ML20248D090
| ML20248D090 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 10/31/1994 |
| From: | EQE INTERNATIONAL |
| To: | |
| Shared Package | |
| ML20248C969 | List: |
| References | |
| NUDOCS 9806020320 | |
| Download: ML20248D090 (82) | |
Text
._
1940-98-20188 p
i
)
ArrACHMENT S SEISMIC FRAGILITIES OF CIVIL STRUCTURES AT OYSTER CREEK NUCLEAR GENERATING STATION
)
i IPEEERAI DOC 1
9906020320 990521 PDR ADOCK 05000219i F
PM i i
IN&G 990-23/4
~
a.%;,- m.
INTERNATIONAL i
.I O.'.Bf",tt FRACLTf ES OF CIVIL
- H't.iCTURES AT OYSTER CREEK
.,.qt.< !.Wi.'i.!ERATING STATION 4
Dei.obe* '. 304 l
1 i.*;>ared for:
: k.
- .- r c_AR CCt*'.*0FWI Sid C 'a Uppe; lind Road T' art.;p!. '.,r, New Jersey 07054 t'u F r.', ject Number
- 50124.04 EQE INTER #/t (IONAL l
t
l TABLE OF CONTEh*TS EAQsl 1.
I N TR O D U CTI O N.............................................................................. 11 2.
GENERAL CRITERIA FOR DEVELOPMENT OF MEDIAN SEISMIC.
S A F ETY F A C T O R S.......................................................................... 2-1 l
l 2.1-De finit i on o f Fa il u re............................................................... 22 2.2 Basis for Safety Factors Derived in Study................................. 22 2.3 Formulation Used f or Fragility Curves....................................... 23 2.4 D e s ig n a nd Construction Errors................................................ 2-6 2.5 Correlation Between Failure Modes.......................................... 2-6 3.
PARAMETERS USED IN THE EVALUATION OF THE SEISMIC FRAGILITY......................................................................................
3-1 3.1 Strength...............................................................................
3-1 3.2 Ductility................................................................................
32 3.3 S ys t e m R e s p on s e.................................................................. 3-3 3.3.1 Ea rth q ua ke Cha racte ristic s............................................ 3-3
- 3. 3. 2 S yst e m Da m pi n g......................................................... 34 3.3.3 S oi! Structure Int eraction.............................................. 3-5
- 3. 3.4 Load C om bina ti ons...................................................... 35
- 3. 3. 5 M odal C om bina ti ons..................................................... 3-6 4
1 3.3.6 Combination of Responses for Earthquake Directional C om pone n t s................................................................ 3-6 3.4 Earthquake induced Soil Liquefaction and Permanent Ground De f o r m a t i on s......................................................................... 3-7 l
3.5 Consistency Between Hazard and Fragility Descriptions............. 3-8
- Hoi t 4snelayettoe.wtw i
M_.R
@W '" 9 I
t i
l TABLE OF CONTENTS (Continued)
EAEA 4.
STRUCTURES.................................................................................
4-1 l
4.1 Median Safety Factors and Logarithmic Standard Deviations...... 4-1 l
l l
4.1.1 S t ructure Ca pa city................................................... 4-4 l
4.1.2 Structure inelastic Energy Absorption........................ 4-14 4.1.3 S truct ure Sliding..................................................... 4-2 2 4.1.4 Structure Response Factors..................................... 4-25 4.2 S tr uc ture Fra gilitie s............................................................... 4-2 9 1
4.2.1 R e a c t o r Buil din g...................................................... 4-3 0 4.2.2 Tur bin e Build ' ng...................................................... 4-31 4.2.3 I n t a k e S t ru c t ur e...................................................... 4-3 2 4.2.4 Emergency Diesel Generator Building........................ 4-33 4.2.5 Bus Duct Bank From the Diesel Generator Building to the Tur bine Building................................................ 4 3 5 4.2.6 Fire Pond Pump House, Fire Pond Dam, Fire Pond Pi pi ng.................................................................... 4 - 3 6 4.2.7 Circulating Water intake and Discharge Tunnels and Outfall Discharge Structure...................................... 4-37 i
4.2.8 Combustion Turbines, Fuel Tank, Gas Supply Piping... 4-38 4.2.9 Condensate Transfer Building................................... 4-40 i
4.2.10 Ve ntila ti on S t a c k.................................................... 4-41 5.'
REFERENCES..................................................................................5-1
' 2Ho1146nbloyettoc.wfw ij gg
@d
TABLE OF CONTENTS (Continued)
TABLES Eaga 41 REACTO.9 BUILDING SEISMIC FRAGILITY PARAMETERS.................... 4 42 42 REACTOR BUILDING SEISWC FRAGILITY PARAMETERS-
.. 443 3
TURBINE BUILDING SEISMIC FRAGILITY PARAMETERS.....................
4-44 4
44 TURBINE BUILDING SEISMIC FRAGILITY PARAMETERS..................... 4-45 45 TURBlNE PEDESTAL SEISMIC FRAGILITY PARAMETERS.................... 4-46 46 INTAKE STRUCTURE SEISMIC FRAGILITY PARAMETERS................... 4-47 4-7 EMERGENCY DIESEL GENERATOR BUILDING SEISMIC FRAGILITY P A F, A M ET E R S................................................................................ 4 - 4 8 4-8 FIRE POND PUMP HOUSE SEISMIC FRAGILITY PARAMETERS............ 4-49 49 SEISMIC FRAGILITY PARAMETERS FOR OTHER STRUCTURES........... 4 50 FIGURES 2-1 Fra gility Curve Re pre senta tions......................................................... 2-8 31 Uniform Hazard Spectra for the 10-4 Annual Probability of l
Exceedance.....................................................................................
39 1
!~
41 Effects of Time and Curing Conditions on Concrete Strength............... 4-51 i
l 4-2 Str ength of Concrete She ar Wall...................................................... 4-5 2 43 Schematic Force - Deflection Relationship for Reinforced Concrete Bilinear Relationship and Equivalent Elastic-Perfectly Plastic l
R e l a t i on s hi p................................................................................... 4 5 3 l
2HD1146nbloyettoc.wfw iii
- 1. INTRODUCTION As part of the IPEEE program for the Oyster Creek Nuclear Generating Station (OCNGS), a seismic probabilistic risk assessment (SPRA)is being conducted by GPUN. The SPRA combines the probabilistic definition of seismic capacity, the probabilistic definition of seismic hazard, and a characterization of the operating systems of the plant to determine the probability distribution for the frequency of seismic induced plant damage states and core melt. In this evaluation, systems models, event trees, and fault trees are utilized to characterize the plant systems.
This report discusses the methodology and presents the results for the seismic fragibty evaluation of selected civil structures to be included in the SPRA risk models. The seismic fragilities represent the probabilistic definitions of seismic capacity for the civil structures and are reported in terms of the peak free field acceleration of the reference earthquake. The civil structures and components included in the evaluation documented here are:
Reactor Building, Turbine building, including the Turbine Pedestal, and the outdoor 1
HVAC platform on the nbrth side, intake Structure, e
Emergency Diesel Generator Bu:Iding, l
Duct bank from the Emergency Diesel Generator Building to the I
e Turbine Building, Fire Pump House, the Fire Pond Dam, and the Fire Pond Piping from the Fire Pump House to the plant fire protection lines, Circulating Water intake and Discharge Tunnels, including the Discharge Outfall Structure, l
Combustion Turbines, including their fuel oil tank and gas supply l
- piping, 50 2HD1146nbloyotntro.wfw 11 L
D
' Condensate Transfer Building, e
Ventilation Stack e
in the above, the seismic' fragility for a single system composed of several components is based on the most vulnerable component or weak link whose capacity controls the functionality of the system. For some structures, the seismic capacity is judged to be high and beyond the range of interest based on a review of the postulated seismic loading and the credible failure modes.
A walkdown was conducted on January 11,1994, to review the above structures and components. The walkdown team consisted of one engineer each from EOE and Geomatrix, as well as two representatives from GPU None of the buildings were entered as the walkdown focused on reviewing the generallayout of the structures and identifying important features of the structures and the adjacent areas significant to the postulated earthquake ground motion.
Potential soil-related seismic failure modes (liquefaction, seismic induced settlements, and slope deformation) were examined as part of the current evaluation for the IPEEE program. The results of the OCNGS liquefaction study are documented in Reference 4. Soil related issues were found to ta significant for several of the above structures and components. As a result, fragilities were evaluated for failure modes associated with soilliquefaction induced deformations, I
where applicable, in addition to those associated with direct seismic induced (inertial) effects.
The Oyster Creek site is classified as a soil site in References 1 and 2. The seismic hazard is described in terms of annual probability of exceedance curves for the peak ground acceleration in addition to the hazard curves, free field response spectra are provided at the 15th,50th, and 85th percentiles for return periods ranging from 1000 years to 100,000 years. The fragilities described in this report are anchored to the mean peak ground acceleration. This acceleration is the average,of the peak accelerations from two horizontal components. For this evaluation, the spectral shape' associated with the 50th percentile,- 10,000 year return period response.
spectrum in Reference 1. is taken to be median centered for all fragilities.
For most of the structures considered here, the fragilities are based upon the latest probabilistic seismic response analyses (Reference 3). In Reference 3, soil-structure bb
- se t 14sndio.. intro.-v.
12 c:: m L..
m__
interaction analyses are performed for the Reactor Building, the Turbine Building and Turbine Pedestal, the intake Structure, and the Emergency Diesel Generator Building, which included multiple time history analyses at peak free field grounc' acceleration * "els of 1SSE (0.184g), 2SSE, (0.368g), and 3SSE (0.552g). Soil stiffness ar. damping properties consistent with these ground motions are used in the analyses, in the multiple time history analyses, the soil stiffness and damping properties as well as the structure stiffness and damping properties are varied. The selection of the values of these properties was based on Latin hypercube sampling techniques. Median structure responses and the total variability associated with the earthauake ground motion, the soil stiffness and damping, and the structure frequency and damping are evaluated. Possible nonlinear structure effects such as concrete cracking, buckling of compression members, etc., at higher acceleration levels are not included and hence, all of the response analyses represent elastic structure response. The approach adopted in this fragility evaluation is to determine the median factor of safety and its statistical variability which exists for the 3SSE earthquake in order to estimate the expected response at failure. Although inelastic energy dissipation is included in determining the factor of safety, no nonlinear analyses of the structures is conducted, and all evaluations are based on elastic analysis and load distributions.
The fragility results can be used together with the estimated annual frequency of l
occurrence of various ground motion levels to determine the frequency of seismic-induced failure for each safety-related structure in the plant. In the total pro':abilistic risk assessment, these conditional structure failure frequencies,
)
together with similar equipment fragilities, are used with systems models to determine the probability of core melt frequencies and radioactive release frequencies. These results are then combined with the results of the consequence analysis to determine the risk induced by earthquakes.
l l'
fEOE 2HD1146nblayetnern wfw j-3
[gy"q l
r i
f l
- 2. GENERAL CRITERIA FOR DEVELOPMENT OF MEDIAN SEISMIC SAFETY FACTORS The factor of safety of a structure o' component is defined herein as the resistance capacity divided by the response associated with the reference earthquake of 3SSE or 0.552g effective peak acceleration used.n the latest (Ref,3) analysis of the structures. The development of seisrnic safety factors associated with the reference earthquake is based on consideration of several variables. The variability of dynamic response to the specified acceleration and the strength capacity of the structure or equipment component are the two basic considerations in determining the variability in the factor of safety. Several variables are involved in determining both the structural response and the structural capacity, and each such variable, in I
turn, has a median factor of safety and variability associated with it. The overall factor of safety is the product of the factors of safety for sach variable. The median of the overall factor of safety is the product of the median safety factors'of all the variables. The variabilities of the individual variables also combine to determine that of the overall safety factor.
Variables influencing the factor of safety on structural capacity to withstand seismic-induced vibration include the strength of the structure compared to the reference earthquake stress level and the inelastic energy absorption capacity (ductility) of a structure or its ability to carry load beyond yield. The variability in computed structural response for a given effective peak free field ground acceleration is made up of many factors. The more significant factors include variability in (1) ground motion and the associated ground response spectra for a given peak free field ground acceleration, (2) energy dissipation (damping), (3) structural modeling, (4) method of analysis, (5) combination of modes, (6) combination of earthquake direction components, and (7) soil structure interaction.
For structures which may be susceptible to sliding, the variability in the amount of sliding is also significant. Similarly, for structures vulnerable to earthquake-induced soil settlements, the variability in the amount of settlement is significant. The ratio between the median value of each of these factors and the value used in the reference earthquake analyses of the structures and the variability of each factor are quantitatively estimated and provided in Chapter 4 for various structures. For some parameters associated with structure response, the variabilities of each
)
2Ho t i4snetov.treh2.wfw -
2-1 ggg LC.G25 9
. =..
...._....._......._.........._................__.._.6
l 1
individual parameter cannot be individually determined since they are combined in the overall struc,ture response variability in Reference 3. These estimates are based on available test data, the Oyster Creek structures, elastic analysis, and engineering judgment and experience in the analysis of nuclear power plants and components.
2.1 DEFINITION OF FAILURE To estimate the median factor of safety against the structure or component failure for the reference earthquake effective peak acceleration, it is necessary to define fai!ure.
For the purposes of this study, structures are considered to fail functionally when l
inelastic deformations of the structure under seismic load are estimated to be sufficient to potentially interfere with the operability of safety-related equipment attached to the structure. These limits on inelastic energy absorption capability l
(ductility limits) chosen for structures are estimated to correspond to the onset of significant structural damage. For many potential modes of failure, this is believed to represent a conservative bound on the level of inelastic structural deformation which might interfere with the operability of components hcused within the structure. It is important to note that considerably greater margins of safety against structural collapse exist for many of these structures than reported within this study. Thus, the conditional probabilities of failure for a given free field acceleration reported herein for structures are considered appropriate for equipment operability limits and should not necessarily be inferred as corresponding to structure collapse. Structures which are susceptible to sliding or soil settlements are considered to have failed when sufficient sliding displacement or soil deformation is incurred to fail piping or electrical duct banks or to cause sufficient damage resulting from structure-to-structure impact to interfere with equipment functionality.
2.2 8 ASIS FOR SAFETY FACTORS DERIVED IN STUDY Since existing codes and standards do not require the determination of ultimate seismic capacities, detailed information related to these capacities is not available for specific structures. Therefore, estimated, median safety factors, variabilities, and conditional frequencies of failure are based on existing analyses, qualified engineering judgment, and experience gained as a result of previous PRAs. The 2HD 1146tibloystreh2.wfw 22 hb te m
fragilities are based on linear elastic models. As noted previously, no nonlinear time-history analyses of the structures have been conducted.
2.3 FORMULATION USED FOR FRAGILITY CURVES Seismic failures of nuclear power plant structures have not occurred. Thus, fragility curves must be developed primarily from analysis combined with engineering judgment which is supported by limited test data.
