PSAM 16 - Session Th03 Overview

PSAM 16 Conference Session Th03 Overview

Session Chair: Zoltan Kovacs (kovacs@relko.sk)

Paper 1 J.145

Lead Author: Jorge Luis Hernandez Co-author(s): Brinkman, Johannes. L. - brinkman@nrg.eu McLean Rob (RR) - rob.mclean@brucepower.com Mandelli, Diego - diego.mandelli@inl.gov Minibaev, Ruslan - R.Minibaev@iaea.org Jeon, Hojun - jeonhojun@khnp.co.kr Hortal, Francisco Javier - jav.hortal@gmail.com Guigueno, Yves - yves.guigueno@irsn.fr Nitoi, Mirela - mirela.nitoi@nuclear.ro Rowekamp, Marina - Marina.Roewekamp@grs.de Schneider, Raymond - schneire@westinghouse.com Siu, Nathan - nosiubiz@gmail.com

Presenter of this paper: Marina Roewekamp (marina.roewekamp@grs.de)

Advantages and challenges in implementing advanced Probabilistic Safety Assessment approaches and applications for Nuclear Power Plants: IAEA overview
Authors: Brinkman, J. L.; Jeon, H.; Guigueno, Y.; Hortal, J.; Luis Hernandez, J.; Mandelli, D.; McLean R.; Minibaev, R.; Nitoi, M.; Röwekamp, M.; Schneider, R.; Siu, N.. Considerable progress has been made in recent years on enhancing and extending the probabilistic safety assessment (PSA) approach as well as in its applications. Aiming to collect current experiences in Member States (MS) related to enhancements and developments in PSA and to develop a related technical document which can be used as a support for future updates of the related IAEA Safety Guides on PSA, IAEA started in 2018 a project supported by extrabudgetary funds from the USA. Consequently, the IAEA performed several Consultancy Meetings from 2019 to 2021 and two major Technical Meetings in 2019 and 2020, to draft the technical document compiling current status and experiences in MS regarding new areas considered as advanced PSA approaches and applications. The advanced PSA approaches aim at expanding the traditional PSA approaches by incorporating time related dependencies into the quasi-static Boolean logic structures used in their models. The PSA approaches considered in the scope of this technical document are the dynamic PSA, the incorporation of ageing aspects in the PSA model, and digital I&C reliability and modelling within PSA. Additionally, the technical document covers the Level 3 PSA that might be considered as a particular “advanced PSA approach”. Those advanced PSA approaches have been subject of many research studies and scientific publications aiming at demonstrating their level of establishment and maturity to provide new risk insights such as better understanding of accident progression or better representation of the risk profile of the plant in time. On the other hand, advanced applications of PSA enhance the traditional PSA approach by incorporating the combination of hazards, the modelling of non-permanent equipment in PSA, and the use of Level 2 PSA in the development of Severe Accident Management Guidelines (SAMG). It is widely accepted that the introduction of those advanced applications in the PSA models provides new insights in relation to both system analysis and reliability data (including equipment, human performance and common cause failures data). The results of the project have highlighted that currently those advanced PSA approaches and applications have been consolidated however there is no overall consensus among MS and PSA practitioners regarding their implementation acceptance. Therefore, for each of the selected approaches and applications, the technical document focuses on providing an overview of the current status of development, their advantages as well as on challenges and open issues related to their implementation. This paper focuses on summarizing the advantages and challenges identified through this project on each of the selected advanced PSA approaches and applications. The paper concludes with an analysis aiming at enhancing the understanding of advantages and challenges of incorporating such advanced PSA approaches into the safety assessment for nuclear power plants with the hope to contribute to the efforts of strengthening nuclear safety.

