Overview of Seismic and Volcanic Hazard Evaluation Methods for Nuclear Power Plants. Aybars Gürpinar, IAEA Consultant

Transcription

1 Overview of Seismic and Volcanic Hazard Evaluation Methods for Nuclear Power Plants REGIONAL WORKSHOP HANOI, VIETNAM, 7 11 JUNE 2010 Aybars Gürpinar, IAEA Consultant

2 Background Information The following few slides will provide background information regarding the preparation of the new revision of the Safety Guide on Seismic Hazards in Site Evaluation of Nuclear Facilities 2

3 IAEA SAFETY STANDARDS ON SITE EVALUATION Safety Standards Series hierarchy SITE EVALUATION Safety Fundamentals Requirements Safety Guides 8/25/ International Atomic Energy Agency REQUIREMENTS SAFETY GUIDES GUIDES SG-S9 to be revised DS

4 IAEA SAFETY STANDARDS - SEISMIC SAFETY SITE EVALUATION DESIGN new installations OPERATION operating/existing installations Evaluation of Seismic Safety - Existing NI NS-G-2.13 REVISION DS422 Seismic Hazard NS-G-3.3 Seismic Design and Qualification NS-G-1.6 Periodic Safety Review The complete lifetime of the installation (t) 4

5 Generation I 50-SG-S1 (1979) Distinction between low and high seismicity countries (the Guide was valid for high seismicity countries) Confusion between probabilistic and statistical approaches Collection of varied and sometimes inconsistent national approaches Recommendation for generic response spectra (USNRC RG 1.60) 5

6 Generation II 50-SG-S1 (Rev. 1, 1991) Seismotectonic modelling using a four-scale approach; regional, near regional, site vicinity, site area Applicable to all countries (no distinction between low/high seismicity) Seismogenic structures and zones of diffuse seismicity Deterministic with an option for probabilistic Minimum requirement for 0.1g design Clear definition of a capable fault Site specific response spectra 6

7 Generation III NS-G-3.3 (2002) More emphasis on uncertainties More guidance on new topics of data generation such as paleoseismology More guidance on probabilistic seismic hazard analysis Decoupling of design response spectra and the hazard based response spectra (site specific) 7

8 Why we need a revision now? Feedback from seismic safety reviews since 2002 (about 20 missions) some involving PSHA Need to include other nuclear installations International experience on PSHA such as Pegasos Recent strong motion recordings in California and especially Japan Exceedance of hazard in Japan (Onagawa and K-K) Preparation for the new build 8

9 Why we need a revision now? Need of a better treatment of uncertainties in both deterministic and probabilistic analyses. Distinction between uncertainties that can be reduced through site specific investigations and those that are imported. Some recent PSHA studies have used approaches with significant human and financial resources. This is not always possible and alternative methodologies are needed to properly account for uncertainties. More attention is needed on organizational and management aspects. 9

10 Why we need a revision now? Evaluation of the potential for fault displacement in the site area or vicinity for existing nuclear installations using a probabilistic approach. This Safety Guide is included in the longterm structure of safety standards. 10

11 Contents of the Presentation Objectives of a SHA Building up of a Database Dealing with Uncertainties DSHA vs PSHA Using PSHA in S-PSA Using PSHA in Design Site Specific vs Regional PSHA Checking the results Use of CAV as filter in SHA Probabilistic Fault Displacement Analysis Conclusions 11

12 Objectives of a SHA Derivation of the Design Basis Ground Motion Values for New NPPs Considerations for fault displaement hazards Seismic Evaluation for NPPs Seismic PSA Seismic Margin Analysis Considerations for Nuclear Installations or Facilities other than NPPs (including Waste Repositories) This presentation will only consider NPPs 12

13 DS422 OBJECTIVES To provide guidance on evaluating seismic hazards at a nuclear installation site and, in particular, on how to determine: (a) the vibratory ground motion hazards in order to establish the design basis ground motions and other relevant parameters for both new and existing nuclear installations, and (b) the potential for fault displacement and the rate of fault displacement that could affect the feasibility of the site or safe operation of the installation at that site. It is intended for use by regulatory bodies and for operating organizations. 13

14 Building up of a Database Introduction of 4 scales of investigation Regional (R~150 km) 1 : Near regional (R~25 km) 1 : Site vicinity (R~5 km) 1 : 5000 Site area (fenced area) 1: 500 This recommendation has been in the IAEA SG since The USNRC RG has similar requirements (1996) 14

15 Seismic Hazard Evaluation Scales of investigations Geological, geophysical and geotechnical databases Site vicinity Objectives: Neotectonic fault history Potential for surface faulting Near regional scale Objectives: Detailed seismotectonic characterization Latest faults movements Regional scale Objectives: General geodynamic setting Characterization of geological features Delineation of seismogenic sources Site area (~1 km 2 ) Objectives: Permanent ground displacement Dynamic properties of foundation materials 5 km (maps scale 1:5 000) 25 km (maps scale 1:50 000) A need for application of increased efforts >150 km (maps scale 1: ) 15

16 The Near Field Issue The site for the NPP is generally chosen at a relatively aseismic part of the country. This generally means that well known seismogenic sources are more than at least 50 kms from the site. Consequently the seismic source that contains the site is a zone of diffuse seismicity (to use the terminology of the IAEA Safety Guide). Because there are few dispersed epicentres and that these are not well correlated with tectonic structures, these areas generally do not attract the interest of researchers and therefore contain the least amount of both geological and seismicity data that is available prior to the selection of the site. 16

17 Schematic diagram of offshore and onshore seismic prospecting Vibration generation: Deep sea probe vessel Kairei (7800cu.in.) etc. Deep sea area Approx. 50m under water Sea land border area (Shallow sea + Coastal area) Streamer vessel (with an airgun) Vessel with airgun Observation vehicle Vibrator Seabed cable Digital telemetry recording system Standalone recording system 17 Vibration generation: 4 large vibrators 17

18 Offshore and onshore prospecting line Vibration generation: [Sea area] Airgun (7800cu.in., 1500cu.in., 480cu.in.) [Land area] 4 large vibrators, etc. Generator interval: [Sea area] 25m [Land area] 25m, 50m, 100m, 500m 1km according to the distance from the site Vibration reception: [Sea area] Digital seabed cables [Land area] Standalone digital telemetry Receiver interval: [Sea area] 25m [Land area] 25m, 50m according to the distance from the site Prospecting line length: Approx. 20km in total length(for 2 prospecting lines) Vibration generation: Airgun (7800cu.in.) Generator interval: 50m Vibration reception: Streamer cables(444ch) Receiver interval: 12.5m Prospecting line length: Approx. 40km in total length (for 2 prospecting lines) ML08 1 ML

19 Dealing with Uncertainties Random (aleatory) uncertainties inherent in the variable Modelling (epistemic) uncertainties Balance between data generation and coping with uncertainties Only some part of the uncertainty can be reduced by additional data imported uncertainties cannot be reduced. 19

20 DSHA vs PSHA Both methods need to transform the seismic event to ground motion. This transformation is the major source of variability In PSHA the rate of earthquake recurrence is an important parameter In DSHA it is not a parameter but it may be used to distinguish between seismic sources 20

21 DSHA vs PSHA It is difficult to say which method is more conservative depends on the safety factors (in the DSHA) and the probability of exceedance considered The treatment of uncertainties (both aleatory and epistemic) should be similar in PSHA and DSHA In general, the differences in earlier and more recent studies are due to the fact that uncertainties have not been accounted for in the earlier studies (not because one is DSHA and the other is PSHA) 21

22 11. Maximum acceleration value of standard ground motion Ss (Unit: Gal) Standard ground motion unit 1 unit 2 unit 3 unit 4 unit 5 unit 6 unit 7 Ss-1 (F-B fault / JEA spectrum) Horizontal: 2280 Vertical: 1010 Horizontal: 1040 Vertical: 630 Ss-2 (F-B fault / Empirical Green's function) Ss-3 (Nagaoka plain western boundary fault zone / JEA spectrum) Ss-4 (Nagaoka plain western boundary fault zone / Empirical Green's function) Horizontal: 1354 Vertical: 402 Horizontal: 600 Vertical: 400 Horizontal: 589 Vertical: 314 Horizontal: 1156 Vertical: 501 Horizontal: 600 Vertical: 400 Horizontal: 826 Vertical: 332 (The horizontal figures represent the greater of the figures for the NS and EW components.) 22 22

23 Revised New Seismic Hazard at the K-K NPP Site Assessment currently under review 23

24 Using PSHA in S-PSA A plant PSA is a requirement of IAEA Safety Standard NS-R-1 (including external initiators) PSHA for a SPSA (for an operating NPP) needs to conform to overall project constraints: Generally no new data therefore need to cope with greater uncertainties Consistency in the treatment of epistemic uncertainties throughout the PSA (e.g. PSHA and fragilities) Overall scope and budget Requirements originating from the plant CDF 24

25 Using PSHA in Design Need to identify reference values that correspond to different design levels Use of a performance based approach USNRC RG (i.e. first onset of inelastic deformation FOSID) 25

26 Site Specific vs Regional PSHA Site specific PSHA (and of course DSHA) should be performed using the 4 scale approach (region, near region, site vicinity and site area). Regional PSHA can take into consideration only the regional database (maybe also near regional to a limited extent). Differences between the two may be attributable to the difference in the database For single site the attenuation relationships may have different σ values (one site variability) 26

27 Checking the results Historical records of large areas show that both spatial and temporal clustering occur (results of the project on the historical seismology of the Mediterranean Basin) The assumptions of stationarity and ergodicity of earthquake recurrence should be used with caution when the results of PSHA are checked against actual data. 27

28 Use of CAV as filter in SHA There is a need to use a lower bound filter in SHA because of the cut-off imposed by attenuation relationships (M~ ) In practice this also serves as a filter for earthquakes the occurrence of which will be of no consequence to the NPP (even if they occur very close) As the hazard metric is ground motion and not earthquake magnitude, a filter using a ground motion based parameter (such as CAV) will be more consistent with the aims of a SHA Eventually, it is possible to use a CAV distribution (instead of one number) and integrate this filter explicitly into PSHA. 28

29 Probabilistic Fault Displacement Analysis Draft DS 422 recommends a probabilistic approach to the fault displacement issue for existing NPPs There are a number of NPPs with such issues. TEPCO is conducting these studies on the K-K NPP. 29

30 Attenuation function for principal fault displacement P * ( D dmr, ) P ( Slipmr, ). P ( D dmrslip,, ) kn kn kn Conditional probability of slip f ( x) a bm Conditional probability of exceedance P (,, ) kn D d m r Slip quoted from Youns(2003) quoted from Youns(2003) 30

31 Attenuation function for distributed fault displacement P * ( D dmr, ) P ( Slipmr, ). P ( D dmrslip,, ) kn kn kn Conditional probability of slip Eq.(7) Eq.(8) Conditional probability of exceedance P (,, ) kn D d m r Slip is obtained by convolving the resulting a distribution function(e.g. below) with a distribution for maximum displacement on the principal rupture. m:magnitude r:distance h:1(hanging wall 0(foot wall) τzi:a term that varies randomly quoted from Youns(2003) 31 quoted from Youns(2003)

32 Conclusions SHA (for a NPP) is always part of a larger project (e.g. SPSA or design basis derivation) and needs to be considered within that context, i.e. Safety approach Performance and safety goals Treatment of uncertainties Scope and effort involved Preferably both PSHA and DSHA should be used. The database requirements, the uncertainty treatment and expert involvement should be similar in a PSHA and DSHA to the extent possible. 32

33 Emerging Issues Can DSHA and PSHA be treated within the same study? Are regional studies compatible with the 4 scaled approach? Is it reasonable to expect the regional studies and site specific studies to produce similar results? (differences in the details of database and larger sigmas for regional studies) 33

34 DS405 Volcanic Hazards in Site Evaluation OBJECTIVE To present the new safety guide on Volcanic Hazards in the site evaluation for nuclear installations. This Safety Guide is the revision and upgrade of the former Provisional Safety Standard Series Nº 1, published in 1997, as decided by NUSSC 19 th and as included in the long term structure of Safety Standards. 34

35 DS405 - BACKGROUND The current Safety Requirements on Site Evaluation for Nuclear Installations (NS-R-3) states, paragraph 3.52.: Historical data concerning phenomena that have potential to give rise to adverse effects on the safety of the nuclear installation such as volcanism, sand storms, severe precipitation, snow, ice, hail, and subsurface freezing of subcooled water (frazil), shall be collected and assessed. If the potential is confirmed, the hazard shall be assessed and design bases for these events shall be derived. 35

36 DS405 - BACKGROUND Since 1993, and due to the contribution of the safety reevaluation project in Armenia and the siting project in Indonesia, IAEA established the first integrated approach for conducting volcanic hazard evaluation for nuclear installations, in line with the treatment given to other external events. Thus, between 1993 and 1997, a Document was prepared, reviewed and issued with the aim to be part of the IAEA SSS as a Safety Guide. Approved for publication in March 1997 as Provisional Standards Series No 1. 36

37 DS405 - BACKGROUND The Provisional Safety Standards Series No. 1, Volcanoes and Associated Topics in Relation to Nuclear Power Plant Siting, (July 1997) was the first to provide any guidance on volcanic hazard assessment. Since its publication, several developments justify its revision: The science of volcanology has been transformed by significant advances in numerical modelling of volcanic processes and hazard assessment based, mainly, on stochastic models of these processes. The volcanology community has gained experience in applying these models in national hazard assessments and in national volcanic hazard mitigation programmes. Probabilistic volcanic hazards assessments for nuclear facilities have been conducted (e.g., in the USA, Japan, Armenia, Germany, Philippines, Indonesia) with important practical experience gained as a result. NEW

38 DS405 - BACKGROUND - 19 th NUSSC Meeting Secretariat s Proposal:.... Having in mind the experience gained and the need to close the gap in the Safety Guides, the Secretariat made the proposal to revise and to upgrade the document to a Safety Guide. NUSSC position: In the discussion, some members stated that they feel uncomfortable with the idea to have such a guide, because NPPs should not be sited or built near volcanoes. On the other hand, the Committee, of course, was aware of the fact that there are already NPPs existing in these areas and, therefore, agreed that such guidance is prepared in order to help the concerned Member States in a way that be compatible with the existing safety requirements.... NUSSC agreed that the Provisional Safety Standard Series No.1: Volcanoes and Associated Topics in relation to NPP Siting be promoted to a Safety Standard, subject to the Secretariat incorporating the Committee s comments and proposals. 38

39 VOLCANIC PHENOMENA AND ASSOCIATED HAZARDS Volcanic phenomena and associated hazards for nuclear installations, with implications for site selection and evaluation, and design: Some of these hazards constitutes a site exclusion criterion in the site selection and evaluation process. For other hazards the mitigation may be possible by either design or operational planning. Although a design basis may be achievable, sites with this hazard are usually avoided. VOLCANIC PHENOMENA: 1. Tephra falls 2. Pyroclastic flows and surges 3. Lava flows and domes 4. Debris avalanches, landslides and slope failures 5. Debris flows, lahars and floods 6. Opening of new vents 7. Volcano generated missiles 8. Volcanic gases 9. Tsunamis and seiches 10. Atmospheric phenomena 11. Ground deformation 12. Volcanic earthquakes and seismic events 13. Hydrothermal systems and ground water anomalies 39

40 VOLCANIC HAZARDS IN SITE EVALUATION GENERAL REMARKS: Detailed guidance is essential because volcanic processes are complex and varied, requiring multidisciplinary expertise and specialist knowledge. The Safety Guide will provide a staged approach for assessing these complex volcanic hazards. A staged approach will allow the hazard assessment to focus on phenomena that represent credible hazards to the site, rather than require an equivalent level of investigation and support for all types of volcanic phenomena and their hazards. This staged approach recognizes the need for increasing levels of information for increasing levels of potential hazard. This approach recognizes that sites located far from potentially active volcanoes may need to consider only a limited subset of potential volcanic hazards, whereas sites located closer to potentially active volcanoes may need to consider a full range of potential hazards. Thus, the Safety Guide is intended to clarify procedures and focus investigations on assessment of such credible external hazards. 40

41 Example Stage 3: Screening Volcanic Hazards Example: Numerical simulations of pyroclastic flow hazard at the BNPP site. Pyroclastic flows associated with small eruptions are not likely to reach the site (left), but those associated with large eruptions are likely to reach the site (right). The Safety Guide recommends focusing on those phenomena that represent credible hazards, in this case large eruptions of Mt. Natib. Figure from: Volentik et al., in press, Aspects of volcanic hazards assessment for the Bataan Nuclear Power Plant site, Philippines In: Volcanic and Tectonic Hazard Assessment for Nuclear Facilities, Cambridge University Press. 41

42 Who is in need or already using the provisional guidance document? Armenia: Japan: Use of guidelines to assess volcanic hazard to NPP site (Ministry of Environment) and Terms of Reference for site evaluation of new NPP unit. Kyushu assessment of tephra fallout hazard at NPPs on Kyushu by WestJec Consulting company (Mr. Nishizono, Mr. Inakura). GENNEN full assessment of the Rokkasho fuel cycle facility (Professor Nakamura). Philippines: Used to assess hazards at the Bataan NPP (Volentik et al., 2009, in press). The Government of Philippines requested ( ) to IAEA to conduct a site safety review mission to assist in the on updated criteria, methodologies and IAEA safety standards for a comprehensive site evaluation of BNPP site (February 2010). USA : Idaho - used to assess hazards at the Idaho National Laboratory (Wetmore et al., 2009, in press) Yucca Mountain expert elicitation for probabilistic hazard assessment consistent with draft guidelines Indonesia: Use of guidelines before Muria project suspended 42

