PSAM 16 - Session W15 Overview

PSAM 16 Conference Session W15 Overview

Session Chair: Marina Roewekamp (marina.roewekamp@grs.de)

Paper 1 JG277

Lead Author: Jan Grobbelaar Co-author(s): Zander Mausolff, amausolff@terrapower.com Brian Johnson, bjohnson@terrapower.com Brandon Chisholm, BMCHISHO@SOUTHERNCO.COM

Probabilistic Risk Assessment of the Molten Chloride Reactor Experiment Conceptual Design
A probabilistic risk assessment (PRA) is being developed for the Molten Chloride Reactor Experiment (MCRE). The MCRE is a low power, fast reactor which will provide key reactor physics data to support the design and licensing of the commercial Molten Chloride Fast Reactor (MCFR). The reactor uses sodium chloride and highly enriched uranium trichloride (NaCl-UCl3) eutectic as its fuel. The United States Department of Energy (DOE) awarded a contract to Southern Company-led team including TerraPower (TP) in 2020 to support the design and development of the MCRE. It is proposed to site MCRE within a testbed at the Idaho National Laboratory (INL). The final operating location at INL will be determined following the NEPA review. DOE review and approval of the MCRE safety basis will be required. The approach chosen for developing the analyses to support the MCRE safety basis is based on the Licensing Modernization Project’s (LMP) risk-informed performance-based (RIPB) approach documented in NEI 18-04. The longer-term goal in applying the LMP RIPB process on MCRE is to build experience for the future licensing of the MCFR. To support the RIPB approach, a PRA is being developed in conjunction with plant design. The scope of the PRA is all-modes all-hazards. The internal events PRA for the MCRE Conceptual Design was completed in 2021. Initiating events have been identified from hazards analysis of the Aircraft Reactor Test, the Oak Ridge National Laboratory Molten Salt Reactor Experiment, recent industry-publications, and a master logic diagram developed at TP. With little operational and component reliability data available for molten salt reactors, available liquid sodium component reliability data have been used for salt-wetted components. The safe stable end states modeled in the PRA are with the fuel subcritical in the fuel salt drain tank (cold shutdown) or maintained subcritical in the reactor vessel and heated to prevent freezing (hot shutdown). In the current model, core freezing in the reactor vessel has conservatively been assumed to be an undesired end state due to neutronic and mechanical challenges. However, this assumption will be refined as the maturity of the design information and safety analyses increases. Deterministic thermal-hydraulic analyses are in progress to confirm the transient evolutions. To this end, a new version of GOTHIC has been developed to better model a liquid core where the delayed neutron fraction is a function of flow rate, for example. Event trees and fault trees have been developed on a safety functional basis for reactivity control, decay heat removal control, fuel pump trip, core offload, and prevention of fuel freezing. In contrast to typical LWR PRAs, the MCRE PRA is much smaller as the event sequences are not complicated and there are far fewer systems to be modelled. For example, active decay heat removal is not needed as decay heat is removed passively via system losses. Standby safety injection systems are also not included in the design. A self-assessment was performed against the requirements of the ASME/ANS PRA Standard for Advanced Non-LWR Nuclear Power Plants (ASME/ANS RA-S-1.4-2021). The PRA has been applied to select the safety basis events (SBEs) and the preliminary results were plotted against frequency-consequence targets based upon relevant DOE regulatory limits. Although the safety classification of the MCRE SSCs will be informed by a comparison between the SBE results and the regulatory limits, additional details and analysis is required for elements such as uncertainties and defense-in-depth before the final SSC classifications can be made. The other PRA hazard groups are scheduled for analysis in 2022. Acknowledgment: "This material is based upon work supported by the Department of Energy under Award Number DE-NE0009045.” Disclaimer: "This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof."