Advantages and challenges in implementing advanced Probabilistic Safety Assessment approaches and applications for Nuclear Power Plants: IAEA overview

Authors: Brinkman, J. L.; Jeon, H.; Guigueno, Y.; Hortal, J.; Luis Hernandez, J.; Mandelli, D.; McLean R.; Minibaev, R.; Nitoi, M.; Röwekamp, M.; Schneider, R.; Siu, N.. Considerable progress has been made in recent years on enhancing and extending the probabilistic safety assessment (PSA) approach as well as in its applications. Aiming to collect current experiences in Member States (MS) related to enhancements and developments in PSA and to develop a related technical document which can be used as a support for future updates of the related IAEA Safety Guides on PSA, IAEA started in 2018 a project supported by extrabudgetary funds from the USA. Consequently, the IAEA performed several Consultancy Meetings from 2019 to 2021 and two major Technical Meetings in 2019 and 2020, to draft the technical document compiling current status and experiences in MS regarding new areas considered as advanced PSA approaches and applications. The advanced PSA approaches aim at expanding the traditional PSA approaches by incorporating time related dependencies into the quasi-static Boolean logic structures used in their models. The PSA approaches considered in the scope of this technical document are the dynamic PSA, the incorporation of ageing aspects in the PSA model, and digital I&C reliability and modelling within PSA. Additionally, the technical document covers the Level 3 PSA that might be considered as a particular “advanced PSA approach”. Those advanced PSA approaches have been subject of many research studies and scientific publications aiming at demonstrating their level of establishment and maturity to provide new risk insights such as better understanding of accident progression or better representation of the risk profile of the plant in time. On the other hand, advanced applications of PSA enhance the traditional PSA approach by incorporating the combination of hazards, the modelling of non-permanent equipment in PSA, and the use of Level 2 PSA in the development of Severe Accident Management Guidelines (SAMG). It is widely accepted that the introduction of those advanced applications in the PSA models provides new insights in relation to both system analysis and reliability data (including equipment, human performance and common cause failures data). The results of the project have highlighted that currently those advanced PSA approaches and applications have been consolidated however there is no overall consensus among MS and PSA practitioners regarding their implementation acceptance. Therefore, for each of the selected approaches and applications, the technical document focuses on providing an overview of the current status of development, their advantages as well as on challenges and open issues related to their implementation. This paper focuses on summarizing the advantages and challenges identified through this project on each of the selected advanced PSA approaches and applications. The paper concludes with an analysis aiming at enhancing the understanding of advantages and challenges of incorporating such advanced PSA approaches into the safety assessment for nuclear power plants with the hope to contribute to the efforts of strengthening nuclear safety.

Paper J.145 | Download the paper file. | Download the presentation PowerPoint file.

Name: Jorge Luis Hernandez (J.Luis-Hernandez@iaea.org)

Bio: Mr. Jorge Luis Hernandez is a Nuclear Safety Officer at the IAEA where he contributes to and lead the development of several publications in the field of safety assessment such as the IAEA safety guide on Level 2 PSA and the TECDOCs on Advanced PSA Approaches and Applications for NPPs, and Approach and Methodology for the Development of Regulatory Safety Requirements for Advanced Nuclear Power Reactors: case study on SMRs. He holds a M.Sc. in Nuclear Safety Engineering and has 28 years of experience in the field of safety assessment for nuclear reactors, including SMRs. He worked for the nuclear safety authority of Cuba and for technical support organizations in France and Germany. He led and contributed to European Commission INSC projects to support the harmonization of best practices in nuclear safety around the world. He has actively contributed to several international working groups such as MDEP, ETSON, CSNI and the SMR Regulators’ Forum.