43 IAEA SAFETY STANDARDS ON SITE EVALUATION Safety Standards Series hierarchy SITE EVALUATION Safety Fundamentals Requirements Safety Guides 8/25/ International Atomic Energy Agency REQUIREMENTS SAFETY GUIDES GUIDES SG-S9 to be revised DS

44 VOLCANIC HAZARDS IN SITE EVALUATION TITLE: According to the current updated versions of the group of IAEA Safety Guides for Site Evaluation, it is proposed the following title: Volcanic Hazards in Site Evaluation for Nuclear Installations. OBJECTIVES: to provide recommendations and guidance on assessing the volcanic hazards at a nuclear installation site, so as to enable the identification and characterization in a comprehensive manner of all potentially hazardous phenomena that may be associated with future volcanic events. These volcanic phenomena may affect the acceptability of the selected site during the survey and selection process and some of which may determine corresponding design basis parameters for the installation. 44

45 VOLCANIC HAZARDS IN SITE EVALUATION SCOPE: site evaluation for nuclear installations using a graded approach. This Safety Guide is intended to be used mainly during the site selection process of new nuclear installations. It may also be used for existing nuclear installations for a retrospective assessment of the volcanic hazards external to the installation that may affect it. This Safety Guide is included in the proposed long-term structure of safety standards. 45

46 VOLCANIC HAZARDS IN SITE EVALUATION DOCUMENT STRUCTURE: 1. Introduction 2. Overview of volcanic hazard assessment 3. General recommendations 4. Necessary information and investigations (Database) 5. Screening volcanic hazards 6. Site-specific volcanic hazard assessment 7. Nuclear installations other than power plants 8. Monitoring and preparation for response 9. Management of volcanic hazards evaluation Appendix 1: Description of types of volcanic phenomena. Annex 1: Volcanic Hazard Scenarios Annex 2: Worldwide sources of information Glossary 46

47 Recommended Stages in the Volcanic Hazard Assessment STAGE 1 STAGE 2 STAGE 3 STAGE 4 Initial assessment Characterize sources of volcanic activity as initiating events Hazards screening Evaluate hazards at site Volcanism <10Myr In appropriate region? Yes Is there current volcanic activity? Yes Potential for any volcanic hazard at the site? Yes Volcano(s) capable Develop sitespecific Volcanic Hazard Model No No Is there Holocene volcanic activity? No Yes No If not, (i.e. 10 Myr to 0.01 Myr), is future volcanic activity credible? Yes No Not a design basis event: no further investigation needed Increasing need for substantiation Site suitability decision, inputs for design basis 47

48 VOLCANIC HAZARDS IN SITE EVALUATION SOME COMMENTS ON VOLCANIC HAZARDS FOR SUBSURFACE INSTALLATIONS-GEOLOGICAL REPOSITORIES: Igneous intrusions processes (magma flows) and development of hydrothermal systems, should be treated in great detail; The conceptual model of volcanism should be adequate to other time scales ( yr) than for surface installations. Different treatment of uncertainties. 48

49 VOLCANIC HAZARDS IN SITE EVALUATION Since 2006, the DRAFT was prepared following NUSSC s decision in Finished in Jan09. The background and status of development of the DRAFT was presented to NUSSC in Nov 2008 with the decision that it will be presented in June The Secretariat realized that no DPP was requested and so a DPP was issued to formalize the process, in parallel with the DRAFT. Precisely, the main concern expressed in all NUSSC meetings is related to the contents and objectives of the Guide and the review of the prepared DRAFT assures that was done following instructions and comments from NUSSC. The DPP is fully compatible with the DRAFT. 49

50 VOLCANIC HAZARDS IN SITE EVALUATION SUMMARY of NUSSC 27 th Meeting (June 2009) : 1. The DPP was reviewed, commented and approved for submitting to CSS approval. 2. The DRAFT was also and simultaneouslyreviewed, commented and approved by NUSSC. 3. It was agreed to not hold up the DRAFT unnecessarily and it could be submitted to Member States for comments (to be done after approval by CSS of the DPP, see 1 above). 4. The Secretariat accepted that, in future, DPPs will be produced and endorsed by CSS before the drafting of standards begins. 50

51 International Atomic Energy Agency Thank you for your attention 51