Fragility of a structure or component is defined as the conditional frequency of its failure for a given value of the ground motion parameter (for example, peak free-field ground acceleration). Thus, the fragility evaluation is based on the estimation of the median ground acceleration value for which the seismic response of a given structure or component exceeds its capacity, resulting in failure. Because there are many sources of variability in the estimation of the median ground acceleration capacity, the component fragility is described by means of a family of fragility curves. Figure 21 depicts such curves, showing the median (50 percent confidence, C2 = 0) curve with its shape governed by randomness variability, (Sg),
and showing the relative position of the curve for other confidence levels greater than or less than 50 percent. The properties of the fragility curves and the general approach to their development are defined in previous works (References 5 and 6).
Employing the characteristics of the lognormal distribution as described in these references, the entire family of fragility curves for any mode of failure is defined in terms of a median estimate of the structure capacity expressed as a free-field ground acceleration, A (Figure 2-1), times the product of randomness and uncertainty variables, en and s, which have unit median values and are lognormally u
distributed with logarithmic standard deviations of Sg and By, respectively.
A nu (2-1)
=
c s
f The capacity in terms of the free field ground acceleration (3SSE or 0.552g for the reference earthquake) is A
FA3SSE (2 2)
=
where F equals the overall fector of safety based on response to the reference earthquake, and A3SSE equals the peak free field ground acceleration of the reference earthquake (0.552g). The overall factor of safety has a median value, 2ND 1146ttleystech2.wfw 2-3
I i
i b, and randomness and uncertainty variabilities (Gg and G ). In contrast, the U
average reference peak free field acceleration is a deterministic quantity (0.552g),
although variabilities were calculated in Reference 3 associated with this level of earthquake. Thus, the product of these terms (Equation 2-2) results in a capacity expressed in terms of a free field acceleration which has a median value,4, and randomness and uncertainty variabilities which are equal to the corresponding variabilities associated with the overall factor of safety. As a result, the structure capacity free field acceleration at any point within the family of fragility curves is computed as:
d (C GR+C8I (2 3) 1 2U A
=
g l
. where C1 and C2 are the statistical constants associated with the failure fraction and confidence level of interests (Figure 21).
In this study, the guidelines used to estimate the values of Gg and 8 f
0 or each variable affecting A are based on considering the inherent randomness, Gs, to be associated primarily with the earthquake characteristics themselves, and the uncertainty, Gu, to be associated with lack of knowledge of the model and material parameters. Thus, the variability resulting from earthquake response spectra shape and amplification, earthquake duration, numbers and phasing of peak excitation cycles, together with their contributions to structure ductility and response characteristics is attributed to randomness. In general, it is not considered possible to significantly reduce randomness by additional analysis or test based on current state-of-the art techniques. Uncertainty, on the other hand, is considered to result primarily from analytical model;ng assumptions and other knowledge gaps concerning variables such as material strength and damping that could in many cases be reduced by additional study or test.
The lognormal distribution can be justified as a reasonable distribution since the statistical variation of many material properties (References 7 and 8) and seismic response variables (Reference 9) may be reasonably represented by this distribution.
In addition, the central limit theorem states that a distribution consisting'of products and quotients of distributions of several variables tends to be lognormal even if the individual distributions are not lognormal.
2MD1146P !hk$t$h eNfg 7 e g's
It should be noted that the use of the lognormal distribution for estimating failure fractions on the order of five percent or greater is considered to be quite reasonable. However, lower fraction estimates which are associated with the extreme tails of the distributions must be considered less accurate. Use of the lognormal distribution in these regions is conservative since the low frequency tails of the lognormal distribution generally extend farther from the median than actual structural resistance or response data might indicate. The degree of conservatism introduced into the probability of release is dependent not only on the conservatism in the fragility description, but also on the seismic hazard description at low seismic levels. If the seismic hazard for low seismic input levels is large enough,it is apparent that very low level earthqu'akes can govern the seismic-induced release.
This is considered unrealistic for engineered structures and equipment found in nuclear power plants since such structures and equipment are subjected to various low level dynamic loads on a repetitive basis from a number of sources such as wind and low level earthquakes which have never been known to produce nuclear power plant failures.
A number of variables, such as material strength, do not typically fall far below the median value but instead normally exhibit some distinct lower bound (References 7 and 8). Damping data also exhibit lower and upper bounds so that cut-offs on the tails of the fragilities could be expected to exist. Based upon a review of the results of nonlinear time history analyses (Reference 10),it is observed that the response due to seismic excitation begins to deviate from the lognormal distribution in the lower tails of the curve. Thus, the following judgmental recommendations are made with regard to the handling of the variability parameters in the tails of the lognormal distribution.
Uncertainty variability (B ) should not be truncated.
1.
U 2.
For ductile failure modes which include most civil structure failures, the randomness variability (B ) can reasonably be R
truncated at about 1 % failure probability or -2.3Ga.
3.
For comparatively brittle failure modes, the randomness variability (Gg) can reasonably be truncated at about 0.1 %
failure probability or -3.1Ga.
2HD 1146nbloystrch2.wfw 25 ggg
$M*M
2.4 DESIGN AND CONSTRUCTION ERRORS Consideration is given in this study to potential design and construction errors at Oyster Creek. It is concluded that if undetected errors exist, they have to be very substantial to impact the plant seismic fragility. The conclusion is that small errors or errors involving local areas of construction at Oyster Creek could not substantially reduce the seismic capacity. Extensive errors in design or construction could, at least theoretically, impact the capacity particularly under the occurrence of rare events, but the extensive quality assurance and quality control programs used at Oyster Creek are highly likely to have discovered any large design and l
construction errors. Based on this logic, seismic design and construction errors are not considered further in the fragilities evaluation.
l 2.5 CORRELATION BETWEEN FAILURE MODES Some potential failure modes of the Oyster Creek structures may not be completely independent. For instance, if the capacity of the lateral force resisting system (i.e.,
the shear walls) is actually higher or lower than the value used in the analysis, the acceleration capacities of all failure modes (including different structures) govemed by the shear walls could be proportionately higher or lower. The actual capacity of the force resisting system may be different from that used in the evaluation due to differences in strength or modeling assumptions. These effects are included in the variabilities associated with each mode of failure for a given structure or component. However, different degrees of correlation may exist from failure mode to failure mode. For instance, for a given structure with given concrete and reinforcing steel strengths, the variability on strength from failure mode to ftilure mode may be strongly correlated, while different modeling assurnptions may result in little correlation for different failure modes.
For failure modes with little contribution to risk, consideration of correlation between modes is probably unimportant. However, consideration should be given to possible correlation between controlling seismically-~ induced failure modes. This
- effect can be included in the development of the seismic risk model.
- There is a certain degree of interdependency between structural and sliding modes l
of failure that could be considered. Fragilities are estimated for failure modes
_ associated with structure sliding or failure of the structure itself (i.e., shear wall gg w,,......_
2.e G
failure). These fragilities are developed assuming that the sliding and structural failure modes are completely independent. That is, structural failure acceleration capacities are based upon seismic loads with the structure bonded to the supporting j
soil, even if sliding is expected at lesser acceleration levels. In reality, the l
occurrence of sliding willlimit the structure inertialloads since accelerations in excess of those corresponding to sliding cannot be transmitted through the structure / soil interface. If sliding is expected to occur at lesser acceleration levels than those expected to cause structural failure, the fragilities associated with sliding are regarded as the lowest fragility associated with that structure. Treatment of the structure fragilities which incorporate the probability that sliding does occur would likely result in higher structure capacities, and therefore are conservatively not included in this evaluation.
A similar failure mode dependence can be considered for structural and soil deformation related failure modes. For those structures for which both types of failure modes are credible, fragilities are presented for both the structure and soil related failure. Structure failure modes are evaluated assuming that soilliquefaction or settlements do not occur if the soil related failure mode is expected at a lower acceleration level than the structural failure mode, then the soil related failure is considered to control. In most cases, just one type of failure, either structure inertia, or soil deformation, controls the fragility. However, the occurrence of liquefaction is a probabilistic event and,in the case of the Diesel Generator building, different fragilities are estimated given that liquefaction does or does not occur.
The dependence of the failure modes and the capacities conditional on the occurrence of liquefaction are discussed case-by, case in Section 4.
2Ho 1146nbloyettch2.wfw 27 twe
1.0 -
'5t Confidence 0.8 Median Frarliitty Curve m
e o
e f
s' a
r 54 Confidence 30.6
.n 04 Uxe C1=0 0.5 E3 o,4 2L i
Median Ground C.2 Acceleration i
C s = -1.65 0.05 f.
/
/
t a
llCLPF A
PEAK CROUND ACCELERATION (G)
I
\\
l Figure 2-1: Fragility Curve Representations
- h..
1 II, 2HD 1146rt/oystech2.wfw 28 p.
- 3. PARAMETERS USED IN THE EVALUATION OF THE SEISMIC FRAGILITY The seismic design of the Oyster Creek structures was based on accepted methodology and criteria in conformance with USNRC licensing requirements at the time of the plant design. These criteria and methods together with the design codes (c.f. Reference 11)in use at the time of the design formed a conservative design basis and ensured that substantial factors of safety were introduced at various stages in the design procedure.
The exact magnitude of many of the safety factors is still a matter of considerable discussion. Nevertheless, in order to establish a realistic value of the actual seismic capacity of a structure or equipment component, the amount of conservatism along with its variability must be established as accurately as possible, in this chapter, the most important parameters affecting seismic capacity are identified, and the general methods used in obtaining more realistic values associated with very high seismic response levels are discussed. The methods of determining these parameters are described in more detailin Chapter 4. The estimated seismic capacities of the most probable failure modes are also developed in Chapter 4.
The general approach in the evaluation of the Oyster Creek seismic capacities is based on the development of the overall factor of safety associated with each important potential f ailure mode. Based on the median reference earthquake response parameters, a median seismic capacity was obtained in terms of the peak free field acceleration of the reference earthquake. The overall factor of safety is composed of severalimportant contributions such as strength, allowance for inelastic energy absorption (ductility), and differences between the reference earthquake and responses resulting from higher accelerations.
3.1 STRENGTH The design strength of a structure or an equipment component is typically determined from applicable codes and standards such as the ACI building codes for concrete structures. Inherent in these design codes is a factor of safety on material strength. Sometimes this factor is known reasonably accurately, while at other times, it is less well defined or may be a function of the geometry or other physical characteristics of the component such as for reinforced concrete shear walls.
2Ho 1146nbloyotrch3.wfw 31
- - --A
For reinforced concrete shear walls, for example, the strength of the element is a function of the concrete strength, the amount and strength of the reinforcing steel, and the configuration of the element including the element geometry and reinforcing steel details. In establishing the strength and seismic capacity of concrete components, the results of concrete compression tests and reinforcing steel stre.7gth and elongation tests provide a valuable basis for establishing the element strength capacity. However, the increase in concrete strength with age together with the specific geometric details of the element must also be considered. These of fects are discussed in more detail in Chapter 4.
3.2 DUCTILITY -
To establish realistic seismic capacity levels for most structures and components, an assessment of the inelastic energy absorption must usually be considered.
Exceptions to this are some modes of failure involving sliding or elastic buckling.
However, most failures due to seismic response involve at least some degree of yielding, and hence energy absorption.
Consideration of structure ductility typically results in an increase in the predicted seismic capacity over that which would be predicted using linear elastic techniques.
All analyses of the Oyster Creek structures described in Reference 3 are linear elastic.
Although inelastic analyses would be desirable in order to more accurately quantify the inelastic effects for the remaining structures, the dissipation of inelastic energy may be adequately accounted for by the use of the ductility-modified response spectrum approach (References 12,13 and 14) together with a knowledge of the elastic model results and the expected ductility ratios associated with the critical i
elements of the structure or component. This approach is based on a series of nonlinear time history analyses using single degree of-freedom models with various nonlinear resistance functions and levels of damping. For different levels of ductility, the reduction in seismic response for the nonlinear system compared to the equivalent elastic system response is calculated. This reduction has been shown to be a function of the frequency and damping of the system as well as the ductility (References 13 and 14). However, a reasonably accurate assessment of
~
the response reduction of a structure or component can be made provided the results of the elastic analysis are available and a realistic evaluation of the system l-2Ho 1146nbleyearch3.wfw 3-2
ductility can be made for multi degree-of-freedom structures (References 10 and 14). For the Oyster Creek structures controlled by failure modes that include inelastic energy absorption, the elastic analysis results are readily available and estimates of system ductility can be made.
For the major structures included in the Reference 3 analyses, soil structure interaction effects are significant. For the postulated earthquake motions, substantial amounts of energy are dissipated by the soil damping and radiation effects. At structure response levels beyond yield, energy dissipation is provided by both the structure hysteresis and the soil. Also, due to the soil flexibility, the structures can experience significant displacements relative to the free field. These effects must be incorporated in the estimation of the system ductility. While a given structural element may have the capacity for substantial ductility, the inclusion of the soil effects reduces the overall system ductility. Details on the methods used to estimate the system ductility are discussed in Section 4.
3.3 SYSTEM RESPONSE A number of parameters must be evaluated when considering the expected system response near failure compared to the reference earthquake analysis conditions (Reference 3). Among these are the earthquake characteristics, directional combinations, system damping, load combinations, and system modeling approaches and assumptions. Although the values used in the analysis of the Oyster Creek structures for some of these parameters were selected to be j
essentially median-centered and therefore introduce little change in the expected seismic capacity, other criteria are more conservative. Several of the more j
important parameters required in evaluating the system seismic response at high response levels are discussed below. The factors of safety associated with these parameters are developed in the following chapters for the specific failure modes identified for the various Oyster Creek structures.
3.3.1 Earthauake Characteristics I
The reference ground spectra used for the structural seismic fragilities are the 10,000 year return period spectral shape from Reference 1. As previously noted,
)
these spectra are appropriate at the free surface of the soil. The 3SSE reference j
earthquake analysis (Reference 3)is based on time history analyses with free field 2Ho 1146nbloyotrch3.wfw 33
)
motion scaled to 0.5529. The reference response spectra are shown in Figure 3-1.
These are smoothed spectra whereas actual earthquake response spectra have peaks and valleys. This introduces some randomness in the structure response since the exact locations and magnitude of these peaks and valleys is unknown. In Reference 3, the calculated variability in the structure response includer the variability in the earthquake excitation. However, only the total variability in response is reported in the results of Reference 3. Thus, the contributions from other sources such as modal phasing, soil stiffness and damping and structure stiffness and damping cannot be separately determined. All of these parameters are varied in the probabilistic response analyses. Therefore, for the structures analyzed in Reference 3, a single variability represents the total variability introduced by these parameters. This variability is dominated by the randomness variability associated with the shape of 'he input response spectrum. Previous studies (Reference 10) have shown that a slightly conservative result is obtained when the variability associated with spectral shape, stiffness, and damping are separately determined. Therefore, there is no need to separate the variabilities. For the structures not imivoso in Reference 3, the variabilities associated with spectral shape, stiffness, and damping are sepreately estimated.