Probabilistic Risk Assessment of the Molten Chloride Reactor Experiment Conceptual Design

A probabilistic risk assessment (PRA) is being developed for the Molten Chloride Reactor Experiment (MCRE). The MCRE is a low power, fast reactor which will provide key reactor physics data to support the design and licensing of the commercial Molten Chloride Fast Reactor (MCFR). The reactor uses sodium chloride and highly enriched uranium trichloride (NaCl-UCl3) eutectic as its fuel. The United States Department of Energy (DOE) awarded a contract to Southern Company-led team including TerraPower (TP) in 2020 to support the design and development of the MCRE. It is proposed to site MCRE within a testbed at the Idaho National Laboratory (INL). The final operating location at INL will be determined following the NEPA review. DOE review and approval of the MCRE safety basis will be required. The approach chosen for developing the analyses to support the MCRE safety basis is based on the Licensing Modernization Project’s (LMP) risk-informed performance-based (RIPB) approach documented in NEI 18-04. The longer-term goal in applying the LMP RIPB process on MCRE is to build experience for the future licensing of the MCFR. To support the RIPB approach, a PRA is being developed in conjunction with plant design. The scope of the PRA is all-modes all-hazards. The internal events PRA for the MCRE Conceptual Design was completed in 2021. Initiating events have been identified from hazards analysis of the Aircraft Reactor Test, the Oak Ridge National Laboratory Molten Salt Reactor Experiment, recent industry-publications, and a master logic diagram developed at TP. With little operational and component reliability data available for molten salt reactors, available liquid sodium component reliability data have been used for salt-wetted components. The safe stable end states modeled in the PRA are with the fuel subcritical in the fuel salt drain tank (cold shutdown) or maintained subcritical in the reactor vessel and heated to prevent freezing (hot shutdown). In the current model, core freezing in the reactor vessel has conservatively been assumed to be an undesired end state due to neutronic and mechanical challenges. However, this assumption will be refined as the maturity of the design information and safety analyses increases. Deterministic thermal-hydraulic analyses are in progress to confirm the transient evolutions. To this end, a new version of GOTHIC has been developed to better model a liquid core where the delayed neutron fraction is a function of flow rate, for example. Event trees and fault trees have been developed on a safety functional basis for reactivity control, decay heat removal control, fuel pump trip, core offload, and prevention of fuel freezing. In contrast to typical LWR PRAs, the MCRE PRA is much smaller as the event sequences are not complicated and there are far fewer systems to be modelled. For example, active decay heat removal is not needed as decay heat is removed passively via system losses. Standby safety injection systems are also not included in the design. A self-assessment was performed against the requirements of the ASME/ANS PRA Standard for Advanced Non-LWR Nuclear Power Plants (ASME/ANS RA-S-1.4-2021). The PRA has been applied to select the safety basis events (SBEs) and the preliminary results were plotted against frequency-consequence targets based upon relevant DOE regulatory limits. Although the safety classification of the MCRE SSCs will be informed by a comparison between the SBE results and the regulatory limits, additional details and analysis is required for elements such as uncertainties and defense-in-depth before the final SSC classifications can be made. The other PRA hazard groups are scheduled for analysis in 2022. Acknowledgment: "This material is based upon work supported by the Department of Energy under Award Number DE-NE0009045.” Disclaimer: "This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof."

Paper JG277 | Download the paper file. | Download the presentation PowerPoint file.

Name: Jan Grobbelaar (jgrobbelaar@terrapower.com)

Bio: Mr. Jan Grobbelaar has been with Terrapower since March 2021. He is the lead in developing the PRA for the Molten Chloride Reactor Experiment to support the request for DOE approval of the MCRE safety basis. Prior to joining Terrapower, he started his career at a PWR utility as a principal developer of their PRA for about 10 years, followed by working as a PRA consultant in the light water reactor industry for about 20 years. During this time he had participated as a peer reviewer in ASME/ANS PRA Standard peer reviews both in the USA and abroad. He holds a B.Sc. degree in nuclear engineering.

Country: USA
Company: Terrapower, Inc.
Job Title: Principal PRA Engineer

Paper 2 CL108

Lead Author: Claire Blackett Co-author(s): Maren H. Rø Eitrheim, maren.eitrheim@ife.no Andreas Bye, andreas.bye@ife.no

The Challenge of Assessing Human Performance and Human Reliability for First-of-a-Kind Technologies
There is growing interest worldwide in the potential of advanced reactor technologies such as Small Modular Reactors (SMRs) as a more competitive and efficient means of meeting future energy needs. SMRs represent a somewhat radical departure from the design of current nuclear power plants, with the promise of unique design attributes such as a smaller physical footprint, a smaller reactor core, simplification of the design, increased use of passive and/or inherent safety systems, and modular construction. Such attributes aim to minimize the potential for severe accidents to occur. The development of first-of-a-kind (FOAK) technologies such as SMRs inevitably invite consideration of the impact of such designs on how these plants will be operated, as compared to current generation plants. For example, the NuScale design proposes a plant of up to 12 reactor modules operated by a minimum shift crew of two senior reactor operators (SROs) and one reactor operator (RO), from a single control room. This is a significant change from current reactor designs that typically feature a minimum crew of three control room operators per reactor, with only one reactor per control room. What effects will this radical change in operating philosophy have on the conduct of operations in the control room, and on human performance and human reliability? Human performance and human reliability assessments for current generation plants rely on well-defined and well-documented scenarios within which potential human failure events can be modelled and evaluated, using operating experience as a valuable input to understand how operators typically respond when things go wrong. The challenge for FOAK designs is that there is no operating experience available yet to verify and validate predictive analyses of human performance and human reliability in these new operating situations. In 2018, the Halden Reactor Project (HRP) - now called the Halden Human Technology Organisation (HTO) project - initiated a research activity to understand and investigate the intended conduct of operations for SMR control rooms, and the subsequent effects these new ways of working may have on human performance. Previous experimental studies in the Halden Man-Machine Laboratory (HAMMLAB) have also yielded results that may be analogous to the potential human performance issues associated with the anticipated changes in operation of SMRs. In this paper, we will describe how the HRP/HTO project can contribute valuable knowledge to this new area which can be used to inform human performance and human reliability assessments in the absence of actual operating experience.

The Challenge of Assessing Human Performance and Human Reliability for First-of-a-Kind Technologies

There is growing interest worldwide in the potential of advanced reactor technologies such as Small Modular Reactors (SMRs) as a more competitive and efficient means of meeting future energy needs. SMRs represent a somewhat radical departure from the design of current nuclear power plants, with the promise of unique design attributes such as a smaller physical footprint, a smaller reactor core, simplification of the design, increased use of passive and/or inherent safety systems, and modular construction. Such attributes aim to minimize the potential for severe accidents to occur. The development of first-of-a-kind (FOAK) technologies such as SMRs inevitably invite consideration of the impact of such designs on how these plants will be operated, as compared to current generation plants. For example, the NuScale design proposes a plant of up to 12 reactor modules operated by a minimum shift crew of two senior reactor operators (SROs) and one reactor operator (RO), from a single control room. This is a significant change from current reactor designs that typically feature a minimum crew of three control room operators per reactor, with only one reactor per control room. What effects will this radical change in operating philosophy have on the conduct of operations in the control room, and on human performance and human reliability? Human performance and human reliability assessments for current generation plants rely on well-defined and well-documented scenarios within which potential human failure events can be modelled and evaluated, using operating experience as a valuable input to understand how operators typically respond when things go wrong. The challenge for FOAK designs is that there is no operating experience available yet to verify and validate predictive analyses of human performance and human reliability in these new operating situations. In 2018, the Halden Reactor Project (HRP) - now called the Halden Human Technology Organisation (HTO) project - initiated a research activity to understand and investigate the intended conduct of operations for SMR control rooms, and the subsequent effects these new ways of working may have on human performance. Previous experimental studies in the Halden Man-Machine Laboratory (HAMMLAB) have also yielded results that may be analogous to the potential human performance issues associated with the anticipated changes in operation of SMRs. In this paper, we will describe how the HRP/HTO project can contribute valuable knowledge to this new area which can be used to inform human performance and human reliability assessments in the absence of actual operating experience.

Paper CL108 | Download the paper file. | Download the presentation PowerPoint file.

Name: Claire Blackett (claire.blackett@ife.no)

Bio: Dr. Claire Blackett is a Senior Research Scientist in the department of Humans and Automation at the Institute for Energy Technology (IFE) in Halden, Norway. Claire’s research interests include human-automation and human-robot interaction, human performance and reliability in technologically advanced environments, and the ethical use of artificial intelligence in society. Before moving to Norway, Claire worked for several years as a human factors specialist in the UK nuclear industry, providing human factors and human reliability analysis support to nuclear safety cases, as well as conducting human factors engineering assessments and providing input to event investigations at existing nuclear sites. She continues to work within the nuclear industry today, as well as the petroleum, maritime, rail, healthcare, and process industries. Claire has a background in root cause analysis and accident investigation methods, with a PhD in this topic from University College Dublin, Ireland.

Country: NOR
Company: Institute for Energy Technology
Job Title: Senior Research Scientist

Paper 3 CL109

Lead Author: Claire Blackett Co-author(s): Maren H. Rø Eitrheim, maren.eitrheim@ife.no Robert McDonald, robert.mcdonald@ife.no Marten Bloch, marten.bloch@ife.no

Human Performance in Operation of Small Modular Reactors
The interest in Small Modular Reactors (SMRs) continues to grow worldwide, with even more countries and organisations exploring how SMRs could meet future energy needs. According to the International Atomic Energy Agency (IAEA), in 2020 there were over 70 SMR designs in development around the world. The interest in how SMR technology could be deployed has extended beyond energy production, and now includes using SMRs for e.g., district heating, desalination and even hydrogen production. There have been significant advancements in recent years in the development of SMR technology, with the NuScale light water SMR being the first commercial reactor of this type to have received design approval by a regulatory body in 2020. Despite these advancements, the IAEA notes that there are still issues related to control room staffing and human factors engineering for multi-module SMR plant designs that require “considerable” attention. Due to the highly commercial nature of the SMR industry, publicly available information about plans for the conduct of operations in SMR control rooms has been very limited resulting in a somewhat “black box” effect for those who wish to understand more about these issues. A research activity was initiated in 2018 within the Halden Reactor Project (HRP) - now called the Halden Human Technology Organisation (HTO) project - with a focus on understanding and investigation of human performance aspects of the operation of SMRs. One of the goals of this activity is to collect information via document reviews, workshops and short surveys and examine this to determine aspects such as: the state-of-the art in SMR developments with respect to conduct of operations; the human factors and human performance issues that are of the most interest to stakeholders such as regulators and researchers; whether there are lessons that can be learned from other industries that have implemented similar work strategies, e.g., monitoring and operation of multiple units from a single control room. In this paper, we will present and discuss the key human performance topics, uncertainties and research questions that have been identified so far in the HRP/HTO research activity on SMRs. We will describe the results from a pilot experimental study conducted in 2019 in a basic principle integral pressurised water reactor (iPWR) simulator, and how we intend to expand our experimental program to collect empirical data on human performance in the operation of SMRs.

