ML19309G229
| ML19309G229 | |
| Person / Time | |
|---|---|
| Site: | Vallecitos File:GEH Hitachi icon.png |
| Issue date: | 04/29/1980 |
| From: | JACK R. BENJAMIN & ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML19309G227 | List: |
| References | |
| NUDOCS 8005050221 | |
| Download: ML19309G229 (18) | |
Text
_ _
9005 0 50 2.2.l p J BA-111-014-01 Probability Analysis For Combined Surface Rupture Offset and Vibratory Ground Motion General Electric Test Reactor l
\\
4,aw
--- m
..cu-
~
l
(
OFFICIAL :
1, e-
,c
)i -. f. M> _ V:nlNI A C. CAc^"~ 7 i, j
.t r;oTAay pueuc - cAUTCRN;A (
!'d prepared for General Electric Company l
San Jose, California Jack R. Benjamin & Associates,Inc.
g Consulting Engineers Coud House Plcro Budding. Suite 205 I
260 Shendon Ae, Poto A:fo. Cokforno 94306
?
l l
PROBA8ILITY ANALYSIS FOR COMBINED SURFACE RUPTURE OFFSET AND VIBRATORY GROUND MOTION GENERAL ELECTRIC TEST REACTOR Prepared for General Electric Company San Jose, California i
April 29, 1980 l
I w
Y x, a <ume <_.,s
. ~ ~,~.
x 4Eyr
):
Aw pucuc - cAUTOR.TA Y ~>2).o:
a
'\\c.
ntua n cou u f.
'?y comm, e pires f/H1 S, ?' " h m,-_-=~
~r l
Jack R. Benjamin & Associates,Inc.
3 Consulting Engineers Court House Plaza Building. Suite 205 I
j 260 Shendan Ave, iblo Alto. Cot:fornia 94306 L
TABLE OF CONTENTS Page 1
EXECUTIVE SU MARY jj 1.
INTRODUCTION 1_1 2.
PROBABILISTIC MODEL 2-1 3.
RESULTS AND SELECTION OF COMBINED PARAMETERS 3-1 REFERENCES R-1 i
I l
1 1
e"~
OFFICIAL l' t
.-.. eu i
- "*J N
t difhp^vcht'c I^ h t!y comm. Sch,,0jjiY
?
AWMDA, CC N,;g;f I
,,. a y
1 Jack R. Benjamin & Associales,Inc.
E Consulting Engineers 3
l
EXECUTIVE
SUMMARY
A probability analysis was conducted to determine parameter values for combined surface rupture offset and vibratory ground motion for the General Electric Test Reactor (GETR) reactor building, which is located at the Vallecitos Nuclear Center.
In two previous analyses, the annual probability of an offset occurring at any point beneath the reactor building foundation was f6und to be approximately 10-6 (Ref. I and 2).
Based on these results, General Electric Company believes that surface rupture offset of any size beneath the reactor building should be excluded as a design basis event.
If the calculated value of 10-6 annual probability were used, no offset would be combined with vibratory ground motion. Hence, for the purposes of this analysis, it was very conservatively assumed that the annual probability of an offset beneath the reactor building is 10-5, which is an upper bound value obtained from sensitivity analyses (Ref. 2).
By combining the probability for surface rupture offset with a lognormal probability distribution for effective ground acceleration (EGA) (median equal to 0.3 g and geometric standard deviation equal to
/f) an interaction curve for surface rupture offset (expressed in terms of unsupported cantilever length for the reactor building) and EGA was obtained.
The combined values have a joint annual probability of occurrence of 10-6 which is the criterion value selected in the previous analyses (Refs.1 and 2).
Based on these results, a family of cantilever lengths and an EGA value were selected and used in a structural analysis of the reactor building.
The effect of varying the probability of a surface rupture offset beneath the reactor building (i.e.10-5) was investigated and is given in this report.
For realistic geologic parameter values, the 10-5 probability used in the analysis is very conservative and represents a maximtsn value.
a.
4 OFFICIAL '
t t JJ 3 VI?7M C. CA" 9 jj Jack R. Benjamin & ASSOClotes,Inc.
E
< - '; g roount puauc - CAL lTCRr"A ;/
Consulting Engineers B
's
/
A!.AMED i COU! n Y I
d?
My tor,1m. excl:2$ f/A'i 7
!"~'
- xmz= nmmwn r
1.
INTRODUCTION A probability analysis was conducted to determine parameter values for combined surface rupture offset and vibratory ground motion.
In a previous analysis, the likelihood of a surface rupture offset occurring beneath the General Electric Test Reactor (GETR) reactor building located at the Vallecitos Nuclear Center was determined (Ref. 1).
Based on the request of the United States Nuclear Regulatory Commission (USNRC), additional probability analyses were conducted (Ref 2).
In these two analyses, the annual probability of an offset occurring at any point beneath the reactor building foundation was found to be approximately 10-6 Based on these results, General Electric Company believes that surface rupture offset of any size beneath the reactor building should be excluded as a design basis event.
However, for the purposes of this analysis a higher calculated probability is assuned.
This report starts with the results presented in Reference 2 and uses them as the basis for selecting parameter values for combined surface rupture offset and vibratory ground motion.
In this analysis, it is very conservative 1,v assumed that the annual probability of an offset occurring beneath the reactor building is 10-5, which is based on using an upper bound value obtained from sensitivity analyses (Ref.
2). The parameter values selected as a result of this analysis form one of the bases used in the current structural analysis of the reactor building.
Figure 1-1 shows the location of the existing shears, which were discovered in recent trench excavations, in relationship to the GETR (Ref. 3). Figure 1-2 shows a cross-section of the GETR site and the relationship between the reactor building, the trenches, and the two shears.
CFMCIAL I.
h
~
cc c
f Jack R. Benjamin & Associates,Inc.
I l
t/
n wm ccu'n <
j'jM (3 -( 1_1 Consulting Engineers 3
i The results of the previous probability analyses (Refs.1 and 2) indicate that it is highly unlikely that a future offset will occur beneath the reactor building (i.e. probability is approximately 10-6 per year but is very conservatively assuned to be 10-5 in this report). Due to the configuration of structural components in the reactor building, the critical surface rupture offset analysis cases occur when the offset intersects the reactor building foundation and cause the concrete core structure to cantilever.
This can happen only if an offset intersects a 20-foot wide area beneath the reactor building foundation, which is 72 feet wide.
In order to create the maximum cantilever length, the offset must intersect one specific area beneath the foundation. This case is more unlikely than an offset intersecting any point beneath the building as determined in the previous analyses (Refs. 1 and 2).
Some vibratory ground motion will occur during a surface rupture offset; although the amount is uncertain.
In the analysis discussed in the following text, the effective ground acceleration (EGA) parameter is used to represent vibratory ground motion.
The probabilistic model which combines the effects of vibratory ground motion and surf ace rupture offset is discussed in Section 2.
The results of the analysis and selection of combined parameters is presented in Section 3.
l
[D,
^bIFICI AL.
bD )y f.OTARY PUCUC - CALFCh 4
v'mlNIA C CWC 7>
' 'n y
l
')
. var *E M COU' f k
L_,
h cor.o. Mi et !!!4 *-
'_v m
Jack R. Benjamin & Associates,Inc.
I 1-2 Consulting Engineers 3
l l
l l
l s.
)
N g k
N r
B-1 B-3
[GETR 1/B-3 shear N B-2 & side trem2es O
20,00 1
feet FIGURE 1-1 LOCATION OF SHEARS IN RELATIONSHIP TO GETR 1
l t
1 l
l
--=a g -
t
(
OFFICIAL.
I 4 -'f'h. virSINtA C. CA"?
' ^ '
l; 7-
. r;OTARY PU3UC - CAUIC R A -'
- g-c,gratD % CCU* d'i
.' %.Q,,.,?.
f.?y CofW. Cifif81 I'I'E '*
..- m =
l-3 Jack R. Benjamin & Associates,Inc.
I Consulting Engineers E
NM
.I i
I 272' l
1320' Trench B-1
_Y[/
Trench B-2 R
REACTOR BULDfMG) I I
Shear D-1/8-3 72' Shear B-2 SCALE: Hortrental = Vertical 0
200 400 000 000 1000 8
I I
i g
g feet i
l FIGURE 1-2 CROSS-SECTION OF GETR SITE I
1 l
)
l 1
~ - <.
..-..w.
),'t Of FiCI,si,.
., v!"7P;;c, C.
C'^
'.L '?MY PL S LIC. C f.i i
e,
'^~
'M' m n
.y _, r.,,
1-4 Jack R. Benjamin & Associates,Inc.