Country: AUT
Company: International Atomic Energy Agency
Job Title: Nuclear Safety Officer

Presenter Name: Marina Roewekamp (marina.roewekamp@grs.de)

Bio: - Diploma in Physics PhD (Dr. rer. nat. In Physical Chemistry / Materials Science) from University of Bonn - Senior Chief Expert for Hazards and PSA at GRS – the Federal German Nuclear Technical Safety Organization – for > 33 years - PSA work: mainly performing and/or reviewing Level 1 PSA, particularly for Internal and External Hazards (incl. hazard combinations) - Member of the German PSA Expert Panel for > 15 years - Former Chair and actual Vice Chair of OECD/NEA/CSNI Working Group on Risk Assessment (WGRISK) - Chair of OECD/NEA CSNI Expert Group on Fire Risk (EGFR) and of Management Board of OECD/NEA FIRE (Fire Events Records Exchange) Project - Consultant and/or reviewer for various IAEA Guides (SSG-64, SSG-67, SSG-68, DS523 (revision of SSG-3 on Level 1 PSA), DS528 (revision of SSG-4 on Level 2 PSA), TECDOCS on MUPSA, Advanced PSA Methods, Safety Assessment of Nuclear Installations Against Combinations of External Hazards, etc. - IAPSAM Board of Directors member since

Country: Germany
Company: Gesellschaft für Anlagen- und Reaktorsicherheit (GRS) gGmbH
Job Title: Chief Senior Expert

Paper 2 SM70

Lead Author: Eunseo So Co-author(s): Yunyeong Heo, Yunyeong.Heo@inl.gov Mohammad Abdo, Mohammad.Abdo@inl.gov Yong-Joon Choi, Yong-Joon.Choi@inl.gov

Development of Genetic Algorithms for Plant Reload Optimization for an Operating Pressurized Water Reactor
This paper summarizes the development, results, and enhancement activity of the Artificial Intelligence (AI) based automated nuclear power plant fuel reload optimization platform under the guidance of the United States Department of Energy, Light Water Reactor Sustainability Program, and Risk-Informed Systems Analysis Pathway. The research focuses on the optimization of the fuel arrangement to maximize the fuel cycle length. The AI-based Genetic Algorithm works with both convex and non convex, constrained or unconstrained problems. This can help explain the relationship between the fuel arrangement and fuel cycle length, in particular, the surrogate models used to reconstruct the Multiphysics problem maps the features/inputs of the problem to the fuel cycle length to provide such explanation. The Genetic Algorithm is composed of several evolutionary processes: fitness evaluation, parent selection, crossover, mutation, survivor selection, and termination. Crossover and mutation are the main steps responsible for injecting randomness/heuristics to prevent the algorithm from getting stuck in local minima. In this paper, roulette wheel parent selection, one-point crossover, swap mutation, and fitness-based survivor selection are used for demonstration to convert the fuel arrangement problem from the physical world (phenotype space) to the computational word (genotype space) via a user performed encoding/decoding step. Here, the search variables (genes) are the fuel locations in the core, whereas the values each variable takes, represent the fuel identification that will be placed in that specific location. The optimization process was demonstrated with a 1/4 core initial loading problem. In the core, 56 locations are loaded with five types of fuel assemblies, each type has different amount of enrichment and burnable poisons. As a result, the fuel cycle length increased to over 590 days, which is very close to the expected value. The results and enhancements in the optimization algorithm are also discussed in this paper.

Development of Genetic Algorithms for Plant Reload Optimization for an Operating Pressurized Water Reactor

This paper summarizes the development, results, and enhancement activity of the Artificial Intelligence (AI) based automated nuclear power plant fuel reload optimization platform under the guidance of the United States Department of Energy, Light Water Reactor Sustainability Program, and Risk-Informed Systems Analysis Pathway. The research focuses on the optimization of the fuel arrangement to maximize the fuel cycle length. The AI-based Genetic Algorithm works with both convex and non convex, constrained or unconstrained problems. This can help explain the relationship between the fuel arrangement and fuel cycle length, in particular, the surrogate models used to reconstruct the Multiphysics problem maps the features/inputs of the problem to the fuel cycle length to provide such explanation. The Genetic Algorithm is composed of several evolutionary processes: fitness evaluation, parent selection, crossover, mutation, survivor selection, and termination. Crossover and mutation are the main steps responsible for injecting randomness/heuristics to prevent the algorithm from getting stuck in local minima. In this paper, roulette wheel parent selection, one-point crossover, swap mutation, and fitness-based survivor selection are used for demonstration to convert the fuel arrangement problem from the physical world (phenotype space) to the computational word (genotype space) via a user performed encoding/decoding step. Here, the search variables (genes) are the fuel locations in the core, whereas the values each variable takes, represent the fuel identification that will be placed in that specific location. The optimization process was demonstrated with a 1/4 core initial loading problem. In the core, 56 locations are loaded with five types of fuel assemblies, each type has different amount of enrichment and burnable poisons. As a result, the fuel cycle length increased to over 590 days, which is very close to the expected value. The results and enhancements in the optimization algorithm are also discussed in this paper.