Human Performance in Operation of Small Modular Reactors

The interest in Small Modular Reactors (SMRs) continues to grow worldwide, with even more countries and organisations exploring how SMRs could meet future energy needs. According to the International Atomic Energy Agency (IAEA), in 2020 there were over 70 SMR designs in development around the world. The interest in how SMR technology could be deployed has extended beyond energy production, and now includes using SMRs for e.g., district heating, desalination and even hydrogen production. There have been significant advancements in recent years in the development of SMR technology, with the NuScale light water SMR being the first commercial reactor of this type to have received design approval by a regulatory body in 2020. Despite these advancements, the IAEA notes that there are still issues related to control room staffing and human factors engineering for multi-module SMR plant designs that require “considerable” attention. Due to the highly commercial nature of the SMR industry, publicly available information about plans for the conduct of operations in SMR control rooms has been very limited resulting in a somewhat “black box” effect for those who wish to understand more about these issues. A research activity was initiated in 2018 within the Halden Reactor Project (HRP) - now called the Halden Human Technology Organisation (HTO) project - with a focus on understanding and investigation of human performance aspects of the operation of SMRs. One of the goals of this activity is to collect information via document reviews, workshops and short surveys and examine this to determine aspects such as: the state-of-the art in SMR developments with respect to conduct of operations; the human factors and human performance issues that are of the most interest to stakeholders such as regulators and researchers; whether there are lessons that can be learned from other industries that have implemented similar work strategies, e.g., monitoring and operation of multiple units from a single control room. In this paper, we will present and discuss the key human performance topics, uncertainties and research questions that have been identified so far in the HRP/HTO research activity on SMRs. We will describe the results from a pilot experimental study conducted in 2019 in a basic principle integral pressurised water reactor (iPWR) simulator, and how we intend to expand our experimental program to collect empirical data on human performance in the operation of SMRs.

Paper CL109 | Download the paper file. | Download the presentation PowerPoint file.

Paper 4 OL137

Lead Author: Ola Bäckström Co-author(s): Pavel Krcal Pavel.Krcal@lr.org Xuhong He Xuhong.He@lr.org

Use of PSA for Small Modular Reactors
PRA modelling approaches for nuclear installations of the current type have evolved over many years. A significant focus of the methodology has been devoted to the management of the complexity of the systems within a station and to the question how to appropriately estimate the metrics for an individual station. The acceptance framework for nuclear reactors is also focusing on individual stations in isolation. Further assumptions include the typical mission times of 24 respective 48 hours and to a great extent also independence of failures within an accident sequence. The booming interest in small modular reactors driven by their cost efficiency and increased safety might challenge the established methodology and bring new impulses at multiple aspects. Amongst them can be mentioned: Risk metric and safe state, component types, passive design, software (digital) control systems, multi-unit analysis. Some of the challenges are also of interest to existing reactors, especially in the effort to modernize them. The challenges will affect the risk assessment on different levels. This paper will discuss topics related to PRA quantification, like the need to manage longer mission times, multi-unit risk and digital control systems. Our aim is to demonstrate how such issues can be dealt with within the existing or extended frameworks.

Use of PSA for Small Modular Reactors

PRA modelling approaches for nuclear installations of the current type have evolved over many years. A significant focus of the methodology has been devoted to the management of the complexity of the systems within a station and to the question how to appropriately estimate the metrics for an individual station. The acceptance framework for nuclear reactors is also focusing on individual stations in isolation. Further assumptions include the typical mission times of 24 respective 48 hours and to a great extent also independence of failures within an accident sequence. The booming interest in small modular reactors driven by their cost efficiency and increased safety might challenge the established methodology and bring new impulses at multiple aspects. Amongst them can be mentioned: Risk metric and safe state, component types, passive design, software (digital) control systems, multi-unit analysis. Some of the challenges are also of interest to existing reactors, especially in the effort to modernize them. The challenges will affect the risk assessment on different levels. This paper will discuss topics related to PRA quantification, like the need to manage longer mission times, multi-unit risk and digital control systems. Our aim is to demonstrate how such issues can be dealt with within the existing or extended frameworks.

Paper OL137 | Download the paper file. | Download the presentation pdf file.

Name: Ola Bäckström (Ola.Backstrom@lr.org)

Bio: Ola holds a master of science in mechanical engineering within reactor technology. He is a specialist in PSA/PRA and has been involved in a variety of PSA and reliability projects since the early 90´s. From year 2000 and onwards Ola has been responsible for the development of RiskSpectrum and has for the past 15 years been the business owner of the RiskSpectrum product family.

Country: SWE
Company: LR RiskSpectrum
Job Title: VP RiskSpectrum