E Consulting Engineers 3
r--w*
w
~-
w,-w w-mw-~w w
2.
PROBABILISTIC MODEL 1
The model for determining the probability of surface rupture offset at any point beneath the reactor building was developed in References 1 and 2.
The probability, P, of the occurrence of a future surface rupture offset at any point beneath the reactor building is defined by the following equation.
(2-1)
P=P P
BS RBlBS where:
P
= Probability that an offset will occur between shears BS B-1/B-3 and B-2 PRBlBS=Probabilitythatanoffsetwilloccurbeneaththe reactor building given that an offset cccurs between shears B-1/B-3 and B-2 The probability PRBjBS is given by the following equation which was developed in References 1 and 2.
RBlBS"L-b (2-2)
P where:
f f.,
OFFICIAL i
' J '.\\ VmnWIA C. CA'7
D 1 = width of the reactor building
~ '
j m:m rucuc - uvecm 3 4
/
AtAV!.? \\ COU!n Y
'i
l?/ Cor.1m. cxpires l'M 8,1701 h
-+-~ - _ _.
L = distance between the two shears b = width of the offset at the ground surface 2-1 Jack R. Benlomin & Associates, Inc.
E Consulting Engineers 3
i
,+
It was estimated from the geological observations of existing shears seen in the recent trench excavations that the width of a future surface rupture offset at the ground level will vary between 2 and 4 feet (Ref.1); thus, it was conservatively assumed in the analysis that the effective (i.e., in terms of affecting the structural response of the reactor building) offset width, b, is 4 feet. For values of 72 feet and 1,320 feet fer 1 and L, respectively, P is calculated to be RBlBS 0.058.
Figure 2-1 shows the relationship between an offset, soil pressure diagram, and reactor building cantilever length. The configuration shown is highly conservative since in reality the soil pressure diagram and cantilever length will vary as ground vibration occurs. The cantilever length, l, is defined as the length between the edge of c
the conservative soil pressure diagram and the outside surface of the reactor building wall.
In this analysis, 1 is measured at the time c
of zero acceleration in the reactor building. This provides a reference value that can be included with the specified vibratory ground motion in the combined structural analysis.
The equation which relates the probability of combined occurrence of cantilever length and the EGA is given as follows:
p (2 I )
p (3 a) = P
(-)
c er where:
p (>, l ) = Probability that an offset occurs and that the c
cantilever length is equal to or greater than 1 c p (> a)
= Probability that the effective ground acceleration is equal to or greater than a, given that an offset occurs j, gcJe OFFICIAL i
- ! q ik VI?GINfA C CWT
.3 a
r ? -i P
= Criteria probability value
/i.
3 NOTAny pucuc - cAuicats m.
n,, cou.ny er s
{ M7 My comm. expi es na e, r;:.
2-2 Jack R. Benlomin & Associates,Inc.
E Consulting Engineers B
l l
i Figure 2-2 shows the conceptual models for determining probabilities p(1 c) and p (1 a).
In Figure 2-2a the surface rupture offset model l
l 1s shown. The cantilever distance I is the maximum cantilever c max l
distance. This value was found in the structural analyses of the reactor building to be 20 feet (Ref. 4).
If an offset intersects the foundation slightly to the right of the l position, the reactor e max building will rotate counterclockwise, and the left edge of the building will be supported by the soil. This case is much less severe than a cantilever configuration.
The distance, i, is a cantilever length c
which is equal to or less than l From Figure 2-2a it is seen e max.
that the probabiliy p(>_ i ) is the probability that an offset will c
intersect the reactor foundation in the distance x, where x is equal to l
minus 1. Using equations 2-1 and 2-2, p(> 1 ) can be e max c
c expressed in the following form, p(1 l ) " PI*)
c hence I
\\
x+4 72+4[
P51 I ) = P
- c 1320 - 4 1320 4
or 4
p(1 l ) = P -
(2-4) c 76 l
l Figure 2-2b shows the probability density function for acceleration a, which is the effective ground acceleration value.
This function is conditional on the occurrence of an offset.
A lognormal distribution l
was selected with a median value, a, equal to 0.3g and a geometric standard deviation of d.