Paper SM70 | |

A PSAM Profile is not yet available for this author.

Paper 3 SS329

Lead Author: Ji Suk Kim Co-author(s): Man Cheol Kim, charleskim@cau.ac.kr

Effect of the Number of Ground Motion Subintervals on the Seismic PSA
There are two approaches to quantify a seismic probabilistic safety assessment(PSA) model. One is simulation approach such as Monte Carlo simulation and Latin Hypercube sampling, and the other is discrete approach[1]. The discrete approach has several advantages including the compatibility with internal event PSA quantification, preservation of logical links between the primary seismic event tree and secondary seismic event trees, and use of conventional PSA software[1,2]. However, in the discrete approach, the number of ground motion subintervals have been used with the lack of backgrounds for the effect on the seismic risks. In addition, the number of subintervals is limited in practice because the seismic PSA model should be quantified for each subinterval. In this study, we examine the effect of the number of ground motion subintervals on the seismic risk and identify the possibility that the seismic risk may be underestimated when the number of subintervals is small. We also provide the method for finding the regions of ground motion level that the underestimation of the seismic risk may be occur. Finally, we can suggest the quantification method for a large number of subintervals to improve the accuracy of results. This study can provide the backgrounds for determining the number of ground motion subintervals in seismic PSA. [1] International Atomic Energy Agency, Probabilistic Safety Assessment for Seismic Events. 2020, INTERNATIONAL ATOMIC ENERGY AGENCY: Vienna. [2] J.S. Kim. and M.C. Kim, Insights gained from applying negate-down during quantification for seismic probabilistic safety assessment. Nuclear Engineering and Technology, 2022.

Effect of the Number of Ground Motion Subintervals on the Seismic PSA

There are two approaches to quantify a seismic probabilistic safety assessment(PSA) model. One is simulation approach such as Monte Carlo simulation and Latin Hypercube sampling, and the other is discrete approach[1]. The discrete approach has several advantages including the compatibility with internal event PSA quantification, preservation of logical links between the primary seismic event tree and secondary seismic event trees, and use of conventional PSA software[1,2]. However, in the discrete approach, the number of ground motion subintervals have been used with the lack of backgrounds for the effect on the seismic risks. In addition, the number of subintervals is limited in practice because the seismic PSA model should be quantified for each subinterval. In this study, we examine the effect of the number of ground motion subintervals on the seismic risk and identify the possibility that the seismic risk may be underestimated when the number of subintervals is small. We also provide the method for finding the regions of ground motion level that the underestimation of the seismic risk may be occur. Finally, we can suggest the quantification method for a large number of subintervals to improve the accuracy of results. This study can provide the backgrounds for determining the number of ground motion subintervals in seismic PSA. [1] International Atomic Energy Agency, Probabilistic Safety Assessment for Seismic Events. 2020, INTERNATIONAL ATOMIC ENERGY AGENCY: Vienna. [2] J.S. Kim. and M.C. Kim, Insights gained from applying negate-down during quantification for seismic probabilistic safety assessment. Nuclear Engineering and Technology, 2022.

Paper SS329 | |

Name: Ji Suk Kim (sssuke@cau.ac.kr)

Bio:

Country: ---
Company: Chung-Ang university
Job Title: