ML19309H086
| ML19309H086 | |
| Person / Time | |
|---|---|
| Issue date: | 04/29/1980 |
| From: | Levin H Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| NUDOCS 8005080341 | |
| Download: ML19309H086 (8) | |
Text
e 8 0 0 5 0 8 OM/
NRC SYSTEMATIC EVALUATION PROGRAM SEISMIC REVIEW HOWARD A. LEVIN U. S. NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C.
20555 ABSTRACT The NRC Systematic Evaluation Program is currently making an assessment of the seismic design safety of 11 older nuclear power plant facilities. The general review philosophy and review criteria relative to seismic input, structural response, and equipment functionability are presented, including the rationale for the development of these guidelines considering the signif-icant evolution of seismic design criteria since these plants were ariginally licensed.
Technical approaches thought more realistic in light or current knowledge are utilized.
Initial findings for plants designed to early seismic design procedures suggest that with minor exceptions, these plants possess adequate seismic design margins when evaluated against the " intent" of current criteria. However, seismic qualification of electrical equipment has been identified as a subject which requires more in-depth evaluation. Plants originally designed to local building codes without explicit seismic design consideration may require substantial retrofitting to achieve acceptable margins.
INTRODUCTION In October 1977, the Nuclear Regulatory Commission approved initiation of Phase II of the Systematic Evaluation Program (SEP) which c<nsists of a plant-specific reassessment of the safety of 11 older operating nuclear reactors.
Many safety criteria have rapidly evolved since the time of initial licensing of these plants.
The purpose of the SEP is to develop a current documented basis for the safety of these older facilities by comparing them to current criteria.
Phase I of the SEP developed a comprehensive list of 137 topics of safety significance which collec +.ively affect the plant's capability to respond to various Design Basis Ev<.nts (DSEs).
This paper summarizes aspects of the ongoing SEP seismic DIE ev.iluations and highlights initial findings of the seismic review.
The nuclear power plant facilities under raview in the SEP received construction permits between 1956 and 1967.
Seismic design procedures evolved significantly during and after this period and through the publication of the Standard Review Plan (SRP) 1975) which represents current analytical and design review criteria along with the Regulations 10 CFR 50, Appendix A, and 10 CFR 100, Appendix A.
As a result, the original seismic design bases of the SEP facilities vary in degree.from Uniform Building Code considerations (static analysis) up through and approaching current standards (dynamic analysis).
Recognizing this evolution, the NRC has found it necessary to make an assessment of the seismic design safety of the SEP facilities.
GENERAL PHILOSOPHY The primary objective of the SEP seismic review program is to make an overall seismic safety assessment and where necessary, to improve seismic satety margins, consider backfitting in accordance with 10 CFR 50.109 of the Regulations.
This Regulation specifies that backfitting will be required only if substantial, additional protection to the public health and safety is required.
To assist in making these decisions, current licensing criteria, as defined by the Standard Review Plan (SRP), are utilized as a baseline to qualitatively assess safety margins.
Compliance with all sections of the SRP would certainly imply acceptability; however, this cannot generally be expected in view of the fact that the SEP plants were originally designed to different criteria.
Furthermore, demonstrating compliance with specific individual criteria in the SRP is not necessarily an indication of acceptability when considered outside the context of the SRP " package," since individual criteria may not generally control broad safety issues.
The important SEP review concept is to make a determination whether or not the plant meets the " intent" of current criteria considered with respect to the general level of safety these criteria dictate as a " package."
For example, current criteria strive to ensure that elements of nuclear power plants remain elastic or nearly elastic in order to assure that they meet their functional requirements. The intent of this goal is important and is maintained in the SEP evaluations.
However, seismic resistance does not imply a total absence of permanent deformation.
Certain structures, piping, and equipment may suffer damage provided that the entire system can achieve and maintain a safe shutdown condition.
Therefore, the SEP evaluations require an assessment of broad safety issues considering the various systems interactions in the context of overall plant safety.
Potential accident sequences considering the behavior of the total nuclear power system are evaluated by SEP systems engineerr. For. example, one major area of review is that associated with possible seismic-induced pipe breaks. The review addresses the subsequent ~
impact on other systems and components, assuming single failures, as well as the possible loss of offsite power with its implications on system safety.
The SEP seismic review process recognizes and-attempts to deal with the inherent and often unquantifiable capability of these facilities to resist seismic forces and the conservatisms associated with current evaluation methods.
The SEP evaluations attempt to more realistically quantify unclaimed factors which contribute to seismic resistance capability.
