International Nuclear Safety Group Activities on Multi-unit Risk Assessment

International Nuclear Safety Group (INSAG) 24-25 October 2017 Vienna Austria IAEA Activities on Multi-unit Risk Assessment Greg Rzentkowski Director Division of Nuclear Installation Safety Department of Nuclear Safety and Security

Background Methodology for MUPSA: requested by Member States involved in EESS-EBP NSNI conducted review of technical approaches available worldwide and identified challenges before starting a project Structure and work plan of such a project developed based on the review results: 3 phases (methodology / case study / refinement) Project initiated in December 2016; 4 CS meetings conducted up-to- now About 30 experts from 13 Member States (Belgium, Canada, Czech Republic, France, Germany, Hungary, Japan, Republic of Korea, Romania, Russia, Slovak Republic, UK, USA) Project report for the first phase: by end of 2017 Case study: to test the methodology; will be developed in 2018 TECDOC in 2019 with refined methodology and case study results for multi-unit probabilistic safety assessment. 2

References IAEA General Conference Resolution for Safety, GC(60)/RES/9: Encourages Member States that have not already done so to perform safety assessments in accordance with international best practice and relevant IAEA Safety Standards, including at multiunit sites . (paragraph 54) GSR Part 4 Safety Assessment for Facilities and Activities For sites with multiple facilities or multiple activities, account shall be taken in the safety assessment of the effects of external events on all facilities and activities (paragraph 4.36A) For facilities on a site that would share resources , the safety assessment shall demonstrate that the required safety functions can be fulfilled at each facility (paragraph 4.36B) SSR-2/1 Safety of Nuclear Power Plants: Design For multiple unit plant sites, the design shall take due account of the potential for specific hazards to give rise to impacts on several or even all units on the site simultaneously (Requirement 17 para 5.15B) DS484*/SSR-1 (revision of NS-R-3 Site Evaluation for Nuclear Installations) The site evaluation shall consider the potential of external natural and human induced hazards that can affect multiple installations at the same site as well as co-located site (Requirement 9) * DS484 is in Step 8 was sent to MS for comments. 3

Work Plan 1. Introduction 3.2.5. IE screening for MUPSA 3.2.6. Accident sequence analysis 3.2.7. Plant response analysis - HRA - Failure Correlations 3.2.8. Quantification & interpretation of results 3.2.9. Peer review 1.1. Background Work plan for 3 years includes 3 phases Phase 1 (2017) methodology for Multiunit PSA compiled and elaborated Phase 2 (2018) conduct a case study to test the methodology Phase 3 (2019) propose and develop a TECDOC with refined methodology and case study results 1.4. Objectives 1.5. Scope 2. Framework for MUPSA 2.1 Summary of existing technical basis 2.2 Overview of the technical approach as reflected in previous work 2.3 Identified challenges detailed 4. Level 2 and 3 MUPSA 3. Technical Approach for Level 1 MUPSA 3.1. Assumptions 3.2. MUPSA steps 3.2.1. Risk metrics 3.2.2. Risk aggregation 5. Safety goals 6. Case Study 7. Path Forward 8. References ANNEX. Current Approach and practice in MS Methodology based on actual PSA state of practice MUPSA technical approaches available worldwide consideration of the challenges identified 3.2.3. Analysis and interpretation of the risk profile for the single units 3.2.4. IE identification for MUPSA 4

Summary and Current Status Request for methodology for MUPSA from Member States NSNI review of technical approaches available and identification of challenges 3 phases project (initiated in December 2016) About 30 experts from 13 Member States Project progressing according to the work plan and schedule Phase 1 90% (methodology elaborated, first round comments addressed, finalization by end of 2017) Phase 2 - Technical Specification for the Case Study, kick-off meeting in February 2018 Major outcome: TECDOC with refined methodology and case study results (in 2019) 5

Other Activities - New NE Research Project PSA Benchmark for Multi-Unit, Multi-Reactor Sites At many nuclear sites world-wide, several NPPs, either of the same or of different types, designs, or age, are co-located on a single site. Regulations generally recognize the potential for multiunit accidents, but PSA of NPPs have mainly focused on estimating the risk arising from damage to a single NPP. Safety assessments, if based on deterministic and probabilistic approaches where the risk from a site with multiple reactors are represented by summing up the risks of individual units . is a too-simplified approach with several limitations ignores complex interactions during a severe event impacting a multiunit site. Example Atucha, Argentina: 2 unique PHWRs operating 1 SMR under construction 1 CANDU planned 1 PWR possible Fukushima accident lesson learned: Need to improve PSA methodologies when applied to multi-unit, multi-reactor- type nuclear sites. Several methods are being explored around the world to extend or translate per-unit PSA results to multi-unit site PSA results, such as core damage and large release frequencies. This CRP will bring together experts from LWR and HWR MSs to benchmark their practices, compare assumptions and results from Multi-unit Site PSA One NPP site with five different reactor types

Thank you!