3.3.2 Svstem Damoina i
J Recent studies (Reference 16) indicate a value in the range of 7 percent of critical may be expected in concrete structures at response levels corresponding approximately to code allowables. A median value of 7 percent of criticalis used in the Reference 3 analyses.
At levels of seismic response corresponding to failure as defined in a fragility evaluation, an increase in the above damping values is expected. This increase is observed in using the spectral averaging method outlined in Reference 14, in which an average effective damping is calculated based on the sum of the elastic and hysteretic damping associated with a typical nonlinear response cycle. In the current fragility evaluation, a median clastic valce of 7 percent of criticalis used for structure damping, with 5 percent as a minus 1G standard deviation value.
)
.p hk 2Ho 1146nbloystrch3.wf w 3-t' I
i
E l
}
3.3.3 Soil Structure Interaction Soil stneture interaction analyses were performed for the structures included in Reference 3. These analyses include the effects of soil compliance, wave scattering, and embedment, with soil properties consistent with the intensity of the postulated ground motion. These analyses utilize state-of-the-art SSI techniques and are assumed to be essentially median centered. Statisticalincoherence of the ground motion wave is not included in the soil structure interaction enalyses. This effect is not expected to lead to significant increases in the structure fragilities and is neglected in this investigation, with some conservatism as e result.
1 For the structures not included in Reference 3, only the evaluation of the diesel fuel oil tank for the combustion turbines includes soil structure interaction. The simplified analysis methods described in References 18 and 19 are used to incorporate soil structure interaction into the response analysis of the tank. The other structures are controlled either by structure sliding or earthquake induced
]
permanent soil deformations. Soil structure interaction effects are conservatively not included in the assessment of the Ventilation Stack. Additional detail on the analysis methods for the structures not included in Refaunce 3 is provided in Section 4.
3.3.4 Lagd Combinations For structures and much of the equipment contained within the drywell, the design basis load combinations and those specified by current licensing criteria include a combination of a loss-of coolant-accident (LOCA) and the SSE loads. Random LOCA events have an extremely low frequency of occurrence as do seismic events.
l 1
The frequency of both events occurring simultaneously is so small that their inclusion is judged to be not important to the risk analysis results. Therefore, for the Oyster Creek fragility evaluation, LOCA loads are not combined with earthquake
- loads, i
i 2Ho 1146nbloystrch3.wf w 3-5
3.3.5 Modal Combinations In the reference earthquake analysis reported in Reference 3, structure loads are developed from three dWensional time history analyses. The structure modal combinations are implicitly incorporated and are median centered.
3.3.6 Combination of Resoonses For Earthouake Directional Comoonents l
l The analyses conducted in Reference 3 uses three simultaneous earthquake records as excitation. For these structures, the combination of directional inputs used in Reference 3 is taken as median centered.
Current NRC design procedures specified in Regulatory Guide 1.92 (Reference 17) require that the response from the two horizontal directions be combined with the vertical response by the SRSS. Other mett.ods of combining directional components such as the one delineated in Newmark and Hall (Reference 15) also yield realistic results. This approach recommends adding 100% of the response from one directional component to 40% of the response from the remaining two components. This method is easy to use and retains a consistent relationship between loads and stresses. The SRSS method is considered to be essentially median-centered. Although the 100 %,40 %,40 % method is slightly conservative, both methods are taken as median-centered for the Oyster Creek fragility evaluations for those structures not analyzed in Reference 3.
In general, a ratio of the vertical to horizontal peak ground accelerations of 2/3 is taken to be median centered. A nominal verticalload increases the shear and moment capacities of concrete structures and this is included in the fragility evaluations using 40% of the vertical earthquake to partially overcome dead weight i
effects. However, for failure modes such as sliding, the effect of reduced dead weight forces due to vertical earthquake upward forces may reduce the friction capacity substantially at high horizontal accelerations. Recorded earthquake records have shown that while the veNical to horizontal peak ground acceleration ratio of 2/3 is a reesonable median value, instances have been recorded with higher ratios, and in some instances, the vertical component may exceed the horizontal.
{
For failure modes such as sliding where the vertical earthquake component is important, an upper bound ratio of 1.5 is assumed for the ratio of the vertical to I
l horizontal directional components.
l 2ND 1146ab4ystech3.wfw 3S MM L%W%
r-1 l
l 3.4 EARTHQUAKE INDUCED SOIL LlOUEFACTION AND PERMANENT GROUND DEFORMATIONS The potential for earthquake induced soil liquefaction and permanent ground deformation for the Oyster Creek structures within the scope of this study is described in Reference 4. For those locations where the potential for liquefaction exists, the median ground accelerations to induce liquefaction are estimated along with uncertainty bounds. Permanent ground settlements and deformations are also I
provided with uncertainty bounds.
1 Significant liquefaction or settlement at the Reactor building is very unlikely for any level of ground shaking. For the Turbine building, the free field peak ground acceleration required to cause liquefaction of the compacted fill beneath the foundation exceeds 0.70g, and therefore liqueft.ction is not expected to affect the fragility of the Turbine building. Significant liquefaction, settlements, or lateral deformations are also found to be unlikely for the intake structure, the Discharge Outfall structure, the Circulating Wa'ter intake and Discharge tunnels, the Fire Pond pump house, the Fire Pond dam, the Combustion Turbines and their fuel oil tank, the Condensate Transfer building, and the Ventilation Stack.
Soilliquefaction and permanent ground deformations are found to be likely and significant potential failure modes for the Diesel Generator building, the bus duct.
bank between the Diesel Generator building and the Turbine building, the fire pond piping, and the gas supply piping for the combustion turbines. For the buried duct bank and piping, the most credible failure modes are associated with the earthquake induced ground deformations. For the Diesel Generator building, soilliquefaction in the slope fill between the discharge canal and the building could lead to failure of the sheet pile structure retaining the slope which,in turn, results in failure of the building. Soil deformations are also estimated for the slope in the non-liquefied condition. The seismic fragility of the Diesel Generator building, therefore, includes different failure modes conditional on the occurrence of soilliquefaction: (1) soil-related failure given liquefaction occurs, and (2) structural failure or soit deformation related failure given that liquefaction does not occur. Additional discussion is included in the sections of the following chapter whie,h discuse the respective structures.
2HD 1146nbioystrch3.wf w 3-7 N3
3.5 CONSISTENCY BETWEEN HAZARD AND FRAGILITY DESCRIPTIONS As previously discussed, the fragility descriptions of all structures are reported in terms of the peak free field ground acceleration. The median response spectra shape for a 10,000 year return period from Reference 1 is selected as the basis for the fragilities, with variabilities determined from the time history response results from Reference 3. The hazard curves from Reference 1 are directly compatible with the fragility descriptions reported here.
If a comparison of seismic rick is desirable using a different description of the hazard, as for instance from Reference 2, the use of the fragility descriptions presented here will introduce some degree of inexactness. For instance, the shapes of the median 10.000 year return period spectra from References 1 and 2 are somewhat different, as would be the variabilities developed from time history runs using a different base median spectrum. However, these differences are overshadowed by the differences in the hazard descriptions. The median annual -
probability of exceedance of the Reference 2 hazard characterization exceeds the Reference 1 hazard by about a factor of 2 in the acceleration range of interest, and the 85th percentile Reference 2 hazard exceeds that of Reference 1 by about a factor of 5. Both hazard descriptions are keyed to the peak ground acceleration.
Since the Reference 2 spectral shapes are different and since the fragilities are based upon the median spectral shape, some different fragilities would be appropriate for the Reference 2 hazard. However,it is recommended that for an initial comparison of the seisniic 'isk, the fragilities presented here may be used interchangeably with the two ha zard descriptions.
4 gg mo, _....._ _
s.,
OYSTER CREEK 10-4 spectra 10 7e m
A 108 E
S' i
- 85 4
1 o
10 o
- 50 w
.1 15
<a:
F 10 uw c
rn 10-1 10-8 10-2 10 10 PERIOD (sec) 4 Figure 31:
Uniform Hazard Spectra For the 10-4 Annual Probability of Exceedance (reproduced from Figure 3196 of Reference 1) v I
L..
2MO INaeloyotech3,wfw 39
[
L
- 4. STRUCTURES i
The methods used in determining the median seismic factors of safety and logarithmic standard deviations for the Oyster Creek civil structures are described in this section. Based on these factors of safety, median capacities associated with seismic failure are presented. The estimated median capacities are anchored to the peak free field ground acceleration. For most structures, the linear dynamic analysis results from Reference 3 were used to determine the median factors of safety and logarithmic standard deviations for each variable affecting the structural response. For structures not analyzed in Reference 3, simplified analytical models were developed.
4.1 MEDIAN SAFETY FACTORS AND LOGARITHMIC STANDARD DEVIATIONS As discussed in Section 2.3, the seismic fragilities of structures are described in terms of the median free field peak ground acceleration A, and random and uncertainty logarithmic standard deviations, pg and pu. In estimating these fragility parameters, it is computationally attractive to work in teims of an intermediate random variable called the factor of safety, F. The factor of safety is defined as the ratio of the free field acceleration capacity, to the 3SSE reference earthquake acceleration used in the Reference 3 analysis. It is easier to estimate the median factor of safety, 5, and variability parameters, pg and pu, based upon the analysis thari it is to directly estimate the fragility parameters. Thus, A
FA (4-1)
=
3SSE From the Reference 3 analyses of the important structures together with a knowledge of the ultirnate seismic capacity behavior of structures, median factors of safety associated with the reference earthquake ground acceleration of 0.552g can be estimated. These are most conveniently separated into those factors l
associated with the seismic strength capacity and inelastic energy absorption capability of the structure and those factors associated with the expected building response.
The factor of safety for the structure seismic capacity Fcap, consists of the l
following parts:
~
p O
1 I
I
\\
The strength factor, F, based on the ratio of actual member 1.
S strength to the 0.552g reference earthquake forces 2.
The inelastic energy absorption factor, F, related to the ductility of the structure and to the magnitude range that is believed to contribute to most of the seismic risk Associated with the median strength factor, F, and the median ductility factor, F,
)
3 p
are the corresponding logarithmic standard deviations, ps, and p. The structure p
strength factors of safety and logarithmic standard deviations vary from structure-to-structure and according to the different failure modes of a given structure.
Factors of safety for the most important modes of failure are summarized in subsequent sections.
The factor of safety, Fn, related to building response includes consideration of a number of variables which include:
1.
The response spectra used as the basis for analysis 2.
Damping used in the analysis compared with damping expected at failure 3.
Modal combination methods 4.
Combination of earthquake components 5.
Modeling accuracy 6.
Soil-structure interaction effects In Reference 3, values of these parameters were selected to be median centered.
Thus, at the 3SSE motion, the median factors of safety associated with most of these parameters are essentially unity. Some variability about the median values is inherent, however. Some of this variability is calculated in the Reference 3 analyses while other portions are estimated by the fragility analyst.
Based on the characteristics of the lognormal distribution, median factors of safety and logarithmic standard deviations for the various contributing effects can be combined to yield the overall estimates. For instance, the capacity factor of safety l
(s]w(c; 2Sh P 2-o n.e. n,
4.w w e2
of a structure, Fcap,is obtained from the product of the strength and inelastic energy absorption factors of safety which, in turn, may include effects of more than one variable.
(4-2)
F ap =FSxFp e
The methods of determining these safety factors are discussed in the following sections. The logarithmic standard deviation on capacity, pcap, is found by:
Os + pj (4-3) pcap As discussed in Section 2.3, the logarithmic standard deviations are composed of both an inherent randomness and an uncertainty in the median value.
Median factor of safety, F, and variability, pg and PU, estimates are made for each of the parameters affecting capacity and response. These median and variability estimates are then combined using the properties of the lognormal distribution in the same manner as Equations 4-3 and 4 3 to obtain the overall median factor of safety and variability estimates required to define the fragility curve for the structure.
For each variable affecting the factor of safety, the random variability, Pa, and the uncertainty, pu, must be estimated separately. The random variability, pg, represents those sources of dispersion in the factor of safety which cannot be reduced by more detailed evaluation or by gathering more data. Thus, pg is due primarily to the variability of an earthquake time-history and, therefore, to a structure's response when the earthquake is only defined in terms of the peak effective ground acceleration. The uncertainty, pu, represents those sources of l
dispersion which could be reduced through better understanding or more knowledge. The uncertainty, pu,is associated with such items as our lack of ability l
to predict the exact strength of materials (concrete and steel) and of structural elements (shear walls and diaphragms); errors in calculated response due to l
inaccuracies in mass and stiffness representations as well as load distributions; and use of engineering judgment in the absence of plant specific data on fragility levels.
Each of the factors presented in Chapter 3 will be discussed in more detail in the following sections. Examples are included to assist in the understanding of the l
application of the methodology.
N.f5 2HD 1146ttloptu.h4.wf=
4-3 bh i
4.1.1 Structure Canacity The primary lateral load-carrying systems of the structures that are included in the Reference 3 analysis and some of the other plant structures are constructed of reinforced concrete.- For lateralload-carrying systems which are composed of concrete, the structure strength is a function of material strengths associated with the concrete and the reinforcing steel. The dstorminations of these strengths are presented in the following two sections.
4.1.1.1 Concrete Comorassive Stranath The evaluation of the strength of most concrete elements, whether loaded in compression or shear,is based on the concrete compressive strength, f'c' Concrete compressive strength used for design is normally specified as some value at a specific time from mixing (for example,28 or 90 days). This value is verified by laboratory testing of mix samples. The strength must meet specified values allowing a finite number of failures per number of trials. As previously stated, there are two major factors which justify the selection of a median value of concrete strength above the design strength.
1.
To meet the design specifications, the contractor attempts to create a mix that has an " average" strength above the design strength 2.
As concrete ages, it increases in strength The concrete utilized in the construction of the Oyster Creek structures was specified to have a minimum compressive strength of 3000 psi or 4000 psi at 28 days. Test data for the concrete cylinder specimens are not readily available.
4 l
As concrete ages, its strength increases. This must also be accounted for in determining the median strength compared to the design strength. Figure 4-1, i-taken from Figure 10.10 in Reference 20 shows the increase in concrete' L.
compressive st.ength with time in which the concrete poured-in-the-field is represented by the curve designated as " air-cured, dry at-test." In Figure 41, the relative strength is referenced to a sample given continuous standard moist curing which is also moist at test. Typical construction practices require moist curing for an initial short period of time.' Therefore, the in-situ concrete is considered " air S o i %,y w en4.w.w a
y__
I cured." At 28 days, the concrete has a relative strength of 50 percent which o
approaches 60 percent asymptotically. The median factor relating the strength of aged concrete to the 28-day strength'is, therefore,1.2. No information is available on the standard deviation expected for aging. A logarithmic standard deviation associated with the 28-day aging factors was estimated to be 0.10.