The current evaluation techniques are sometimes overly conservative because certain energy dissipating mechanisms are not quantitatively considered.
Nonlinearities below the threshold of overall elastic response are found to dissipate significant energy and in turn reduce design force levels.
Energy absorption below elastic limits is represented through structural damping. Parameters such as damping values have been chosen conservatively because of large uncertainty in their selection.
This is justified when the uncertainty is characteristic of the nature of a particular process; however, often uncertainty results from a 1 ck of our 4
1
knowledge which needs to be refined.
Over the years, more structural damping data has become available.1 Accordingly,.this data is considered in the SEP evaluations.
The SEP review has utilized structural damping values thought more realistic in light of current knowledge to more accurately reflect the true seismic response.
The SEP seismic review focuses on an assessment of the integrity of the reactor coolant pressure boundary and the capability of essential systems and components required to safely shut down the reactor and maintain it in a safe shutdown condition during and after a seismic event or to mitigate the consequences of such an event. Therefore, emphasis is not given to other components and systems, such as the radwaste systems, which ordinarily fall under the scope of Regulatory Guide 1.29, " Seismic Design Classification." This approach was taken to concentrate efforts on those systems which are most important to safety and minimize the impact on review resources.
In most cases, the SEP evaluations should be sufficient to infer the capability of other systems originally classed for seismic consideration such as the radwaste systems because these systems are similarly designed.
The Safe Shutdown Earthquake (SSE) is the only earthquake level evaluated because it represents the most severe safety significant design condition to which the plant may be expected to respond.
Present licensing criteria some-times result in the Operating Basis Earthquake f0BE) (usually 1/2 SSE) controlling the design of various structures, systems, and components because of specified load combinations and lower damping and allowable streas levels.
Relief is usually granted when these criteria do not imply real safety issues but rather have operational implications.
Since a facility designed t, shut down safely following an SSE will obviously be safe for a lesser earthquake, it was deemed unnecessary to investigate further the effects of the OBE.
Furthermore, NRC staff requirements for an inspection to evaluate any damage to the plant will follow the occ 2rrence of any major earthquake.
OVERVIEW OF REVIEW PROCEDURES The 11 SEP plants have been categorized into 2 groups based upon the degree seismic design was originally considered and the quantity of available seimsic design documentation. Two different procedures are in use to review the plants in each respective group.
The five later plants, Oyster Creek, Dresden 2, Ginna, Millstone 1, and Palisades, are categorized under Group 1 while the six earlier plants, Dresden 1, Yankee Rowe, Big Rock Point, Lacrosse, Haddam Neck, and San Onofre 1, are categorized under Group 2.
I The general review procedures employed are summarized as follows:
Group 1 - Detailed NRC staff / consultant review of existing seismic design documents with supplemental review team studies for spot checking of licensee information and confirma-tion of review team judgments.
l
Group 2 - Detailed review of new licensee seismic design evaluations with limited supplemental review team studies for spot checking.
These procedures were developed to optimize the allocation of combined NRC staff and licensee resources and to complete the seismic review within the SEP time frame.
The SEP seismic review was first initiated by conducting a detailed review of the respective plant docket files to provide a core of pertinent informa-tion to be used in the review.
From the Group 1 plant docket searches, it was concluded that the existing documentation including supplemental informa-tion available at the offices of the architect-engineer (AE) and nuclear steam
~
supply system (NSSS) designer would for the most part be sufficient to permit the necessary SEP evaluations. However, due to a lack of available information from all sources for the older Group 2 plants, it was determined that new information should be generated. Therefore, each Group 2 plant licensee was requested by the NRC to initiate a seismic evaluation program to document a more current seismic design basis.
Similarly, in limited situations where existing information is not sufficient to adequately document the seismic safety of the newer Group 1 plants, these licensees have been requested to generate new supporting documentation.
The NRC staff review of all of the Group 1 facilities should be complete by the end of 1980.
Dresden 2 is complete, while Ginna is nearing completion.
Oyster Creek, Palisades, and Millstone 1 are in intermediate stages of review.
The Group 1 reviews necessarily require an evaluation of critical structures to assess the seismic input to equipment.
However, the seismic stresses for the structures are not evaluated at every location.
Spot checks are made at critical locations, but the evaluations generally end at the force level by way of a comparison to the original design force levels.
Stress evaluations in addition to the-spot checks are made only when the newer force levels are predicted to be higher than the original' design values.
A detailed evaluation of the hundreds of individual components within the Group 1 plants is not made. The evaluations rely upon sampling representative components from generic groups of equipment.
This sample is augmented subject to walk-through inspections of the facilities to select additional components based upon a higher potential degree of seismic fragility.
These procedures rely in large part upon the expertise of the review team.
Accordingly, a team of recog-nized seismic design experts under the direction of Dr. N. M. Newmark has been organized by the NRC to assist in these reviews.
To date the licensees' Group 2 seismic evaluation programs have concentrated upon the development of seismological, structural, and mechanical acceptance criteria and evaluation procedures.
The NRC staff is working with these licensees to develop criteria which are consistent with SEP objectives.
k'orking analytical models are in development and detailed evaluations are beginning.
In view of the limited existing seismic design data base for the Group 2 plants, it is expected that the scope of review will be more encompassing.
DETERM1SATION OF THE SEISMIC HAZARD The SEP review includes an intensive evaluation of the seismic hazard at each site.
Various deterministic and probabilistic techniques are under considera-tion to produce site-specific ground response spectra.
This information will be used to evaluate the adequacy of the seismic input originally specified for detign and newer ground response spectra recently proposed by the SEP licensets for use in the SEP evaluations.
The currert methodology for determination of design bases for vibratory ground motion as defined by 10 CFR 100, Appendix A, of the Regulations and the SRP ut:lize the generic Regulatory Guide 1.60 ground response spectra anchored at a zero period (peak) ground acceleration of the Safe Shutdown Earthquake (SSE). The peak ground acceleration is established based upon required investigations of the local and regional geology, seismic history, and engineering characteristics of the site. These criteria are deterministic and necessarily require bounding interpretations of important parameters.
Generally, this procedure is believed to be adequately conservative; however, it has a tendency to produce variable estimates of the seismic hazard from site to site. The penalties associated with being overly conservative are particularly significant when reevaluating older facilities.
Therefore, i
recognizing the inadequacies with the current approach, the NRC has contracued with the Lawrence Livermore Laboratory to develop various alternative seismic hazard methodologies to explore improved approaches.2 Probabilistic seismic risk analysis methodologies provide a unique tool for decision making which can be used for:
1.
explicitly keeping track of important parameters and the sensitivity of their variance; 2.
quantifying predicted risk in a consistent and uniform manner; 3.
estimating relative risk from one seismic hazard level or design level to another, and 4.
evaluating in a limited way the required level of seismic resistance capability to meet overall risk goals.
Although these methods are controversial particularly in estimating absolute levels of risk as implied in item 4, regulatory experience suggests their usefulness for the quantitative comparison of relative risk as suggested in item 3.
If properly used,.this information is valuable in considering whether to backfit a facility to increase seismic safety margins.
The following four approaches have been suggested by LLL for evaluation of all sites except San Onofre 1:
1.
Direct statistical evaluation of response spectra from appropriate groups of real earthquakes; 2.
Scaling of real spectra to peak acceleration values determined for various risk levels;
3.
The Newmark-Hall technique of scaling response spectra to peak accelerations and velocities determined for various risk levels; and 4
Uniform risk technique-scaling spectral ordinates as a function of return period.
Figures I through 4 provide examples of output from these methods.
Figure 5 shows a comparison of the four methods.
An important revieu objective is to assess whether these various methods yield common spectral predictions in the engineering frequency range of interest.
The NRC will incorporate the LLL information, data submitted by the SEP licensees, predictions by alternative methods, and other pertinent data to reach a decision relative to an appropriate seismic input specification.
The San Onofre 1 licensee has proposed a site-specific spectra developed using ground motion modeling techniques from first principles.
BASES FOR REEVALUATION The specific SEP reevaluation criteria are documented in NUREG/CR-0098.3 This document addresses:
1.
Selection of the earthquake hazard; 2.
Design seismic loadings; 3.
Soil-structure interaction; 4.
Damping and energy absorption; 5.
Methods of dynamic analysis; 6.
Review analysis and design procedu'res; 7.
Special topics such as underground piping, tanks and vaults, equipment qualification, etc.
Limitations of space make it impossible to summar'ze all of these issues.
The reevaluation criteria consider the implications of various levels of damage, short of collapse for critical structures and the failure to function for safety-related equipment.
Some active elements of nuclear power plants such as pumps, valves, switch gear, motors and motor control centers must remain elastic or nearly elastic to perform their safety function.
This objective is maintained within the SEP; however, it is recognized that pure linear elastic analysis, even up to yield stress levels, sometimes overestimate design loadings because certain nonlinearities are not rigorously considered.
