ML20136E929
Text
, _
d, acg'o ( b UNITED STATES
!Y )sy b NUCLEAR REGULATORY COMMISSION E WASHINGTON. D. C. 20555 g v ,/ APR 121984 "U
12 F;,1 3 51 y .. m . . n.
MEMORANDUM FOR: Jesse C. Ebersole, Chairman W M: 9W AdvisoryCommitteeonReactorSafeha'rds'^
FROM: William J. Dircks Executive Director for Operations
SUBJECT:
QUANTIFICATION OF SEISMIC DESIGN MARGINS This is in response to your letter of January 18, 1984, regarding ACRS concerns related to seismic design margins. Our progress on developing a program to respond to this issue has been somewhat slower than we had anticipated due to a variety of high priority items (e.g., the Diablo Canyon Independent Design Verfication Program and allegation assessment) which reduced NRR staff resources dedicated to the seismic design margins issue. However, progress is being made and the following paragraphs set forth our approach to address this issue in terms of the safety issue, regulatory needs, research activity, coordination and planning, and organizational interactions.
Safety Issue Based on the general understanding of the inherent seismic capacity of nuclear power plants that has been obtained through the seismic safety margins programs which have been conducted to date, we believe that this issue can be resolved in the longer term and does not constitute an immediate safety issue. The following examples substantiate this con-clusion.
- 1. Seismic probabilistic ri_k analyses published to date, including the Zion nuclear plant seismic risk analysis performed under the NRC sponsored Seismic Safety Margins Research Program (SSMRP), estimate that most of the contribution to seismically induced core melt frequencies comes from earthquakes in the peak ground acceleration range from 2 to 4 times the safe shutdown earthquake (SSE) acccleration. This indicates a significant amount of overall plant safety margin above seismic design levels. These seismic risk studies also indicate there are large margins as measured by the median value of failure at the component level. For eastern U.S.
plants, the median capacities to failure of typical structures and components are generally estimated to be greater than 4 times the SSE level. Passive components such as piping and structures tend to have slightly higher median capacities and active equipment such as pumps and valves tend to have slightly less. Studies such as the SSMRP "Best Estimate versus Evaluation Method (BE-EM) Study" have quantified large margins resulting from the conservative-methods used to predict structural and component response.-
,M [
q+r f M O P ' " T @ 4.. 0 y j
. . . . , c. u y 9=
Om E 5 ,. ! I),..mnnon f...~ It O N O 6h ( f' ,
(.
i Jesse C. Ebersole 2 l
- 2. Scale model tests of prototype Seismic Category I structures mounted on vibration tables have indicated that initial yielding occurs at about the_SSE level with significant structural-damage occurring at levels in excess of 3 times the SSE. At earthquakes greater than 4 times SSE, vertical displacement (or rocking) resulting in basemat uplift may become a concern.
- 3. Tests performed on ASME Code designed piping-have indicated that such piping can sustain earthquakes several times higher than the SSE while at the'same time being subjected to approximately twice the design pressure. No apparent damage or leakage resulted as.a consequence of these tests. Moreover, probabilistic calculations conducted by Lawrence Livermore National Laboratory have demon-strated for PWR primary piping that the likelihood of a pipe rupture t
or pipe5leakage mately times SSE induced by a range is extremely smallof earthquakes (less than 10 ug6)o approxi-4.- Experience from comparable heavy industrial structures and components that have been subjected to large earthquakes have shown that structures, piping and equipment, designed with far less care than nuclear, exhibit capacities far beyond their design levels provided they are adequately supported.
Regulatory Needs As indicated in our April 4, 1983 let'ter to Commissioner Ahearne, we have viewed ACRS' concern as a call for consolidation of seismic review efforts into a systematic appraisal of seismic risk. From a licensing perspec-tive, there is a continuing need for consideration of the inherent quantitative seismic margins in nuclear power plants. This need, for
. example, is often the result of changing perceptions of the seismic hazard of an area based on the acquisition of new knowledge or the nearby occurrence of earthquakes. In addition, our continuing assessment of facilities in high population areas have utilized probabilistic methods to assist in decisionmaking and the assessment of older plants..as accomplished in the Systematic Evaluation Program (SEP). Probabilistic methods will also be used in the proposed Integrated Safety Assessment Program (ISAP). All of these assessments require that judgments be made
- with regard to the seismic adequacy of these facilities. Such assessments will continue to be required as a result of the increasing number of 1- operating plants and as knowledge of the seismic hazard changes. A sound,
- practical seismic margins program utilizing margins to failure analysis j and state-of-art seismic probabilistic risk assessment techniques would
. serve to minimize the need for changing regulatory requirements and licensing actions. In addition, it would provide a sound basis for i
confidence in the seismic capacity of nuclear power plants and, when i necessary, it would serve to better identify where resources can best be i expended to provide the ' greatest overal_1 reduction in seismic risk.
However, in order for such a program to be fruitful, continued development of licensing tools are needed such as validated calculation methods and a broad experimental fragility data base.
. - - , . ~ , - - _ _ . . . - . _ _ - . _ %_m.,_.. . . , , .,._,1_.m.___..~.__. - _ - . . - _ _ _ _ _ . _ , . . . -
. .' I (
Jesse C. Ebersole 3 Research Activity The seismic design margins research effort can be described in terms of short term, high priority research required to support immediate regulatory needs and long term, lower priority research required to ulti-mately resolve the question of seismic design margins through the use of acceptable simplified design approaches.
Within the short term (i.e., the next 3 to 4 years) the research effort will develop validated calculation methods, improve the fragility data base, and develop probabilistic input data (hazard curves) needed to perform seismic margins studies being planned. The results of this re-search will provide the technical basis for regulatory decisions regard-ing the significance of design margins in terms of overall safety and identifying potential weaknesses that might have to be addressed to achieve.the desired level of safety.
The following on-going research programs will support the short term seismic design margins effort being planned and will provide input to the longer term research program:
- 1. Validation of seismic risk calculation methods and modeling techniques. Specific validation requirements and experimental requirements are being identified. The experimental data will be obtained from various test programs including NRC participation in the Heissdampfreactor (HDR) vibration tests in the Federal Republic of Germany. In addition, negotiations are proceeding with the Japanese Ministry of International Trade and Irdustry to cooperate with us in the area of seismic effects. A meeting with MITI will be
-held in July to identify specific programs for cooperative exchange.
Other foreign and domestic test programs will be used to the extent possible to validate the seismic calculation methods. (Completed in 1986)
- 2. Development of comprehensive seismic risk methods under the Seismic Safety Margins Research Program (SSMRP) and a detailed seismic risk analysis of the Zion nuclear power plant including sensitivity studies with current (not fully validated) codes and models.
(Completed in 1984)
- 3. Development of simplified seismic risk methods derived ~from the comprehensive methods developed in (1) above.
(Completed in 1985)
- 4. An overall risk analysis of the LaSalle BWR nuclear power plant under the Risk Methodology Integration and Evaluation Program (RMIEP). The seismic risk analysis will be performed under the SSMRP BWR simplified methods project. (Completed in 1985)
- 5. Research projects in the areas of geology, seismicity and geophysics to gather ground motion data and develop seismic hazard curves for the eastern U.S. (Completed in 1985)
c . '[ ( (
2
~ Jesse C. Ebersole 4 6.. An evaluation and review of the risk analysis (including seismic) of the Kuosheng nuclear power plant in Taiwan. (Completed in 1985)
- 7. Investigation of the inelastic seismic behavior of shear wall structures using reinforced concrete scale models to assess how the '
parameters (e.g., damping, frequency, floor response spectra) used in the design of safety related equipment and structures are affected by earthquake loads'above initial design levels. (Completed in 1988)
- 8. _ Piping research programs to investigate the behavior (respnse and damping values) and capacity (fragility) of nuclear plant piping systems. These pro (Completed in 1985) grams include laboratory and in situ testing.