Based on GPUN's knowledge of practices during the original plant construction and current condition of the plant structures, the median concrete compressive strengths are taken as the 28 day design strengths with an increase for aging. The l
median values and logarithmic standard deviations for the two mixes se listed below:
3000 psi design strength:
3600 psi f'c
=
p 0.10
=
4000 psi design strength:
j'c 4800 psi
=
p 0.10
=
Other effects which could conceivably be included in the concrete strength evaluation include some decrease in strength in the in-place condition as opposed to the test cylinder strength, and some increase in strength resulting from rate of loading at the seismic response frequencies of the structure. Although experimental data on the in-place and rate effects are limited, that which is available would tend
{
to indicate these effects are relatively small and of the same order of magnitude.
Since the two effects are opposite, they are neglected.
4.1.1.2 Reinforcing Steel The reinforcement used in the construction of the Oyster Creek structures was specified to be Grade 40. Site specific test data for the reinforcing bars are not readily available. The median value and logarithmic standard deviation for the yield strength are based on available generic data in the literature (Reference 21). From the data for Grade 40 reinforcing.
!?A%!?
2Ho 1146nbloyotrob4.wfw 45 M
l
t j
48 ksi
=
y pfy 0.11
=
Two other effects must be considered when evaluating the yield strength of reinforcing steel. These are the variations in the cross-sectional areas of the bars and the effects of the rate of loading. A survey of information (Ref. 21) determined that the ratio of actusi to nominal bar area has a mean value of 0.99 and a coefficient of variation of 0.024. The same reference notes that the standard test rate of loading is 34 psi /sec. Accounting for the rate of loading anticipated in seismic response of structures results in a slight decrease in yield strength of reinforcing steelin tension. This effect is neglected in concrete compression.
4.1.1.3 Strenath of Plenar Concrete Shear Walls in Shear Studies show that the shear strength of low-rise concrete shear walls with boundary elements is conservatively predicted by the ACI 318-83 code provisions (Ref. 22). This is particularly true for walls with height to length ratios on the order of 1 or less. Barda (Ref. 23) determined that the ultimate shear strengths of tested low rise walls can be represented by the following relationship:
u Vc + Vs v
8.3 6 -3.46
- 0. 5
+ p.fy + 4 t -t (4-4) st.
where:
Ultimate shear strength, psi v
=
u c
Contribution from concrete, psi v
=
s Contribution from steel reinforcement, psi v
=
f'c Concrete compressive strength, psi
=
hw Wall height, in
=
'w Wall length, in
=
Vertical steel reinforcement ratio
- v
=
2Ho 114s,4ser.i.ch4.we=,
4-6
Axial load, Ib.
N
=
Steel yield strength, psi f
=
y Wall thickness, in t
=
The contribution of the concrete to the ultimste shear strength of the wallis shown l
in Figure 4 2 as a function of h / w. Also shown in Figure 4-2 are the available w
test values (Refs. 23 through 26) and the corresponding ACI 318-83 formulation.
l The tests included load reversals and varying reinforcement ratios and hw/ w l
ratios. Web crushing generally controlled the failure of the test specimens.. Testing was performed with no axialloads, but an increase in shear capacity of N/4l wt is recommended in Reference 23, where N is the uxial load in pounds, and t is the wall thickness in inches. The axial load, N, is comprised of both the dead weight and vertical seismic load Since the vertical seismic component may be either up or down at the instant of maximum horizontal response,it can either add to or subtract from the de,ad weight. In the case of an extreme earthquake, the upward vertical seismic component could overcome a large fraction of the dead weight load so that the increase in wall shear capacity could be minimized. For this reason, the increase in capacity due to the axial load term, N, is conservatively neglected for some structures.
The contribution of the steel to the ultimate shear strength according to ACl 318 83 (Ref. 21)is:
- f s
hy (4-5) v where P
= horizontal steel reinforcement ratio.
h One of the conclusions reached by Oesterie (Ref. 26) is that for sheer walls with l
w/ w 21, vertical steel has no effect, and the entire contribution to shear h
strength is due to the horizontal steel.
In order to estimate the effects that the horizontal and vertical steel have, the steel contribution to wall shear strength was determined from test values for the range of l
0.5 < h l w < 2. Test data from References 23 through 26 were used. The w
effective steel shear strength was taken to be in the form:
'Avsv + Bv h (4-6) v
=
se s
EO,f5 IHD li4Ge6/cystrsh4.wfw -
4-7 g
where A, B are constants and
- v y = vertical steel contribution to shear strength f
v sv (4-7) l
- fhy = horizontal steel contribution to shear strength vh s
l l
The constants A and B were then calculated assuming the concrete contribution to I
ths ultimate strength is given as shown in Equation 4-6. Based on the results of this evaluation, the constants A and B can be shown to be as follows for various l
values of hw/ w:
l A=1 B =0 hw/ w s 0.5 w/ w)-1'0 0.5 s hw/ w s 1.0 l
l l
2.0 (h
= -2.0 (hw/ w) + 2.0
=
l
=0 1
1.0 s hw/ w
=
(4-8) l and the median ultimate shear strength is given by:
v
" Vc + Vse u
1
= 8.3 6 - 3.4 6 '1 - 0.5 h
i N
+ 4 twt +#se Y (4'9I I
ela where Pse = APy+89 with A and B determined as shown above. The h
logarithmic standard deviation is estimated to be 0.20.
The data used to substantiate the median shear strength equations presented above are derived from tests conducted on cantilever walls with the load applied at the top of the wall. The height, hw, for these walls is well known. However, the walls evaluated in this study typically span rnore than one story and are loaded by seismic shear forces at each floor level. In order to apply equation (4-9) in a manner consistent with the data upon which it is based, an equivalent wall height is required to represent the interaction of moment and shear. For the Oyster Creek walls, the equivalent cantilever wall height, hwe, is taken as the ratio of the in-plane moment to the in-plane shear force at the section under consideration. Recall l
that for a tip loaded cantilever, the moment at any cross section is the product of l
the concentrated load and the distance between the load and the section beir g Ms 2Ho 1146nbleystach4Iwiw 4-3 bb a
consideredi The equivalent height, hwe, is used to determine the median wall shear strength.
Equation (4-9) is also applied to concrete floor slabs to evaluate the seismic shear capacity of floor diaphragms.
4.1.1.4 Examole of Shear Wall Failure in Shear The north wall on column line R7 of the Reactor Building just above elevation 23'-
6" is selected as an example. The wallis 18 inches thick and 109 feet long.
Concrete having a 28 day design strength of 3000 psiis used in this segment of the wall. The reinforcing in the wall consists of # 8 bars spaced at 6" at the outside face and
- 8 bars at 12" at the inside face in the vertical direction and # 6 bars at 8" at each face in the horizontal direction. The reinforcing is specified to be Grade 40.
The median material properties are as follows:
f'c = 3600 psi
'f = 48 ksi y
Based on the elastic force distribution in the wall resulting from the reference earthquake response analyses (Reference 3), the equivalent cantilever height of the wallis estimated to be 60 feet. The demand shear force at the reference 3SSE levelis 1307 k. The axial dead load force acting on the wallis:
NDL 6419 k
=
The vertical seismic force corresponding to the reference 0.552g earthquake motion is:
N 1271 k
=
y
- The median verticalload acting on the wall,in terms of the peak ground acceleration, then is given by:
- tso u4endinwa4.*t.
.:.3 l
Y I
l' #
N = 6419 - 0.4 O.552gs
(
'= 6419 - 921. A / g in which A is the peak ground acceleration.
Using equation (4 4),' the median concrete shear stress capacity,in units of psi as a function of A,is found as:
c = 8.3J3600 - 3.443600
- 0.5 + (6419 - 921 A / g) 1000 lb / k
' 60 v
4 18".109'12 in / ft s109 s
= 556 - 9.78 A / g The horizontal and vertical reinforcement ratios are 2
2 0.44 in ph =
= 0.00611 18" 8" 2
2 0.79 in 0.79 in p, =
+
= 0.011 18" 6" 18" 12" The effective height-to-length ratio of the wallis 60/109 = 0.55. Using equation (4-8) gives A = 0.90 and B = 0.10. Then, using equation (4 6), the effective steel reinforcing ratio and the steel shear strength is:
p., = (0.90 0.011) + (0.10 0.00611) l
= 0.0105 vs = 0.0105 48000 psi = 504 psi.
For a rectangular well with uniformly distributed vertical reinforcing, the effective depth, d, frcm the extreme compression fibe' to the resultant of the tensile force is r
taken to b's 0.6t, based e. : commendations in Reference 29.- The median
(
F fBRf2 i'
2Ho 1144ee/ovotrch4.wfw 4-10 GWG
ultimate shear capacity of the wallis found by equating the demand shear to the l
capacity.
1307k-
= [556 - 9.78 A / g + 504) 0.6 109t12 $ 18"+1000 $
^
ft k
O.552g Solving for A, gives A = 6.0g. The resulting ultimate sheer strength of the wallis then:
'8 Vu = 1307k.
= 14200 k O.552g 4.1.1.5 Strenoth of Planar Concrete Shear Walls in Flexure Under in-Plane Forces Equations to predict the overturning (in-plane) moment capacity of rectangular shear walls containing uniformly distributed vertical reinforcement are found in Reference
- 25. These equations are derived from the basic ultimate strength design provisions for reinforced concrete members subjected to flexure and axial loads contained in Section 10.2 of ACI 318 83. These provisions are based upon the satisfaction of force equilibrium and strain compatibility.
Equation 1 of Reference 25 can be used to predict the flexural strength of rectangular walls having uniformly distributed reinforcement. The accuracy of this equation has been verified by testing. Equation 2 of Reference 25 shown as Equation 410 below, is presented as an adequate approximation to Equation 1.
Nu 16C 1
Mu = 0.5 A fs y#w l
in - lb (4-10)
Asfys(
tw s where s
Total area of vertical reinforcement at section, sq. in.
A
=
f Yield strength of vertical reinforcement, pai
=
y l
I w Horizontallength of wall,in.
=
Distance from extreme compressive fiber to neutral axis, in.
c
=
2Ho 1148r4loyettch4.wfw 4 11
Axial load, positive in compression, Ib.
N
=
u Ratio of the depth of the equivalent concrete stress p1
=
block to the distance to the neutral axis (c) inspection of Equation 4-10 reveals that the overturning moment capacity of a rectangular wall can be adequately represented by lumping the total area of the uniformly distributed vertical reinforcement at midlength of the wall and applying the basic design provisions in Section 10.2 ACI 318 83 (Ref. 21).
1 in-lb.
(4-11 )
(A fsy+Nl M
=
u u
s2 2s This approach was typically used to predict the median flexural strength for walls without concentrated reinforcement. Concentrated reinforcement can be embedded steel columns well tied to the concrete wall or the vertical wall reinforcement bars within the effective flanges of the cross walls cast integrally with the wall being evaluated. The compression flange steelis typically neglected since it is near the neutral axis, and its effect on the moment capacity is small. The total moment capacity of reinforced concrete shear walls including concentrated reinforcement is then:
d. B c' (4-12)
(A fsy+Nl
- 1
+Affy 1
M
=
u u
( 2 2s 2s t
where Area of concentrated reinforcement steel Af
=
d Distance from the extreme compressive fiber to the
=
centroid of concentrated reinforcement steel 4.1.1.6 Examnia of Shear Wall Failure in Flexure The same wall segment used for the shear capacity example is again used here to
' illustrate the evaluation of the median flexural capacity.
A review of the drawing showing the detail of the vertical reinforcement in the wall at elevation 23*-6" indicates that the vertical #8 bars at the outer face do not have E-an adequate lap with the dowel bars at the construction joint to develop the N5
?w01146 eleystre%wfw 4 12 1:sN::1 "2:s
strength of the bars. Thus, only the inner face bars are assumed to be effective in resisting the overturning moment flexural tension.
Total vertical reinforcing contributed by the inner face bars ist 2
A, = 0.79 in 109'= 86.1 in, 1,
The depth of the compression block, a, in units of inches, is given by:
2 86.1 in 48 ksi + E419 k - 921. A a=
0.85(3.6 ksi) 18'
= 1916 - 16.7 A in which the axial load is included.
Using equation (4-11), the wall moment capacity is given by:
2 109' 1 '1916 16.7 A' ' '1 + 6419 - 921. A' Mu = 48 ksi 86.1 in
, 2 12s 2 2
s,.
48 8E1 This yields a quadratic equation for M in terms of A, the PGA. The demand u
moment acting on the wall for the reference earthquake motion is 78310 k ft.
Equating the expression for the moment capacity above to the scaled demand moment as:
A
% = 78310 0. 552g Solving the above for A, gives A = 2.74 g. Substituting A back into the expression for Mu gives:
Mu 389,000 k ft
=
This corres' ponds to a median strength factor of safety of 2.74/0.55 = 4.96.
IB'2hfd 2HD 1146nbloystict.4.*fw 4 13 b
5 (f
h i
4.1.2 Structure (nelastic Enerav Absorotion A much more accurate assessment of the seismic capacity of a structure can be obtained if the inelastic energy absorption of the structure is considered in addition to the strength capacity. One tractable method involves the use of ductility l
l modified response spectra to determine the reduced amplification resulting from the inelastic energy dissipation. Early studies indicated the inelastic energy absorption factor was primarily a function of the ductility ratio,, which is defined as the ratio of maximum displacement to displacement at yield. Additional studies (Ref.13) have shown that for single-degree of freedom systems with resistance functions characterized by elastic-perfectly plastic, bilinear, or stiffness-degrading models, the shape of the resistance function is, on the average, not particularly important. For structures subjected to earthquakes which produce more narrow banded response spectra where the spectral accelerations vary over the inelastic frequency range of the structure, the spectral averaging method (Ref.14) has been shown to produce somewhat more accurate results.
4.1.2.1 Ductility Corresoondina to Failure of a Structural Eiement in order to determine the factor of safety associated with inelastic energy absorption, F, it is necessary to determine the ductility p, corresponding to failure of the structural element. For the Oyster Creek structures the median ductility was taken to occur at inelastic deformations corresponding to the beginning of strength degradation of the controlling structural element. This is shown schematically as corresponding to a deformation of A in Figure 4 3. In actuality, force-deformation u
curves determined from cyclic tests of reinforced concrete members seldom show such a clearly defined change in the slope of the force deformation relationship as is shown in Figure 4-3, but rather show a curved transition region from the ultimate strength to degraded strength at larger deformations. Collapse of the structure should n )t be implied at ductilities corresponding to the ultimate strength since the structure retains the capability of resisting lateral load although at increased deformation. Rather, the selection of the ductility ratio corresponds to the onset of severe strength degradation as the median failure ductility is expected to provide a substantial margin against building collapse, but could potentially result in loss of some safety related equipment located within the structure. The deformation at YSh N
M e
4
V g
failure is often described in terms of story drift which is defined as the ratio of the i
deformation, A, divided by the length of the element.
y
. For Oyster Creek shear wall structures controlled by shear failures, the median story drift corresponding to severe strength degradation is expected to be about 0.007.
i Story drifts in the range of 0.007 will exhibit considerable cracking. A more conservative median story drift of 0.005 is used for structures whose controlling l
walls may have significant amounts of safety related piping, cable trays, conduit, or other equipment attached. Story drifts in the 0.005 range correspond to the onset.
of degradation of equipment attached to controlling walls. Since the majority of shear walls in the Oyster Creek structures support safety related equipment,0.005 is used as a median story drift for all Class I structures controlled by shear failures.
For shear wall structures governed by flexural failures, the ductility corresponding to failure was estimated from the available effective plastic hinge rotation capacity at
, the critical section of the wall. The hinge rotation capacity is limited by the maximum flexural strains in the wall, which for this evaluation, are taken as 0.003 for concrete compression and 0.04 for flexural reinforcement steel tension. Shear wall plastic hinge rotations are predicted using methods presented in References 27 and 28. Since the hinge rotation capacities depend on the well dimensions and the reinforcement pattern, they are evaluated on a case-by-case basis.
4.1.2.2 Multi Dearse of Freedom Svatem Ductility The ductility-modified response spectrum approach was developed from single-degree of freedom models. For the multi degree of freedom Oyster Creek structures, it was necessary to establish a system ductility rather than use the story ductility directly. The system ductility is determined from n
E W: Aui Psystem. i=1 (4 13) n IWe A.i i=,1
' where Wj is the tributary story weight, A i
ui s the structure drift using the modified maximum story displacement described in Section 4.1.2.3, and A,i s the elastic i
structure drift at yield.
ESS 2Mo 114enh/evowohe wtw 4 15 L9%.15
___---._.___1_____
For the structures included in the Reference 3 analyses, substantial soil-structure interaction effects are observed. As previously discussed, energy is dissipated by the soil as well as through inelastic structure deformation. At acceleration levels exceeding the yield level of the structure, a significant portion of the structure's total displacements relative to the free field can be attributed to the displacements (translation and rocking) of the structure foundation. The effects of the soil are included in the estimation of the system ductility by including the contributions of the foundation translation and rocking, relative to the free field,in the estimated deflected shapes at the structure yield level and ultimate level as well as by including the foundation mass in equation (4-13). This has the effect of significantly reducing the system ductility from that which would be obtained from a fixed base analysis.
Once the system ductility,# system,is determined from the mass-weighted story drift approach, a factor of safety is obtained using the method described in the following.
4.1.2.3 Inelastic Enerav Absorotion Factor Effective Riddell-Newmark Method)
Ory:e the system ductility corresponding to failure has been determined, the effective Riddell-Newmark method (Reference 10) is used to predict the inelastic energy absorption factor, F. The effective Riddell-Newmark method includes corrections to the original Riddell-Newmark method (Reference 13) for the shape of the yielding portion of the element force-deflection curve, pinched shear wall hysteresis behavior, and long duration ground motion input.
. The original Riddell-Newmark method is based upon the structural element strength defined by the yield st,rength and an elastic-perfectly plastic force-deflection curve.
However,' typical shear wall force-deflection relationships are idealized as bilinear as shown in Figure 4 3. Note that in Figure 4-3, moment, M, could be interchanged with V and curvature, $, could be interchanged with A. From the bilinear curve with an ultimate deflection of Au, an equivalent elastic perfectly plastic relation is obtained by equating the areas under the curves, which leads to values for Ay and A and the effective component ductility ratio, p', which is defined as A '/Ay*-
u u
The modified values, Ay
- and A ', are used in estimating the deflected shape in u
equation (413) to arrive at the effective system ductility p'. The effective p' is used to predict the inelastic energy absorption factor F as follows:
p h
Ho 114sntsov.i,ch4.wi.
4 16 l
._____-__________-a
[pp'-glf F
=
p Where p
q+1
=
3.0 -0.30 in the amplified acceleration region.
q
=
4 2.7 0.40 in the amplified velocity region.
=
4 0.48q-0.08 in the amplified acceleration region.
t
=
0.66 0.04 in the amplified velocity region.
=
4 pe cent of critical damping.
(
=
In accordance with the above relationships, the following definition of the Riddeli-Newmark inelastic absorption factor based on the system ductility is used for the structures whose fundamental frequencies are within the amplified acceleration region:
S..
p E
S.,
S e Spectral acceleration from the elastic response a
=
spectrum for the fundamental structure mode having a frequency in the amplified acceleration region.
S p Deamplified spectral acceleration accounting for a
=
nonlinear structure response.
Greater of Sa,A Or Sa, RIG, where
=
p p
Sa.A "
(P Feff 4) r (S,)
p S
l-a, RIG "
(FeffP (PGA) p i
l p,q.r are from Equation 417 1
j 0.11 7% damping at-
=
0.13 10% damping
=
' 280 i usne.<=vatrons.~<.1w '
.,. t 7
Peak ground acceleration PGA
=
Reinforced concrete walls exhibit pinched hysteresis behavior. This effect was not considered in the nonlinear resistance models used in the Riddell-Newmark analyses. For reinforced concrete elements with severe pinched hysteresis behavior, the inelastic energy dissipated is less than that calculated using the Riddell-Newmark approach. F ' is corrected for the hysteresis pinching and for the effects of long duration input ground motion to obtain the final median value for F as' k
1+CD ( F '- 1)
=
The correction factor CD was found to be approximately 0.6 based on the results of nonlinear time history analyses (Reference 10). This approach has been shown to produce good correlation with results from nonlinear time-history analysis (Ref.10).
Nonlinear time history analysis results for concrete structures with appropriately modeled hysteretic locp effects indicate that an SRSS combination of the randomness for independent ductility, damping, and response spectra effects significantly overestimates the combined randomness predicted by the time history analyses. It was found in Reference 10 that most of the random variability of F is p
attributed to the peak and valley variability of the input response spectrum, which is accounted for in the spectral shape factor variability. References 10 and 29.
provide an estimate for the additional randomness variability associated with F as pg=0.4 0.06 + 0.03($p - 1)
Reference 29 suggests that the uncertainty of F associated with the hysteresis p
behavior can be estimated as:
( = 0.1(F - 1) p 4.1.2.4 Example of the inelastic Enerav Absorotion Factor The reinforced concrete wall considered in the examples for shear and flexure capacity is used to illustrate the evaluation of the inelastic energy absorption factor using the effective Riddell-Newmark method. Flexure is found to be the controlling MO
(
- HD 114Cait,iuppd,4.wtw 4 18
{5 6 i
I
m I
l L
failure mode for the wall. The section at elevation 23'-6" is the most critical location of the wall.
L First, an idealized bilinear moment curvature relationship for the wall is defined. At the ultimate moment, Mu = 389,000 k-ft 4u = h = h =
= 2.10 x 10-4 ft -1
/P1 12.15/'0.85 e
a/
/
in which c is the depth from the extreme compression fiber to the neutral axis at the ultimate moment. Using current strength methods of analysis (Reference 22), e can be expressed as a/p, where a is the depth of the equivalent compressive 3
stress block found in the example in Section 4.1.1.6, and $ is the stress block 3
factor defined in Reference 22. For f'c _< 4000 psi, p, = 0.85.
From a strain compatibility analysis, the estimated yield moment and curvature are:
My = 0.9 Mu = 350,000 k ft
&y = 2.42 x 10 5 ft-1 where 4 is estimated by extrapolating the elastic cracked section moment y
curvature relation of the wall to a moment of 0.9Mu.
Referring to Figure 4-3, the area under the bilinear moment-curvature curve is (M + Mu)(fu - &y)
Area =
M
$y+
y y
5 3.50 x 10 2.42 x 10-5
=
2
+ d(1 + 0.9)t 5 2.10 x 10-4. - 2.42 x 10-5}
j 3.89 x 10 2
ir i
= 72.9 k.- ft/ft ggi
l
' Defining the equivalent elastic-perfectly plastic moment curvature relationship,
'$y = b b y = 2.69 x 10-5 ft 'l j
l l
Equating the area under the bilinear curve with the area for an equivalent elastic-1 1
perfectly plastic curve, l
1
-Mu ty' + Mu (fu'
- fy')
Area
=
72.9
=
Solving for $u', gives (u. - 2.01 x 10-4 ft -1. The effective curvature ductility is then:
2.01 x10-4 9'
= 7.5
=
2.69x10-5 To allow for a cyclic load reduction of the maximum curvature, the effective curvature ductility is reduced to a value of 6.
Since flexure is the controlling yielding behavior, the post-yield deflected shape of the concrete wall is based on the effective plastic hinge rotation at the critical section. The maximum wall plastic hinge rotation is estimated from the effective curvatures using relationships from lieferences 27 and 28 as op t, (+u' - +y'l
=
in which (p is the effective length of the plastic hinge. The effective hinge length is estimated to be approximately 5', which is about twice the length of the dowel lap at the base of the wall. Typical wall effective plastic hinge lengths are generally longer, but since the flexural yielding is limited by the bars at the inner face, a short effective hinge length is used. Using an effective curvature ductility of 6 which gives fu' = 1.61 x'10 ~4 ft ~1, the resulting maximum plastic hinge rotation is estimated to be 0.00673 radians.
The median deflected shape of the structure'at yield is estimated from the median structure deflections obtained in the reference earthquake analysis. As previously mentioned, the effects of the soil are incorporated by including in the deflected
+
FAA r:3 2Ho 1146nblevotech4.wfw 4 20 e
shape the contributions of the foundation translation and rotation measured relative to the free field. The foundation translation and rotation at yield are obtained by scaling up the values from the reference earthquake analyses at the 3SSE level.
Similarly, the ultimate deflected shape is estimated by including the foundation translation and rotation, the wall plastic hinge rotation, and the elastic deflections.
Using the estimated inelastic deflected shape and the masses from the elastic model, the modified system ductility is estimated as follows:
Yield Defl.
Estimated Elevation Mass Mi M x A j.
Aui.
M; x Aui' A j.
i y
y 2
(ft.)
(k sec /ft)
(ft)
(ft) 156.75 16.34
.2.46 40.3 2.79 45.5 138.00 25.85 2.43 62.9 2.74 70.9 119.00 270 2.39 645.5 2.69 725.8 95.25 474 2.38 1126.1 2.66 1259.0 75.25 543 2.36 1282.9 2.63 1427.6 51.25 527 2.35 1235.9 2.59 1367.3 23.50 715 2.32 1661.1 2.55 1826.7 0.00 607 2.31 1402.4 2.54 1542.5 19.00 1660 2.30 3820.5 2.53 4202.5 11278 12468 The effective system ductility is given by 12468 P,sys 11278 l
The modified Riddell-Newmark ductility factor of safety is then:
F' p
[p p'sys - Al'
=
The predominant frequency of the soil-structure system is in the velocity region of the spectrum.
~
2.70 -0.40 = 2.70(7)-0.40 = 1.24 Q-
=
4 2HD 114Snblevetech4.wfw 42) e L
4 I
l q + 1 = 2.24 p
=
0.66 -0.04 = 0.66(7)-0.04 = 0.61 r
=
4
[2.24 1.11 -1.24]O.61 1,14 F'
.=
p Correcting for the pinching effect of the concrete hysteresis and for long duration ground motions gives the median value of the effective Riddell-Newmark factor of safety for inelastic energy dissipation:
F 1 + Co ( Fp' 1) : Co = 0.6
=
p 1 + 0.6 (1.14 - 1)
=
1.08
=
Due to the large contribution of the foundation motions to the deflected shape, variability in the wall plastic hinge rotation will have a very small contribution to the variability of the inelastic energy absorption factor. The randomness and uncertainty variabilities associated with the plastic hinge rotation are neglected.
Therefore, the randomness variability associated with F, which is in addition to the variability associated with the peaks and valleys of the response spectrum is estimated by PR 0.4[0.06 + 0.03 (1.08 1))
=
0.02
=
The uncertainty variability is estimated from 0.1 $p - 1) pU
=
0.01
=
4.1.3 Structure Slidina Resistance to structure sliding is provided by static friction between the structure foundation and the soil below, lateral earth pressures from backfill placed against exterior walls, and any shea'r keys embedded into rock or soil. Gross structure sliding initiates when the base shear acting at the foundation-soil interface equals the available resistance. Initiation of sliding does not constitute structure or equipment failure. As a structure slides as a rigid body,its accelerations and relative story drifts cannot exceed those values occurring at the initiation of sliding.
ECM 2HD 114firelayetteh4 wfur 4 22 L.'::3W "
l l
l Failure modes resulting from structure sliding are displacement-dependent. For example, piping attached at one end to the structure that is sliding and at the other end to some adjacent structures may cause concrete spalling and subsequent damage to equipment or piping mounted near the localized spalling regions.
However, the sliding displacements necessary to cause these failure modes are substantial and can occur only under peak ground accelerations well in excess of acceleration levels initiating sliding. Sliding was found to be the controlling failure mode for the Fire Pond Pump House.
Studies have been performed by Newmark (Ref. 30) and Kausel, et al., (Ref. 31) to predict structure sliding. Newmark (Ref. 30) developed methods for predicting sliding displacements for structures with symmetric sliding resistance (no preferential sliding direction) and unsymmetric sliding resistance (preferential sliding direction). A method has recently been developed by Danay and Adeghe (Reference 32) to predict the sliding displacements of concrete gravity dams.
Because the method was developed for dams,it only applies to sliding systems with a preferential sliding direction. That is, the maximum sliding displacement is the result of sliding motions that ratchet outward. The prediction equation was developed through regression anslysis from the results of riuttiple time history analyses which included parameter variation with empirical and synthetic earthquake ground motions. An important feature of the method is that earthquake ground motions characteristic of both western end eastern North America were used in the parameter studies. The maximum sliding displacement is predicted from:
s(mm) = 251 C v.8 A-0.82 10-3.738, t 1
where maximum sliding displacement, mm.
s
=
C equation confidance factor
=
1 for median displacement estimate
=
2 for 98% confidence sliding displacement
=
peak ground velocity, mm/sec.
v
=
peak ground acceleration, mm/sec2, A
=
rP A rP 2MD 1146nbloyotrch4.wfw 4 23 5N
critical acceleration ratio, a/A.
R
=
critical horizontal quasi-static ground acceleration at a
=
which sliding initiates, including the effect of vertical acceleration.
amplification factor, taken as unity here for rigid L
=
structures.
4.1.3.1 Examole of Slidino induced Failure The critical failure mode for the Fire Pond Pump House is controlled by sliding. This structure is selected to illustrate the estimation of the peak ground acceleration capacity associated with sliding induced failure.
The pump house is constructed of reinforced concrete below grade with a one story, light, structural steel and sheet metal structure above grade, it is embedded in the soil on three sides, with the fire pond water on the fourth side. Based on its configuration, the sliding direction with t ie minimum resistance is toward the water. Static and seismic loads as wel. as sliding resistances for the pump house structure are provided in Reference 4. The seismic induced loads are expressed in terms of PGA.