For active components and other deformation limited items, increased damping levels are considered to more realistically account for energy absorption in the context of overall linear response. The following_ table summarizes the damping values recommended by NUREG/CR-0098 for the SEP and those specified in Regulatory Guide 1.61.
DAMPING (% OF CRITICAL)
SEP Recommended R.G. 1.61 (SSE)
(near yield level)
Reinforced concrete 7
7 to 10 Prestressed concrete 5
5 to 7 Welded assemblies 4
5 to 7 Bolted and riveted assemblies 7
'10 to 15 Piping 2 or 3 2 to 3 101 Cable' trays Equipment 71
s 3
1
. i i. o;
...n, i i......
u ni
. i s
3 C
E f.
3
Mean o to
=
=
0.1 r 4
0.1 y 7
i O
8 1
1 Mean i
0.0001 7
g g
g; y
y 0.001 2
0.01 j 0.01 {
j 0.01,=.
3
.d S
U 5
n t
5% damping -
5% damping
~
i 0001 O.001 O.1 1
10 100 0.1 1
10 100 F requency-Hz Frequency-Hz FIG.1 Real spectra FIG. 2 Real spectra scaled to peak accelerations 1::
1; s ' a e "t i
>>siii*
i i
++iti1 a
i s i i ng iii..:
1
.. i i i s ui e
~
0.0001 3
~
~
0.001 i
0.1 e
0.1 "
0 01 E
.=
. 0.0001
.E.
t 0.001
.t.,
3u o
g 0.01 g
3
_2 0.01
~~
[ 0.01 W
a n
t 5% damping.
5% damping l
0.001 O.001 O.1 1
10 100 0.1 1
10 100 Frequency-Hz F requency-Hz FIG. 3 Newmark-Hall spectra scaled to FIG. 4 Uniform hazard spectra peak accelerations
't "4
3 4
.r
1
", ' %.s?
.; i
'?
.e s.w _._._ 1'
'/
t
'/
4
.ci-
'f i
/
i
/
4 l
k e
l
/ /
i
~
I
/
l i
a
/
]
3...
I4 P
t
-..s:
- om,
~,*ss 3
y
.*.- ami.
e o l
l w s. 8 g 4
- O 001 ee.e l
s a
e mi ei i
- o
,ao FIG. 5 ' Comparison of scenra from FIGS.14
For structures or ppssive components such as pressure vessels, tanks, piping, transformers and heat exchangers, for which their safety-functions are to main-tain leak-tight or structural integrity, energy absorption in the inelastic range is considered in the SEP through use of the " ductility factor." The ductility factor is the ratio of the maximum useful (or design) displacement of a structure to the " effective" elastic limit di.tplacement.3 For vital items that must remain functional and nearly elastic, ductility levels of 1.0 to 1.3 are recommended in NUREG/CR-0098.
Ductility levels of 3 to 8 can be justified to account for inelastic energy absorption when deformation is tolerable from a safety point of view.
It is noted that care must be taken to ensure that local ductilities are maintained at acceptable levels when addressing ductility at a system level.
PRELIMINARY CONCLUSIONS i
Structural - Initial review of the later Group 1 plants suggests that these facilities possess adequate structural margins. On a plant-specific basis, some of the earlier Group 2 facilities may require some retrofitting to attain acceptable structural qargins.
Mechanical and Electrical - Equipment functional qualification has been identified as any area that requires additional documentation.
In certain cases for the Group 1 plants and in more cases for the Group 2 plants, equipment modifications will be necessary.
The issue of anchorage and support of equipment and, in particular, electrical equipment has been identified as an area that requires upgrading.
The SEP licensees are addressing this issue through individual inspection and evaluation programs.
It would appear that piping and mechanical components are adequately supported for the Group 1 plants; however, the Group 2 plants may require substantial upgrading.
REFERENCES 1.
J. D. Stevenson, " Structural Damping Values as a Function of Dynamic Response Stress and Deformation Levels," paper K11/1, 5th International Conference on Structural Mechanics in Reactor Technology (August 1979).
2.
D. L. Bernreuter, " Methods to Develop Site Specific Spectra and a Review of the Important Parameters that Influence the Spectra,"
UCRL-52458, Lawrence Livermore Laboratory (May 1979).
3.
N. M. Newmark, W. J. Hall, " Development of Criteria for Seismic Review of Selected Nuclear Power Plants,"-NUREG/CR-0098 (May 1978).
.