In terms of the standard equation for the lognormal distribution (Ref. 5):
h #Q y,lQl.S*"[,-
- [lna-inf}2 l =3,y ra m rume.cc y
\\
(y
/
,, \\ '. s. J/
/ w.c, : _
P(1 a) =
0' ygy Q Y"'
a 2-3 Jack R. Benjoniin & Associates,Inc.
E Consulting Engineers D
e# is defined to be the geometric standard deviation. Thus the probability p(> a) can be expressed as follows:
p(> a) = Q(n)
(2-5) where:
Q(-) = Complementary cumulated standardized normal distribution In(a/0.3g) n
=
in(/2)
Using equations 2-3, 2-4, and 2-5, values of a and i with equal c
probability of occurrence can be obtained.
The results of the analysis are given in the next section.
)
.; mmme nmaw 6
N
<$,,, OFFICIAL F" ALc cg-~,-, ';
y 4
t-c 4
+
h
).
l l
1 2-4 Jack R. Benjamin & Associates,Inc.
D Consulting Engineers 3
)
l i
~
i b
b i
i i
T 1
T
\\
ss ss ss s%
ss N
Soil Pressure p
s N
s Offset On
/s N
N ew Shear s
s N
i antilever Length, Ic I
FIGURE 2-1 RELATIONSHIP BETWEEN OFFSET, SOIL PRESSURE l
DIAGRAM, AND CANTILEVER LENGTH 7'" ~wn.e =. :
w.
/CN OFFICIAL r
,f, p
(: / cN 4 Vmcl.'UA C. CA'* m/
'!J f!01ARY Pucuc - cAu car i)}
s.
iN.../
m f.cr. cca,,
2-5 Jack R. Benjamin & Associates,Inc.
E Consulting Engineers B
p (> I l'= p(x) c
- x e
%A a a l
x e
le max a Surface Ruoture Offset p(a)
Jb y % S ' * % ws_w, e,.
3-a_.,
s.s,,,
Cl I ICis\\L '
i
\\/ RT,'NI \\ C. C/ "
I
- - l 1J-fj0TARY PUGUC. C/ L r
s :.4
/ iA'/ E3i CCU.
s s
/
p(
'l v: ~
ITi Comm. c;pi:es !lf f? n ' 'T l
.w a
- m..
i g,
- q.
..l a = 0.3g l
i i
b Vibratory Ground Motion i
FIGURE 2-2 PROBABILITY PARAMETERS l
2-6 l
l Jack R. Benjamin & Assoclotes,Inc.
I Consulting Engineers B
3.
RESULTS AND SELECTION OF COMBINED PARAMETERS The results of the probability analysis are shown in Figure 3-1.
The vertical axis is the cantilever length, i, and the horizontal c
axis is the effective ground acceleration, a.
The criterion annual probability, Per, (see Equation 2-3) was selected to be 10-6 The basis for this value is presented in References 1 and 2.
The curve corresponding to P equal to 1 x 10-5 assunes that the annual probability of an offset occurring at any point beneath the reactor building is 1 x 10-5 Based on the results of the probability analyses given in References 1 and 2, this value is very conservative compared to the calculated best estimate value of I x 10-6 As stated in both references, it is believed that 1 x 10-6 is sufficiently conservative and that surf ace rupture offset of any size beneath the reactor building should be excluded as a design basis event.
All cantilever length and effective ground acceleration values that lie on the P equal to 1 x 10-5 curve (and the other curves also) comply with the 10-6 criterion. Note that at the point where the cantilever length is zero, the corresponding EGA value is 0.35 g.
However there are many other locations between the two existing shears where the offset may occur (besides beneath the reactor building);
hence, the EGA value becomes the criterion value of 0.6 g for the hypothetical fault. This value was selected deterministically.
Curves for other values of P are also shown in Figure 3-1.
These curves show the variation of the interaction curves to P.
Based on the results of sensitivity analyses given in Reference 6, the probability of an offset beneath the reactor building is insensitive (in the sense of an order of magnitude) to realistic geologic parameter values. Thus, l
the probability P equal to 1 x 10-5, which is one order o' magnitude i
ou w
. ' ' - N -
OFFICIAL ! " 3.1, 7 C: CQ vinnlNIA C. CMo'7no (
g NOTARY PUBUC
- CAUTCRf0A
(
g
/
AI AVfDA COUmY uf cenm. enrires r/m e, ~n1 t j
[
n-l l
Jack R. Benjamin & Assoclotes,Inc.
E R-1 Consulting Engineers 9