Based on the loads and resistances in Reference 4, the quasi-static peak ground acceleration required to iniate sliding, a, is found to be 0.36 g. This value includes the effects of the static and dynamic active soil forces, the seismic hydrodynamic l
force, the buoyant effect of the hydrostatic pressure acting on the bottom of the i
1 l
foundation, and the vertical earthquake acceleration.
A median sliding displacement to failure is estimated to be 2 inches. During the walkdown, the pump house was not entered, so the connections of the fire pond piping leading away from the pump house could not be examined. Detail drawings of the connections were also not readily available. Thus, the median sliding failure displacement is based on judgment.
Due to the very non-linear form of the sliding displacement equation, an explicit l-solution for the peak ground acceleration, A, corresponding to a given sliding l
l displacement was not obtained. Rather, a range of displacement was estimated for I
fashre 2nD ii46nclovetrch4.wfw 4 24
$Y@
(
' a range of values of A using the median v/A ratio for the Oyster Creek site, which is estimated to be 13 in/sec/g, or equivalently,330.2 mm/sec/g. The median v/a ratio is derived from the median response spectrum shape defined by Reference 1. This lead to a value of A = 1.21g or 11866 mm/sec2 to correspond to a sliding displacement of 2 inches.
a/A R'
=
0.36/1.21 = 0.298
=
(v/A) A v
=
330.2 mm/sec/g X 1.21 g = 399.5 mm/sec
=
Substituting the appropriate values into the sliding displacement equation:
s(mm) = 251 1-(399.5)1.8 (11866)4.82 10-3.73/0.298, j
= 50.7mm
= 2.0in This then corresponds to a strength factor of safety with respect to the 0.552g reference earthquake of:
9 is = 0.552g= 2.19 4.1.4 Structure Response Factors The Oyster Creek structures were analyzed in Reference 3 using median centered properties with elastic models of the structures. Consequently, the median fa.ctors of safety for most of 'the parameters associated with structure response are unity
~
I although variability is inherent in these factors of safety.
Median loads and variabilities were determined for the member shears and moments by calculating the median and 84th percentile values of the peak loads in a given element from the series of time history results. Similarly,' median and 84th percentile nodal displacements were determined from the time history analyses.
The median loads for the 3SSE or 0.552g level obtained in Reference 3 serve as the median response for the fragility evaluations.
perp 2Ho 1144nblovetteh4.wfw 4-25 t-
4.1.4.1 Ground Resoonse Soectra The median ground response spectra which provide the basis for the Reference 3 analysis were selected as the 50th percentile,10000 year return period response spectra from Reference 1. These spectra are smoothed to eliminate the peaks and valleys inherent in the resoonse spectra resulting from natural earthquakes. The time histories selected for the Reference 3 analysis correspond approximately to the median Reference 1 response spectrum, in the Reference 3 analysis, the variability in the response spectra is included in the overall variability in the structure response. Since the structure stiffness and damping are also included in these analyses as random variables, the overall variability in structure response includes their contribution also. The individual contributions from earthquake time history, stiffness, and elastic structure damping cannot readily be separated from the total structure variability for the structures included in the probabilistic SSI analysis. This variability is assigned to be all randomness, PR, and is evaluated from the 84th percentile and median responses as pg=In ( R>
in which the response variable, R, is either the structure force quantity or the spectral acceleration.
i For those structures not included in the probabilistic SSI analyses, the peak and I
valley variability of the ground response spectra is treated separately from the structure response variability (stiffness and damping).
4.1.4.2 Peak Horizontal Resoonse The median shape of the input response spectrum (Reference 1) is based on the average of the two horizontal components. A corresponding rando~ ness variability m
exists since the smoothed spectrum of one component may be greater than the average, while the other is less than the average. For the structures included in the i
Reference 3 analyses, time history analyses were performed in which three 1
components of earthquake ground motion were independently generated.
Therefore, the time history analyses are median centered and the variability associated with the peak horizontal response relative to the average is included in the total response variability reported in Reference 3.
l mo manw,.wh4.wr.
4 26 u
.i For structures not included in the probabilistic analyses, the median ratio of the peak horizontal response relative to the average horizontal response and the associated randomness variability is evaluated separately. Reference 29 provides guidance for these values for a range of relative contributions of the horizontal j
responses to the total response. For a structure whose critical failure mode is dominated by one horizontal component response, the response based on the average horizontal spectrum is median centered, but a randomness variability, PR.
of 0.12 is included to represent the variability.
4.1.4.3 Structure Damoina The Reference 3 analysis is based on a median elastic structure damping of 7% of critical as previously noted., For the fragility of concrete structures,5% of critical is taken as a minus one standard deviation value. Damping was varied in the Reference 3 analyses using latin-hypercube techniques and the overall variability in the response calculated in Reference 3 includes that for damping.
For the structures not included in the probabilistic SSI analyses, the *,ariability in elastic damping can then be approximated as S,(G) p = - (n S,(q_ip)>
where S,(G) is the spectral acceleration at the structure fundamental frequency corresponding to the median damping and S,((_ip) is the spectral acceleration at the same frequency corresponding to the -1p damping value. The median and -1p damping value depend on the structure being evaluated.
l l-4.1.4.4 Modal Combination 1:
The analyses conducted in Reference 3 by use of time history methods implicitly includes the phasing of the individual modal responses although different time histories may result in different phasing. It should be noted that the analyses conducted in Reference 3 includes the variation in the structure stiffness and damping in the response. Thus, the randomness in response includes not only the bb
~
- no 114snbio iech4 of.
4 27 t.::e --
t
variation in earthquake response spectra and phasing but some contribution from stiffness and damping which is not readily separable. For the structures not included in the probabilistic SSI analyses, a nominal Bg of 0.05 was used in some cases to provide for some possible contribution from higher modes. This value is considered sufficient because all such structures are essentially single mode structures.
4.1.4.5 Combination of Earthauake Comoonents The three orthogonal time history records are input simultaneously into the individual structure models in the Reference 3 analysis. Thus, the directional components including torsional response is included in the response results for those structures analyzed in Reference 3.
Combination of earthquake components for structures not analyzed in Reference 3 was accomplished by taking 100 percent of the effects due to motion in one direction and 40 percent of the effects from the two remaining principal directions of motion (Reference 15L This combination is considered to be median centered.
For shear wall structures where the shear walls in the two principal directions act essentially independently and are the controlling elements, the two horizontal loads do not combine to a significant degree except for the torsional coupling. Thus, only the vertical component affects the individual shear wall stress resulting from horizontal motion in one principal direction. A moderate amount of vertical load slightly increases the ultimate shear load carrying capacity of reinforced concrete walls, while the overturning moment capacity may be more significantly affected.
The effect of the vertical dead load on the wall capacities is sometimes conservatively neglected. In these cases, the effect of the vertical seismic component on the capacities and the earthquake component combination variabilities is not included since these capacities already contain a degree of conservatism by not including the dead load. In other cases, where the increase in capacity due to the dead load is included, the effect of the vertical seismic response on the capacity and the earthquake component combination variability is also included.
W 2HD 1146nbloystsch4.wfw 4-25 l':i?2'"'7 1
I I
I l
4.1.4.6 Soil Structure Interaction (SSI)
The Reference 3 analysis includes soil structure interaction effects which include the soil compliance functions and the embedment effects of the structure in the soil. The SSI analysis is considered median-centered, and the variation of the soil properties is included in the structure response variability. The effects of ground motion incoherence are not included iri the Reference 3 analyses. These effects are not expected to be significant and are neglected.
Soil-structure interaction effects are considered separately for the combustion turbine fuel oil tank. The median response and variabilities are estimated based on simplified procedures. For the other structures which are controlled by earthquake induced soilliquefaction or permanent deformation, the median responses and variabilities are given in Reference 4.
4.1.4.7 Structure Modelina Considerations The structure models used in Reference 3 are considered to be essentially median centered in the elastic range. In these analyses, the stiffness is varied using the Latin hypercube technique. The structure stiffness is uniformly varied through the structure material stiffness such that variations of the structure frequency are included in the structure response variability, but variations of the structure mode shape are not. Therefore, an additional uncertainty associated with the structure mode shape is included separately.
4.2 STRUCTURE FRAGILITIES The controlling failure m' las for each of the Oyster Creek structures included in A
this study have been evaluated. The median load demands and variabilities corresponding to the reference earthquake motion,3SSE, are evaluated in the Reference 3 analysis. In some cases, the structure :apacities are found to be high and thus, a statement of high capacity is made without a complete fragility evaluation. The resulting fragilities for the structures are discussed in the following sections. Where applicable, fragilities are reported for both the seismic inertia related failure modes and earthquake induced soilliquefaction or deformation failure modes. The fragility results are summarized in Tables 41 through 4 8. Tables 4-1
^
through 4 7 summarize the results for the major structures. Table 4 8 summarizes 2Ho 114Snbleptsch4.ntw 4 29 b
U
the results for the balance of the structures and components considered in this evaluation.
4.2.1 Reactor Buildina The Reactor Building (RB) consists of four main sub-structures: the main structure, the drywell, the biological reactor shield wall (BSW) and the reactor pressure vessel (RPV). The main structure is a rectangular reinforced concrete structure up to the operating floor at El.119'-3". Above the operating floor, the structure consists of a steel frame with insulated metal siding. The drywell containment vessel is an axisymmetric steel shell surrounded by a heavy concrete shield wall which follows the contour of~the vessel from the foundation of the drywell to the operating floor.
The interface between the RB, Drywell, BSW and RPV includes the RPV stabilizer, the Star Truss, the Drywell lugs, and the radial support beams of two steel platforms at El. 46'-0* and 23' 4.5*.
Failure modes investigated for the reinforced concrete portion of this building include shear and overturning failure of the perimeter and the drywell walls, and diaphragm shear failure. Additional failure modes investigated include shear and overturning failure of the concrete pedestal, failure of the BSW anchorage, failure of the star truss at the top of the BSW, failure of the steel roof trusses, and failure of the steel superstructure column anchorage. The controlling mode of failure for the reinforced concrete is overturning of the North exterior wall between El. 23'-6" and 51' 3". The shear and flexural capacities of this segment of the wall are calculated in sections 4.1.1.4 and 4.1.1.6 of this report.
I As previously noted, the critical section for flexure is located at the construction j
joint at El 23'-6" where the flexural tension capacity is limited to the contribution of only the inner curtain of vertical reinforcing as the outer curtain dowel bars do i
not have an adequate lap splice to develop their strength. Also, at this elevation, the thickness of the wall decreases and the concrete design strength decreases.
The median acce'aration capacity is estimated to be approximately 2.96g. Median f' actors of safety and variabilities are listed in Table 4-1. Failure of this wallis expected to lead to damage to safety related. equipment inside the RB. The controlling failure mode for the steel superstructure is the anchorage of the corner columns. The anchorage consists of 41 %" diameter J-bolts and is controlled by pull out. The median ground acceleration capacity is estimated to be 1.Og. The 2Ho 1146nbloyettch4.wtw 4 30 thiM%
I I
l l
l.
median factors of safety and variabilities are shown in Table 4-2. Failure of the column anchorage may potentially lead to the collapse of the crane supported by the steel superstructure. An evaluation of the crane itself was not performed in this study.
As noted in Reference 4, no significant soil liquefaction or settlement which could affect this structure is expected due to the earthouAs ground motion.
4.2.2 Turbine Buildino Several structural systems of the Turbine Building (TB) are considered in the fragility evaluation. The main structure of the Turbine building includes the lower portion which consists of concrete shear walls with concrete beams and slabs. The reinforced concrete portion is located from the foundation mat to the turbine main operating floor at El. 46'-6". The upper portion is a steel superstructure above El.
46'-6", which consists of steel columns support;ng long span roof trusses running in the E W direction and steel braced frames in the N-S direction. The turbine pedestal is a heavy reinforced concrete frame and shear wall structure which supports the turbine and other equipment. The turbine pedestalis supported on a
)
common foundation mat with the main structure. Also included in the scope of work for the Turbine building is the HVAC platform located on the north side of the building, just outside the control room. The capacity of each of these systems are evaluated separately.
The capacity of the concrete portion of the main Turbine building structure is governed by shear failure of the floor diaphragm at El. 46'-6" due to EW seismic
- loads. The criticallocation is near the south-west corner of the building, where a floor slab opening reduces the effective area for shear transfer. Other potential failure modes that were examined include shear and flexure of the main shear walls and the failure of diaphragms at elevations 23'-6", and 63'-6" (the concrets slab over the control room). The median acceleration capacity is estimated to be approximately 0.88g. Median factors of safety and variabilities are listed in Table l
4 3.
The capacity of the steel portion of the structure is governed by an anchorage failure of the columns on Line F which carry EW disection frame loads and NS l
direction braced frame loads. The median capacity acceleration was found to be j
IE ^
2Ho 1146r@/oyotrCh4.wfw 4 31
l approximately 0.74g. Median factors of safety and variabilities are listed in Table j
i 4-4. Failure of the anchorage is expected to result in a loss of stability of the steel frames, leading to a potential partial collapse of the column or a collapse of the crane. This could result in impacts with the floor slab at El. 46'-6", which could
)
I affect safety related equipment or potentially induce relay chatter in electrical equipment in the control room.
The lateral force resistance of the turbine pedestalis provided by reinforced concrete beams, columns, and shear walls. In the NS direction, the beams and columns form a frame to resist the seismic loads. In the EW direction, three shear walls, which are located approximately at the. center of the pedestal, and four one-
' bay frames resist the seismic loads. Failure modes investigated for the pedestal include shear and flexural failure of the shear walls as well as shear and flexural failure of the besms and columns. The shear capacities of the beams and columns exceed the flexural capacities such that the governing failure mode corresponds to the development of a plastic hinge mechanism of the EW frame at column line 3.
The median ground acceleration capacity is estimated to be 2.17g. The median factors of safety and the logarithmic standard deviations are shown in Table 4 5.
The outdoor HVAC platform, located at El. 41', consists of steel framing with steel grating which supports three fans. The south end of the platform is anchored to the north concrete wall of the Turbine building, while the north end of the platform is supported by steel columns. Steel angle diagonal braces provide the seismic diaphragm for the platform to carry load back to the Turbine building wall. A review of the diaphragm brace and the platform anchorage indicates that the ground acceleration capacity of the platform exceeds that of the main situcture's concrete diaphragm at El. 46' 6". Therefore, the platform does not control.
From Reference 4, the free-field PGA required to cause liquefaction of the fill beneath the Turbine building exceeds 0.70g. This acceleration corresponds to a high confidence value. Significant settlements are unlikely without liquefaction.
4.2.3 Intake Structure The intake Structure (IS)is a partially embedded reinforced concrete structure, L
partly filled with water. The main lateral force resisting system consists of l'
concrete slabs in the water intake tunnel at El. 2'-6", at the operating floor at El. 6'-
N.@
2Mo 1145.bicystrd.4. s 4-32 GW I
i-
r 0", and at the roof, the 2.5' to 3.5' thick concrete walls in the EW direction, and a 3.5' thick wall in the NS direction.