A' longer term research program to reduce the uncertainties in the quantification of seismic design margins to be applied to simplified design approaches is currently being planned. This long term research program will interact with and build upon the research results cited above and is being coordinated with EPRI and AIF. A panel of expert con-sultants will be used to assist in the planning effort which will be com-pleted in FY'1984. Meaningful results from this research program can be achieved only if validated methods and adequate fragility data bases are established. RES has budgeted $10.4 million for the FY 1985 seismic research effort' (most of which supports seismic margins) with $1.4 million targeted to address specific seismic design margins problems such as fragilities.
Coordination and Planning An in-house Seismic Design Margins Working Group has been formed to focus the supporting elements into a coherent and responsive plan that will recommend appropriate licensing action to the Director of NRR and short term research to support the. licensing effort as well as long term research to the Director of RES. The first effort of the group has been to develop a more focused understanding of the' seismic margins issue, including the ACRS concerns as well as exploring methods that could be employed to address this complex problem. An extensive interaction took i- place during a workshop of the ACRS Extreme External Phenomena l
Subcommittee meeting on December 8 and 9,1983. Although the working group has not issued any reports of its activities, it is proceeding with L a program to recommend a course' of action to address the seismic design margins issue. This group also believes it is very important for the l- responsible staff to assess within the next year what we learned in the
- current seismic PRA' effort and the SSMRP program before proceeding with l any significant new programs W her than those now planned.
i The Seismic Design Margins Working Group, which is expected to continue-l through 1989, includes representatives from the relevant divisions in NRR
- and RES who will be working under the line management of those offices.
i
(-
b i.
. .l . (
Jesse C. Ebersole 5 This group will also interact with the panel of expert consultants, established under the seismic design margins research program being planned, who will assist in developing a plan that will be used for further discussions with the Staff and the ACRS.
We concur in the ACRS'~ call for a consolidation of seismic review efforts (past, present and proposed) into a systematic appraisal of seismic risk.
Such an effort is necessarily complex and will have to take place over the next several years and is heavily dependent on the assignment of priority staff resources and industry participation. Specific resource needs will be identified by the working group. We do not believe the formation of a seismic program office nor the designation of a single individual to be responsible for NRC's total seismic effort is necessary or desirable at this time. The diverse responsibilities involving plant licensing and research programs can be adequately coordinated through upper management interactions between NRR and RES Office Directors and appropriate Division Directors. The ad hoc Seismic Design Margins Working Group will have the responsibility to formulate and recommend a plan of action and monitor the subsequent activity. The group will be jointly chaired by James E.
Richardson, Acting Chief, Mecher.ical/ Structural Engineering Branch, RES, and Robert E. Jackson, Chief, Geosciences Branch, NRR, and will report through the existing NRR and RES management structures. Programs recom-mended by the working group that have licensing implications will be referred to NRR for appropriate action on a case by case or generic basis.
Recommended long range research programs to validate methods and quantify margins will be referred to RES for incorporation into the Long Range Re-search Plan and will be carried out within the existing RES organization.
This effort will require a continuing dialogue with the ACRS to further clarify its concerns and assist us in providing direction to the program.
Interactions I believe the key to addressing the ACRS concerns is more frequent and
! complete communications together with getting all appropriate parties l
involved in the process from the very beginning of the program planning.
l This includes not only the NRC offices but the ACRS, EPRI and AIF.
The Seismic Design Margins Working Group will be responsible for assuring that those communications and interactio'ns are complete and timely.
i Messrs Richardson and Jackson will contact the ACRS staff to arrange for a meeting with the appropriate ACRS Suhcommittee in the near future.
N William J. Dircks Executive Director for Operations
. - _ . .- -