PSAM 16 - Session M21 Overview

PSAM 16 Conference Session M21 Overview

Session Chair: Hyun Gook Kang (kangh6@rpi.edu)

Paper 1 EC303

Lead Author: Edward Chen Co-author(s): Bao Han han.bao@inl.gov Tate Shorthill tate.shorthil@inl.gov Nam Dinh ntdinh@ncsu.edu

Failure Mechanism Traceability and Application in Human System Interface of Nuclear Power Plants using RESHA
In recent years, there has been considerable effort to modernize existing and new nuclear power plants with digital instrumentation and control systems (DI&C). However, there has also been considerable concern both by industry and regulatory bodies on the risk and consequence analysis of these systems. Of particular concern is digital common cause failures (CCFs) as a result of failures or “misbehaviors” by the software in both the control and monitoring. While many new methods have been proposed to identify potential failure modes, such as Systems-theoretic Process Analysis (STPA), Hazard and Consequence Analysis for Digital Systems (HAZCADS), etc., these methods are focused primarily on the control action pathway of a system. Unlike the control pathway, the information feedback pathway lacks control actions, which are typically associated as software basic events, and thus assessment of software basic events in such systems is uncertain. In this work, we present the idea of intermediate processors and unsafe information flow (UIF) to help safety analysts trace failure modes in the feedback pathway and how they can be integrated into a fault tree for improved visual assessment. The concepts presented are demonstrated in a comprehensive case study on a representative advanced human system interface, specifically the Qualified Indication and Alarm System for Safety (QIAS-P). The qualitative software basic events are identified, and a fault tree analysis is conducted based on the Redundancy-guided Systems-theoretic Hazard Analysis (RESHA) methodology for tree construction. The case study demonstrates the inclusion of UIF and intermediate processors in the fault tree can significantly improve traceability of software failures that can occur in the highly complex feedback digital instrumentation and can also clarify fault tree construction when multiple sources of information are present in such systems.

Failure Mechanism Traceability and Application in Human System Interface of Nuclear Power Plants using RESHA

In recent years, there has been considerable effort to modernize existing and new nuclear power plants with digital instrumentation and control systems (DI&C). However, there has also been considerable concern both by industry and regulatory bodies on the risk and consequence analysis of these systems. Of particular concern is digital common cause failures (CCFs) as a result of failures or “misbehaviors” by the software in both the control and monitoring. While many new methods have been proposed to identify potential failure modes, such as Systems-theoretic Process Analysis (STPA), Hazard and Consequence Analysis for Digital Systems (HAZCADS), etc., these methods are focused primarily on the control action pathway of a system. Unlike the control pathway, the information feedback pathway lacks control actions, which are typically associated as software basic events, and thus assessment of software basic events in such systems is uncertain. In this work, we present the idea of intermediate processors and unsafe information flow (UIF) to help safety analysts trace failure modes in the feedback pathway and how they can be integrated into a fault tree for improved visual assessment. The concepts presented are demonstrated in a comprehensive case study on a representative advanced human system interface, specifically the Qualified Indication and Alarm System for Safety (QIAS-P). The qualitative software basic events are identified, and a fault tree analysis is conducted based on the Redundancy-guided Systems-theoretic Hazard Analysis (RESHA) methodology for tree construction. The case study demonstrates the inclusion of UIF and intermediate processors in the fault tree can significantly improve traceability of software failures that can occur in the highly complex feedback digital instrumentation and can also clarify fault tree construction when multiple sources of information are present in such systems.

Paper EC303 | Download the paper file. | Download the presentation PowerPoint file.

Name: Edward Chen (echen2@ncsu.edu)

Bio: Edward is a 4th year Ph.D. candidate researching risk and reliability in digital instrumentation and control systems at North Carolina State University under the direction of Dr. Nam Dinh. His primary areas of focus include risk quantification and model development in conventional PLC based as well as data-driven ML control and information systems. He has worked with multiple groups including Kairos power as a simulation developer for transient cases as well as a contractor for Idaho National Laboratories under the Light Water Sustainability Project. He has also worked on ARPA-e projects such as the Near Autonomous Management and Control system and has developed multiple data-driven autonomous safety systems.

Country: USA
Company: North Carolina State University
Job Title: Research Assistant

Paper 2 HA87

Lead Author: Han Bao Co-author(s): Hongbin Zhang: hzhang@terrapower.com Tate Shorthill: Tate.Shorthill@inl.gov Edward Chen: echen2@ncsu.edu

Common Cause Failure Evaluation of High Safety-significant Safety-related Digital Instrumentation and Control Systems using IRADIC Technology
Digital instrumentation and control (DI&C) systems in nuclear power plants (NPPs) have many advantages over analog systems but also pose different engineering and technical challenges, such as potential threats due to common cause failures (CCFs). This paper proposes an integrated risk assessment technology for DI&C systems (IRADIC) developed by Idaho National Laboratory for dealing with potential software CCFs in DI&C systems of NPPs. The methodology development of the IRADIC technology on the quantitative evaluation of software CCFs in high safety-significant safety-related DI&C systems in NPPs is illustrated in this paper. In IRADIC, qualitative hazard analysis and quantitative reliability and consequence analysis are successively implemented to obtain quantitative risk information, compare with respective risk evaluation acceptance criteria, and provide suggestions for risk reduction and design optimization. A comprehensive case study was also performed and documented in this paper. Results show that the IRADIC technology can effectively identify potential digital-based CCFs, estimate their failure probabilities, and evaluate their impacts to system and plant safety.