Failure modes investigated include shear and overturning failure of the perimeter walls, and shear failure of floor diaphragms. The capacity of the IS is governed by a shear failure of the diaphragm at Ei,6' for NS loads. At this elevation, the diaphragm is perforated with numerous openings. The median acceleration capacity is estimated to be approximately 0.82g. Median factors of safety and variabilities are listed in Table 4-6. Failure of this floor slab is expected to lead to damage to safety-related equipment located on'the floor.
i From Reference 4, significant liquefaction or settlements are unlikely beneath the intake structure for any ground motion.
4.2.4 Emeroency Diesel Generator Buildina The Imergency Diesel Generator Building (EDGB) is a reinforced concrete shear wall structure, supported by a concrete foundation mat. While the walls and the foundation slab ars monolithic, the roof consists of a series of precast panels which are bolted to the walls on two sides, but not to one another. The precast roof panels are bolted to the long NS direction walls. No shear transfer continuity is -
provided between the panels to span the EW seismic loads to the EW direction walls. This then forces the EW seismic roof inertia loads near the center of t'he building to be resisted by the out-of-plane bending of the long NS walls, with the walls acting as cantilevers fixed at the foundation. Significant failure modes are identified for both seismic inertia loading and for seismic induced ground deformation.
The seismic inertia induced failure modes investigated for this building include shear and overturning failure of the perimeter walls, shear failure of the roof panels, anchorage failure of the roof panels, and sliding of the building. The controlling mode of failure is out-of-plane bending of the N S walls. The median acceleration capacity is estimated to be approximately 1.18g. Median factors of safety and variabilities are listed in Table 4-7. Failure of these walls is expected to lead to
' ^
damage to safety-related equipment inside the EDGB.
2hD 11'46teioyotron4.wtw
'4 33 hh
_ _ _ _ _ _ _ - _ _ - _ _ = _ _ _ _ _ _ _ _ _ _ _
Potential soil related failures are identified in Reference 4. The soil related failure of the Diesel Generator building is expected to be controlled by the sheet pile system located at the toe of the slope at the discharge canal. The extent of fill beneath the DG building is not known exactly (Reference 4). However, it appears possible that sufficient fill exists under the DG building that failure of the sheet pile restraint system could at least result in failure of the day tank piping.
In addition to the sheet piles, fender piles are located at 10 foot centers and the tie back system consists of a 1-1/2 inch diameter and a 2-1/2 inch diameter rod, each with turnbuckles, at 10 foot centers. Seismic loads acting on the wall for the liquefied and non-liquefied conditions are given in Reference 4. Liquefaction of the fillis expected *" a free-field PGA of about 0.40g. Below is shown the soil-cont :::ed seismic fragility for the DG building with the fillin the liquefied condition.
0.69 g
=
P 0.14
=
R pu 0.28
=
HCLPF =
0.69 e 1.65(0.14 +0.28) = 0.35 g The seismic capacity for the non-liquefied condition is above the range of interest.
In the preceding discussion, the structure failure mode associated with the out-of-plane bending of the NS walls is based on the assumption that liquefaction does not occur. However, the soil deformation failure mode is based on the occurrence of 1
liquefaction. Since the liquefaction is a probabilistic event, different outcomes or failure modes are possible depending on whether liquefaction occurs. in developing the risk model for the seismic failure, a conditional model is required for the Diesel Generator building. This conditional model must include the probability that liquefaction occurs, then include the failure mode given that liquefaction does or does not occur. Specifically, the occurrence of liquefaction is def;ned by a fragility with a median peak ground acceleration capacity and variabilities. The failure mode of the EDGB is conditional on the occurrence of liquefaction. Given that i
l j
l 2*o S ide etcystrche.w*w 4 0 *,
em L
liquefaction does not occur, the governing failure mode of the EDGB is the out-of-plane bending of the NS walls. On the other hand, given that liquefaction occurs.
the failure mode is controlled by the soil deformations caused by the failure of the sheet pile system at the discharge canel.
Based on the results provided in Reference 4, the median PGA at which liquefaction 1
is expected to occur is O.4g and the randomness variability, PR, is 0.14.
4.2.5 Bus Duct Bank From the Diesel Generator Buildina to the Turbine Buildina The bus duct from the DG building to the Turbine building consists of two reinforced concrete duct banks. Each bank is 24 inches deep with 4 inch diameter
' cavities. The first bank is 12 inches wide with three cavities and the second is 18
. inches wide with six cavities. Both are reinforced with #4 bars spaced at 8 inch centers top and bottom with an additional #4 bar at mid-depth at each face. No. 3 bar ties spaced at 18 inches on conter are specified. The reinforcing steelis Grade
- 40. The concrete is assumed to have a 28 day design strength of 3000 psi.
Significant soil settlements are estimated in Reference 4 for the soil backfill adjacent to the Turbine building. The area where such settlements are postulated extends about 50 to 60 feet from the face of the Turbine building. Uncertainty in the compacted properties of the backfillleads to an estimated median settlement of 3 inches with 5% and 95% confidence bounds of 1.5 to 8.3 inches.
Assuming a worst case condition of 42 inches of soil cover above the duct bank and' assuming none of the cover is dislodged by the ground shaking, it was found j
that the duct bank can be expected to form plastic hinges as a result of the most severe ground settlement estimated in Peference 4. However, the required displacement ductility is less than 3 for the N1 = 10 isoil condition 95th percentile settlement of 8.3 inches. Park and Paulay (Reference 27) indicate that much larger -
I ductility ratios can be expected prior to failure for the above concrete and steel strengths. Since the shear capacity of the duct bank is significantly greater than ihe flexural capacity, it is concluded that even if significant soil settlement should occur in the region of the Diesel Generator Tu'rbine building, failure of the electrical cables within the duct bank is unlikely, m.
s l.e6 2HD 114snblevotech4.wfw 4 35 ie
4.2.6 Fire Pond Pumo House, Fire Pond Dam. Fire Pond Pinina Fire Pond Pumo House The Fire Pond Pump House is a fully embedded concrete box-type structure, partly filled with. water. The main lateral force resisting system consists of a concrete slaF at grade, concrete shear walls in the N-S direction, and a 1' shear wall in the.
E W direction. A steel frame supporting a monorail and several other pieces of equipment are anchored to the slab at grade. A sheet metal structure encloses the
- equipment at grade.
The controlling failure mode of this structure is sliding of the building. Sliding initiates when the seismic base shear and the soil active forces overcome the sliding resistance forces. However, sliding-induced failure does not occur until sufficient displacement is developed to cause significant structural damage and/or damage to safety-related equipment. Failure of the pump house is expected to correspond to a sliding displacement of approximately 2 inches. For this displacement, damage to the fire water piping leaving the pump house is expected.
As previously mentioned, the pump house was not entered and the pipe connections were not examined. Therefore, the median displacement to failure is based on judgment. ' The median acceleration capacity is approximately 1.219, with the onset of sliding starting at around 0.36g. Median factors of safety and variabilities are listed in Table 4 8.
Based on Reference 4, significant liquefaction or settlements are not expected for any ground motion.
Fire Water Buried Pinina
- The seismic capacity of the buried fire water piping between the fire pump house and the plant fire protection lines is expected to be controlled by possible ground movement where it crosses the dike separating the intake and discharge canals. It E
is assumed (Reference 4) that fill was placed above the reinforced concrete anchor wall and tiebacks. This fill could be expected to liquefy at about a peak ground
. acceleration of 0.4 g.
l l
+
er e
N
The sheet pile at the dike is restrained by 3-1/2 inch diameter tie rods with turnbuckles on 9 foot centers anchored in a reinforced concrete anchor wall. The wall is in turn supported by vertical and batter piles. Lateral hydrodynamic and soil loads for both the liquefied and non-liquefied soil conditions as a function of PGA are provided in Reference 4. Failure of the sheet pile is assumed to lead to sufficient ground displacement to fail the fire piping.
The seismic capacity of the sheet pile is controlled by the 3-1/2 inch ties. For the liquefied soil condition, this failure is expected at a median PGA of over 1.5 g with a High Confidence of Low Probability of Failure (HCLPF) of over 1 g as shown below.
d 1.57 g
=
0.14 pg
=
pu 0.10
=
HCLPF =
1,57 e 1.65(0.14 +0.10) = 1.06 g The seismic capacity for the non-liquefied soil condition is much higher and will not control.
Fire Pond Dam Based on Reference 4', significant liquefaction, lateral movements, or settlements are not expected for any ground motion. The dam is therefore judged to have capacity beyond the range of interest.
4.2.7 Circulating Water intake and Discharoe Tunnels and Outfall Discharae Structure The Circulating Water intake and Discharge Tunnels are buried reinforced concrete rectangular tunnels. The tunnels run in the EW direction and are founded at El. - 14'. The intake tunnelis located on top of the discharge tunnel. Expansion joints are provided in the tunne!s to allow longitudinal movement. The intake tunnel
(
runs from the east side of the intake structure to the Turbine building. The discharge tunnel _ runs from the Turbine building toward the discharge canal. The 36.5 2nD i 646norovettene.wtw 43/
M
discharge structure is also reinforced concrete and extends from the discharge tunnel at an angle, opening to the discharge canal.
Adequate longitudinal flexibility is provided in the expansion joints to withstand the ground shaking effects. Also, based on the results in Reference 4, significant soil liquefaction or settlements are not expected beneath these structures for any ground shaking. Therefore, the seismic fragility of the intake and Discharge tunnels and the Discharge structure is judged to be beyond the range of interest.
4.2.8 Combustion Turbines Fuel Tank. Gas Sunniv Pinino Based on a review of drawings, Reference 4, and the walkdown observations, the most vulnerable components of the combustion turbine system are the fuel oil tank and the gas supply piping. Significant soil settlements beneath the gas turbine building and the fuel oil tank are not expected at any level of ground motion (Reference 4).
Combustion Turbine Fuel Oil Tank The Combustion Turbine Fuel Oil tank is a 1 million gallon capacity unanchored cylindrical steel tank, with a radius of 30 ft and a height of 48 ft. The shell is f abricated from ASTM A283, Grade C steel, with the wall thickness varying from
+
0.344 inches at the base to 0.25 inches at the top. The tank has a conical roof.
The bottom plate is 0.25 inches thick. The tank is supported on a reinforced concrete ring wall and the bottom plate is supported by a sand cushion over a 9 inch thick concrete slab.
To evaluate the median response, the deterministic procedures recommended in Reference 33 are used with median centered input parameters. Soil-tank interaction effects are included by following Veletsos's recommendations in References 18 and 19. The tank capacity is controlled by shell buckling due to base moment. This failure' mode dominates over other potential failure modes, such as sliding or shell hoop tension failure, because this tank is unanchored and thus has relatively low resistance against seismic-induced base uplift. The base moment capacity is determined following the evaluation procedures recommended in Reference 34. The inelastic energy absorption factor is determined using the procedures recommended in Amndix M of Reference 33.
1P4%f2
' 2HD 1149nbloystrch4 wfw 4 3h
$b
l The fragility parameters for the fuel oil tank are as follows:
1.
0.66 g
=
Pa 0.37
=
PU 0.39
=
HCLPF =
0.66 e-1.65(0.37 + 0.39) o,1 g g d
Gas Sucolv Pioina Soil related failure of the combustion turbine system is expected to be controlled either by horizontal or vertical soil displacement of the soil supporting the gas supply piping. The gas supply pipe is located at the top of the slope on the west side of the intake canal. A minimum of 3 feet of covet is specified. Thus, the pipe may be partially surrounded by black peat or the fine to medium sand of the roadbed.
The pipe consists primarily of 16-inch schedule 30 (std.) pipe with some 10-inch schedule 40 pipe in the vicinity of the combustion turbine building and 16-inch schedule 80 pipe supported by the bridge. The gas pipe supported by the bridge is supported from the main bridge girder. The supply end is connected to the New Jersey Natural Gas Co. metering station by a 16 inch 600 lb. bolted flange. From I
Reference 4, the greatest soil displacement as a function of ground acceleration is expected at Trench no. 4 and 68. Both of these locations are in the 16 inch schedule 30 pipe portion of the line, and the seismic fragility is based on soil displacements for these locations from Reference 4.
Based on a weighted average of the modulus of subgrade reaction.for the sand and l
' the peat, an offset displacement of about 4 inches (10 cm.) is required to develop a fully plastic moment in the pipe. Since the pipe is ductile and can be expected to withstand additional displacement, the 4 inch displacement can be considered a conservative estimate of failure displacement. For the worst case trenches (nos. 4 I
t and 68),4 inches displacement corresponds to a PGA of about 0.75g.
A check was also made on the bending moment capacity of the 600 lb. flange.
4 i
Some slight leakage past the gasket seal may occur at bending moments less than j
\\
WAP 2Ho 1146nbleysttcM vvf.
4-39 M
- a
l l
l the pipe fully plastic bending moment. Assuming the same soil condition at the flange as at Trench no. 4, this would indicate some slight leakage could occur at PGA's less than 0.75g. This minor leakage would not be expected to result in loss of the combustion turbines unless a source of ignition were present near the leak, in order to reach the elastic limit of the flange bolts and hence possibly result in a significant leak, PGA's wellin excess of 1g with the maximum ground offset near the flange are required.
The fragility of the combustion turbine gas supply piping is conservatively estimated to occur once the fully plastic bending moment in the pipe occurs:
d 0.75 g
.=
p 0.14
=
g pu 0.21
=
HCLPF =
0.75 e-1.65(0.14 + 0.21,) = 0.42 g No evaluation of the New Jersey Natural Gas Co. transmission system was conducted in the current program.
4.2.9 Condensate Transfer Buildina The Condensate Transfer Building consists of a reinforced concrete slab on grade with reinforced concrete piers which run downward from the slab and are supported on the circulating water intake tunnel. Above grade, the building consists of steel members with a sheet metal exterior.
Since the steel structure is relatively light, the capacity for seismic inertia loads is expected to be high. Reference 4 notes that significant liquefaction or settlements are not likely for this structure. Therefore, the seismic fragility of the Condensate Transfer building is judged to be beyond the range of interest.
During the plant walkdown, it' was noted that the Condensate Storage tank is til ely to be the most vulnerable component of this. system for seismic loads. However, the condensate storage tank is not included in the scope of this evaluation.
2Ho 1146nbloyotrche.wfw 4-40 lY d
1 4.2.10 Ventilation Stack The Ventilation Stack is a 394 ft. tall reinforced concrete chimney. The stack has a circular cross section which tapers from an outside diameter of 31'-8 3/4" at the base (top of foundation) to 9'-6" at the top. The wall thickness varies from 18" at the base to 6" at the top. The stack is founded on an octagonal spread footing which is 45' across and 7' thick. The stack is partially embedded with the top of the foundation located at elevation -3', which is 26' below the plant grade elevation.
