . //

/

! f f

5 0

4

/

/

O S f /

L N

S E

/ f f //

/

O R f /

m//

Y N

G O

f f /

/

A C

R P ff f /

/

f /

/

O L . . / f f /

A B /

/. f / /

/

I D / s,, f ff /

/ //

. f /

/

/

/

/ ff f /

/ /

/

/ f f /

/ / /

/ Y /r / /V Y

H S S P N D '

O O R

. E l A

O T Z

/ O A .

S M I H Y E T '

S D C I

/ N I M

L I

L U ,

G A O S O I S A R E R E R G O S S F e

.- . 3 PROBABLE DOMINANT CONTRIBUTORS (FROM PHASE II)

MEDIAN VARIABILITY HCLPF COMPONENT 5 A (G) ER BU SA (G)

1. DIESEL GENERATOR 4.22 0.19 0.33 1.79 PLUS PERIPHERALS

2. 4160V SAFEGUARD RELAY 6.34 0.20 0.50 2.00 PANEL

3. 120V AC INVERTERS 4.49 0.21 0.26 2.07

4. REACTOR COOLANT PUMP 5.84 0.30 0.32 2.10

5. CONTAINMENT FAN COOLERS 5.04 0.27 0.28 2.03

6. 4160V TRANSFORMERS 5.39 0.25 0.24 2.40

7. 480V BREAKER CABINET 4.77 0.08 0.23 2.86

8. DIESEL GENERATOR 3.88 0 0.18 2.88 CONTROL CABINET

9. TURBINE BUILDING:

. LOCAL DISTRESS 4.4 0.13 0.25 2.4 OVERALL DISTRESS 6.2 0.30 0.38 2.0 e

I

FINAL PHASE 11 RESULTS - MEAN CORE .

F DAMAGE FREQUENCY (BY INITIATING EVENT) I-

'{ NONSEISMIC,INCt_OTHER EXTERNALS = 2.3 x 10 4

SEISMIC

• 9.8 x 10-5 i 2 x 10 -

TOTAL -

3.3 x 10-5 .

'== -

l PREUMINARY l ~

) , j 4

!k 1.8 x 10 -

I e l

i 6 .

i, >

g 9.4 x 104 - k, l

!t g h 1.2 x 10~" -

I' E I **.

O l d *

• j (, E 1 s to -

-? b l ,

E axir5 -

[ .

! E ~

1i 8 . l ll ex10-s _ ,

!! l 4 x 10-5 _ .

l j

2 x 104 -

l i GENERAL LOCCW SMALL LOOP 480V OTHER I m-1.0g 1.5g h k $$ M-2.0g 2.5g 3.0g >3.25g TRANSIENT (LOASW) LOCA SWITCH.

• EXTERNAL l GEAR EVENTS j /- CC (* MM %41 m NONSEISMIC TL SEISMIC #d D 8. I O .

Pickard, t. owe and Garrick,.inc.

I

\

. [

11/20/86

/- .

NRC WORKING GROUP DN SEISMIC DESIGN MARGINS J. RICHARDSON (C0-CHAIRMN) -

N. ANDERSON (CO-CHAIRMN)

D. GUZY (CURRENT PROJECT MNAGER)

R. KENNEALLY (FUTURE PROJECT MNAGER)

L. REITER C. GRIMES G. BAGCHI  ;

_ f0RMER MEMBERS N. CH0KSHI R. JAEKSON (C0-CHAIRMN)

J. CHEN J. KNIGHT (C0-CHAIRMN)

. L. BERATAN G. LEAR P. K. NIYOGI A. THADANI EXPERT PANFL ON OUANTIFICATION OF SEISMIC MARGINS R. BUDNITZ (CHAIRMN) 0 N LL FORMER MEMBFRS J. REED R. KENNEDY M. SHIN 0ZUKA k HALL JECHNICAL SUPPORT R. MURPHY (LLNL PROJECT MNAGER)

R. CUMMINGS (LLNL)

P. PRASSINOS (LLNL)

J. SAVY (LLNL)

M. RAVINDRA (E0E)

J. JOHNSON (EQE)

G 9

l

_A

- h l Start Time axis 1 Select an earthquake review level Gather information .*.*.{*.

Gather information on the plant. Determine which on systems and sort ,,...t.,,*,'

p /

Group A functions. .

. ,] . : . *

. Interac* ion /

broad classes or youps of

. . . .. . . ----- 3 components have HCLPF

.;.)),. *

/ values greater than the review

.".*.**.* A level. Possibly identify plant-unique features.

1 First plant walkdown:

,.- , . . . . Concentrate on identification of problems.

.4.*. ,

Emphasize systems interaction. Confirm spplicability of screening tools. Complete identification of plant-unique features.

o KEY: Task is performed by: f* *;;', ',*.*;l;;f,' Revision of systems

• *** relationships established 7* .*; Systems Anslyst

• • • . . t .,,,

• f, * .in*/,Step 2. Develop fault

• .,.* * : 1,* ,,.* .,- trees and event trees.

Fragility Analyst u

Both Second plant walkdown:

Primarily fragility analyst for checks.

6[

Collect specific data (sire and other l physical characteristics) of components requiring detailed analysis.

r

! I Determine minimal **, *.* *.* ,* .****

.' [* i* ..

• . Finalize HCLPF value for cut sets for end-g components b Unal 65

• • '** (components not screened out).

point core melt. '., t * * ;[, . * ; *.

)

• .. ~., . - : : . .* .

. .. . /

if .

Margin assessment '

• complete I

SEISMIC MAR'ilNS REVIEW l

l l

L2