A detailed fragility evaluation was not conducted for the stack. Since the objective of the review of the stack is to assess whether there is the potential for failure leading to collapse or partial collapse and impact with the surrounding structures, a screening approach is used to verify that the stack has a sufficiently high HCLPF capacity. This is accomplished using the seismic margins approach outlined in Reference 33.
A simple stick model was generated for the stack, ass'uming a fixed base. Seismic loading was based on response spectrum with the median NUREG/CR-OO98 (Reference 15) shape anchored to a peak ground acceleration of 0.3 g. The shear and moment capacities were found to be adequate for the resulting load demands.
Reference 4 notes that no significant liquefaction or settlements are expected beneath the stack foundation. It is also noted that the soil bearing capacity is very high such that soil bearing failures are not expected.
Therefore. the Ventilation Stack is judged to be non-controlling.
9 9
e
TABLE 4-1 REACTOR BUILDING SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Overturning moment Wall at Column Line R7 at El. 23'-6" Faetor O
h j
R U
Strength 4.96 0
0.20 Inelastic Energy Absorption 1.08 0.02 0.01 Spectral Shape 1.0 0.29 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.05 0
Total 5.36 0.29 0.25
hA3SSE = 5.36 x 0.552g = 2.96g 4 6Rp,+p,) = 1.21 g HCLPF
e
- l. ~
I r
l Nc:
2 o n 4.
,..m
.. 2
TABLE 4-2 REACTOR BUILDING SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Steel Superstructure Column Anchorage Corner Column Factor i
En Ou Strength 1.81 0
0.22 Spectral Shape 1.0 0.35 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.14 0
Total 1.81 0.38 0.27
hA3SSE
1.81 x 0.552g = 1.00g
-16Ep,+p,) = 0.34g HCLPF =
e d
l IP/F%f2 26dD 1146nb!cystreh4.wtw 4 43
'6 6
TABLE 4-3 TURBlNE BUILDING SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Diaphragm Shear South-west corner of the slab at El. 46' 6" Faetor F
hR hU Strength 1.59 0
0.21 Spectral Shape 1.0 0.19 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.05 0
Total 1.59 0.20 0.26 d=$A3SSE =
1.59 x 0.552g = 0.88g
-NO.+0) = 0.41 g HCLPF =
.e l
'(
l
}
um nu:nww.a.,
a.a; 1
L'
TABLE 4-4 l
TURBINE BUILDING SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Anchor Bolt for the Steel Column j
Column Line F at El. 46'-6*
Factor OR hu Strength 1.34 0
0.19 Spectral Shape 1.0 0.34 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.06 0
Total 1.34 0.35 0.24 d=hA3SSE =
1.34 x 0.552g = 0.74g
-16Mp,+p) = 0.28g HCLPF =
e m
8 zwo ti4wov. trow.wv.
4 45 (E@'N
.~
s
I TABLE 4-5
)
l' l
TURBINE PEDESTAL SEISMIC FRAGILITY PARAMETERS l
Critical Failure Mode:
EW Frame Plastic Hinge Mechanism EW Pedestal Frame at Column Line 3 Factor O
O R
U 1
Strength 3.58 0
0.22 1
Inelastic Energy Absorption 1.10 0.03 0.01 Spectral Shape 1.0 0.19 0
l l
Modeling 1.0 0
0.15 i
Earthquake Component Combination 1.0 0.15 0.10 Total 3.94 0.24 0.28
=hA3SSE = 3.94 x 0.552g = 2.17g HCLPF = d e-1 WO.+0.) = 0.92g 9
te
' IHO i146f%icystfCh4.Wfw 4 46
l l
l TABLE 4-6 l
lNTAKE STRUCTURE SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Diaphragm Shear East End at El. 5.5' Factor i
O O
R U
1 l
Strength 1.49 0
0.21 l
i Spectral Shape 1.0 0.17 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.05 0
Total 1.49 0.18 0.26 d=5A3SSE =
1.49 x 0.552g = 0.82g HCLPF = d e-16sp,+p) = 0.40g f56.(2
- i uw...u.4..r.
4 47 gw%
TABLE 4-7 EMERGENCY DIESEL GENERATOR BUILDING SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Out-of-Plane Bending of the NS Walls Structure failure mode conditional on no liquefaction Factor i
On On Strength 1.74 0
0.16 Inela'stic Energy Absorption 1.23 0.05 0.05 Spectral Shape 1.0 0.36 0
Modeling 1.0 0
0.15 Earthquake Component Combination 1.0 0.05 0
Total 2.14 0.37 0.22 1= $ A3SSE = 2.14 x 0.552 g = 1.18 g
-16Ep,+p) = 0.45 g HCLPF =
e Soil related failure mode conditional on the occurrence of liquefaction:
d = 0.69 g En = 0.14 pg = 0.28 HCLPF = 0.69 e-1.65(0.14 +0.28) = 0.35 g
</
e 2HD 1146reloystrch4.wfw 4-48
TABLE 4-8 FIRE POND PUMP HOUSE SEISMIC FRAGILITY PARAMETERS Critical Failure Mode:
Sliding Leading to Failure of the Fire Water Piping Factor En Ou Strength 2.19 0
0.41 Spectral Shape 1.0 0.15' O.10 Peak Horizontal / Average Horizontal 1.0 0.12 0
Mode Combination 1.0 0.05 0
Earthquake Component Combination 1.0 0.05 0
Soil Structure Interaction 1.0 0.14 0.14 Total 2.19 0.25 0.44
=hA3SSE = 2.19 x 0.552g = 1.21g HCLPF = k. e-16Ep,+p) = 0.39g
.O m
I
'2nu 114eneroyotreh4.wfw.
4 43
I TABLE 4-9 SElSMIC FRAGILITY PARAMETERS FOR OTHER STRUCTURES i
b Eu HCLPF Structure / Failure Mode and Comments (g)
(g)
Bus Duct from EDGB to Turbine Bldg
' high Adequate capacity to withstand maximum high predicted settlements Fire Pond Piping Soil deformation initiated by failure of the 1.57 0.14 0.10 1.06 sheet pile Fire Pond Dam high high Circulating Water intake and Discharge Tunnels and Discharge Structure high high Combustion Turbine Fuel Oil Tank unanchored tank controlled by shell 0.66 0.37 0.39 0.19 buckling Combustion Turbine Gas Supply Piping pipe failure induced by earthquake induced 0.75 0.14 0.21 0.42 soil deformation Condensate Transfer Building high high Ventilation Stack high high High = beyond the range of interest or non-controlling P.- b. -,
2HD 1146nbloyettch4.wfw 4 60 M
A 1
1 80 70
-cre.
n-a - e s t 60 l
u 5
p 50 e
, 40 l
.E a
&30
=
1 i
20 f
10 0
0 1
2 3
4 5
6 7
8 9
10 11 12 Time (Months)
Figure 4-1:
Effects of Time and Curing Conditions on Concrete Strength (From' Reference 20).
- wo u4ew.v.im.d*
4 61
A-12 10 OO O. Ref23 O
o nef 24 c.
O x
nef 2s
.8 O. nef 26 l
EQ 4-4 Vc 9
/
- 4. ff 6
O
's O
N '
.6 s
\\
x x
4 N
N O
s
/
\\
2 -
'N X
0 1.0 20 3.0 4,0 hw/lw Figure 4-2: Strength of Concrete Shear Walls 9
9
- 2mD ilooresovoteen" *'"
4 52 1
I L.'_._.___
I i
l l
l l
V-
/--------T--~~
u
/I sk
/I I
I v-l l
y
- g Ii I
I ll l
l l l l
1 k Il l
l l l l
l l l l
l o o.
A-O y y u
u DEFLECTION Figure 4-3:
Schematic Force - Deflection Relationship for Reinforced Concrete Bilinear Relationship and Equivalent Elastic-Perfectly Plastic Relationship.
r;marp 2HD 1146nbicystrch4,wfw 4 53 559R
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ ~
REFERENCES 1.
McGuire, R.K., et al, "Probabilistic Seismic Hazard Evaluations at Nuclear Power Plant Sites in the Central and Eastern United States: Resolution of the Charleston Earthquake issue," EPRI NP-6359-D, Project P10153, Appendix E, April,1989.
2.
Bernreuter, D.L., et al, " Seismic Hazard Characterization of 69 Nuclear Power Plant Sites East of the Rocky Mountains," Volume 2, NUREG/CR-5250 UCID 21517, January,1989.
3.
"Probabilistic Seismic Response Analyses of the Oystar Creek Nuclear Generating Station in Support of IPEEE
- EQE Report No. 50124-R-003, Prepared for GPU Nuclear, July,1994.
4.
" Assessments of Potential for Liquefaction and Permanent Ground Displacements at Designated Facilities, Oyster Creek Nuclear Generating Station, Draft Report," Geomatrix Consultants, Prepared for EOE Engineering Consultants, August,1994.
5.
Kennedy, R.P., et al., "Probabilistic Seismic Safety Study of an Existing Nuclear Power Plant," Nuclear Enaineerina and Desian, Vol. 59, No. 2, pp.
315 338, 6.
Kennedy, R.P., and M.K. Ravindra, " Seismic Fragilities For Nuclear Power Plant Risk Studies," Nuclear Enaineerina and Desian. Vol. 79, No.1, pp. 47-68.
7.
Freudenthal, A.M., J.M. Garrelts, and M. Shinozuka, "The Analysis of j
Structural Safety," Journal of the Structural Division, ASCE, ST 1, pp. 267-325, February 1966.
8.
Kennedy, R.P., A Statistical Analysis of the Shear Stranath of Reinforced Concrete Beams, Technical Report No. 78, Department of Civil Engineering, Stanford University, Stanford, California, April 1967.
2His i146 e/oyettent.wtw b1 l
l 4
L
9.
Newmark., N.M., "A Study of Vertical and Horizontal Earthquake Spectra,"
WASH 1255. Nathan M. Newmark Consulting Engineering Services, prepared for USAEC, April 1973.
I 10.
Kennedy, R.P., D.A. Wesley, and W.H. Tong, "Probabilistic Evaluation of the Diablo Canyon Turbine Building Seismic Capacity Using Nonlinear Time History Analysis," NTS Engineering Report 1643.01, Prepared for PG&E, June 1988.
11.
ACI Committee 318, "Suilding Code Requirements for Reinforced Concrete (ACl 318 63)," American Concrete Institute, Detroit, Michigan (1963).
12.
Newmark, N.M., "!aelastic Design of Nuclear Reactor Structures and its implications on Design of Critical Equipment," SMiRT Paner K4/1.1977 SMiRT Conference, San Francisco, California.
13.
Riddell, R., and N.M. Newmark, " Statistical Analysis of the Response of Nonlinear Systems Subjected to Earthquakes," Department of Civil Engineering, Reoort UILU 7g 2011, Urbana, Illinois, August 1979.
14.
Kennedy, R.P., et al., " Engineering Characterization of Ground Motion," Vol.
1 and 2, NUREG/CR 3805, May 1984, March 1985.
15.
Newmark, N.M., and W.J. Hall, " Development of Criteria for Seismic Review of Selected Nuclear Power Plants," NUREG/CR-OO98, May 1978.
l 16.
Hashimoto, P.S., et al., " Review of Structure Damping Values for Elastic Seismic Analysis of Nuclear Power Plants," NUREG/CR 6011, March 1993.
i 17.
USNRC, " Combining Modal Responses and Spatial Components in Seismic l
Response Analysis," bSNRC Reaulatory Guide 1.92. Rev.1, February 1976.
18.
Veietsos, A.S. and Y. Tang, "The Effects of Soil Structure Interaction on Laterally Excited Liquid Storage Tanks," EPRI NP 6500 Research Project 2907 2, September 19'89.
19.
Veletsos, A.S. and Y. Tang, " Interaction Effects in Verticclly Excited Steel Tanks," Dynamic Resnonne of Structures. Americ6n Society of Civil Engineers,1986, pp. 646-643.
TP.AJP 2Ho 114snWeystrohs.wfw -
52 5Dd
- 20.. Troxell, G.E., and H.E. Davis and J.W. Kelly, Composition and Properties of Concrete, McGraw Hill,1968.
21.
Mirza, S.A., M. Hatzinikolas, and J.G. MacGregor, " Variability of Mechanical' Properties of Reinforcing Bars," Journal of Structural Division, ASCE, May 1979.
22.
ACI 318-83, " Building Code Requirements for Reinforced Concrete,"
American Concrete institute,1983.
23.
Barda, F., J.M. Hanson, and W.G. Corley, " Shear Strength of Low Rise Walls With Boundary Elements," ACI Symposium, " Reinforced Concrete Structures in Seismic Zones," ACl, Detroit, Michigan,1976.
24.
Shiga, T., A. Shibata, and J. Takahashi, " Experimental Study on Dynamic Properties of Reinforced Concrete Shear Walls," 5th World Conference on Earthquake Engineering, Rome, Italy,1973.
25.
Cardenas, A.E., et al., " Design Provisions for Shear Walls," ACI Journal, Vol.
70, No. 3, March 1973.
26.
Oesterle, R.G., et al., " Earthquake Resistant Structural Walis - Tests of isolated Walls - Phase ll" Construction Techrology Laboratories (Division of PCA), Skokie, Illinois, October,1979.
I 27.
Park, R., and T. Paulay, Reinforced Concrete Structures, John Wiley, Inc.,
1975.
28.
Paulay, T. and M.J.N. Priestly, Seismic Desian of Reinforced Concrete and Masonry Spildinas. John Wiley, Inc.1992.
29.
Psed, J.W. and R.P. Kennedy, " Methodology for Developing Seismic Fragilities,' EPRI Research Project RP2722-23, Draft, August,1993.
30.
Newmark, N.M., " Effects of Earthquakes on Dams and Embankments,"
Geotechnious, Vol. XV, No. 2,1965.
l l
l zHD 1146twovetrch5.wfw 53
3 31.
Kausel, E.A., et al., " Seismically Induced Sliding of Massive Structures,
Journal of the Geotechnical Division. ASCE, Vol.105, No. GT12, pp.1471-1485, December,1979.
32.
Danay, A. and L.N. Adeghe, " Seismic induced Slip of Concrete Gravity Dams," Journal of Structural Enaineerina, ASCE, Vol.119, No.1, January, 1993, pp.108129.
33.
"A Methodology for Assessment of Nuclear Power Plant Seismic Margin (Revision 1)." EPRI NP 6041-SL, Rev.1, August,1991.
34.
Bandyopadhyay, K., et al, " Seismic Design and Evaluation Guidelines for Department of Energy High-Level Waste Storage Tanks and Appurtenances," Prepared for the Department of Energy by the DOE Tanks Seismic Experts Panel and Brookhaven National Laboratory, BNL 52361, January,1993.
?
9
~
7Wn 11dAnbicygtrehEMw 6.(
- - - -