Common Cause Failure Evaluation of High Safety-significant Safety-related Digital Instrumentation and Control Systems using IRADIC Technology

Digital instrumentation and control (DI&C) systems in nuclear power plants (NPPs) have many advantages over analog systems but also pose different engineering and technical challenges, such as potential threats due to common cause failures (CCFs). This paper proposes an integrated risk assessment technology for DI&C systems (IRADIC) developed by Idaho National Laboratory for dealing with potential software CCFs in DI&C systems of NPPs. The methodology development of the IRADIC technology on the quantitative evaluation of software CCFs in high safety-significant safety-related DI&C systems in NPPs is illustrated in this paper. In IRADIC, qualitative hazard analysis and quantitative reliability and consequence analysis are successively implemented to obtain quantitative risk information, compare with respective risk evaluation acceptance criteria, and provide suggestions for risk reduction and design optimization. A comprehensive case study was also performed and documented in this paper. Results show that the IRADIC technology can effectively identify potential digital-based CCFs, estimate their failure probabilities, and evaluate their impacts to system and plant safety.

Paper HA87 | Download the paper file. | Download the presentation pdf file.

Name: Han Bao (han.bao@inl.gov)

Bio: Han Bao is a Nuclear Research & Development Scientist from Idaho National Laboratory who has been working on digital I&C risk assessment, machine learning applications in nuclear engineering and digital twin development for almost 10 years. Currently he is leading a digital I&C risk assessment project under the Light Water Reactor Sustainability (LWRS) Program to support transition from analog to digital technologies for nuclear industry and to assure the safety and reliability of vital digital I&C systems in nuclear power plants. He has led the methodology development and demonstration of an integrated risk assessment framework for digital I&C to provide a sustainable scientific basis for enabling industry to balance the digital-related risks, costs, reliability, and safety. He has also been conducting research in risk-informed design optimization of digital twins and AI-guided control systems. Han holds a Ph.D. degree in nuclear engineering from North Carolina State University.

Country: ---
Company: Idaho National Laboratory
Job Title: Nuclear Research & Development Scientist

Paper 3 SM60

Lead Author: Sung-Min Shin Co-author(s): Sang Hun Lee / k753lsh@kins.re.kr Seung Ki Shin / skshin@kaeri.re.kr

A novel approach for quantitative importance analysis of DI&C systems in NPP
The safety-related I&C system of nuclear power plants(NPPs) has quite complex interactions between its components in accordance with the redundancy/diversity design concept applied to ensure their functions, and the complexity is further increased with the recent introduction of digital characteristics. Meanwhile, safety signals can be generated/executed not only automatically but also manually but it is understood that the linkage between them was insufficient in the PSA process, the analysis framework of the existing NPP I&C system. Moreover, it is very difficult to secure quantified failure information of digital components required in analyzing the DI&C system according to the PSA framework. Therefore, this study proposes a new approach to resolve these problems, that is, the complex interactions between system components, the insufficient consideration on the relation between automatic and manual safety signal generation/execution, and the difficulty of securing failure information of digitalized components for PSA analysis. The method proposed in this study basically includes the human element from the system modeling phase to integrate automatic/manual aspects and assigns weights to related elements by dividing the redundancy/diversity characteristics of the DI&C system into macro and micro perspectives. When a specific component is unavailable, the effect of it is calculated in the microscopic perspective first and expanded to the macroscopic point of view, based on the assigned weights. The methodology was explained through simple examples, in addition, to verify its validity, a real-world system was analyzed using the methodology and the result was presented. The methodology is expected can be used to derive useful insights from the design to improvement stages for more diverse I&C systems by enabling quantitative importance analysis of without failure information.

A novel approach for quantitative importance analysis of DI&C systems in NPP

The safety-related I&C system of nuclear power plants(NPPs) has quite complex interactions between its components in accordance with the redundancy/diversity design concept applied to ensure their functions, and the complexity is further increased with the recent introduction of digital characteristics. Meanwhile, safety signals can be generated/executed not only automatically but also manually but it is understood that the linkage between them was insufficient in the PSA process, the analysis framework of the existing NPP I&C system. Moreover, it is very difficult to secure quantified failure information of digital components required in analyzing the DI&C system according to the PSA framework. Therefore, this study proposes a new approach to resolve these problems, that is, the complex interactions between system components, the insufficient consideration on the relation between automatic and manual safety signal generation/execution, and the difficulty of securing failure information of digitalized components for PSA analysis. The method proposed in this study basically includes the human element from the system modeling phase to integrate automatic/manual aspects and assigns weights to related elements by dividing the redundancy/diversity characteristics of the DI&C system into macro and micro perspectives. When a specific component is unavailable, the effect of it is calculated in the microscopic perspective first and expanded to the macroscopic point of view, based on the assigned weights. The methodology was explained through simple examples, in addition, to verify its validity, a real-world system was analyzed using the methodology and the result was presented. The methodology is expected can be used to derive useful insights from the design to improvement stages for more diverse I&C systems by enabling quantitative importance analysis of without failure information.

Paper SM60 | Download the paper file. | Download the presentation pdf file. Download the presentation PowerPoint file.

Name: Sung-Min Shin (smshin@kaeri.re.kr)

Bio: Sung-Min Shin received a Ph.D. degree in nuclear and quantum engineering from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea in 2016. He is currently a senior researcher with the Korea Atomic Energy Research Institute (KAERI). His research interests include the safety assessment of digital instrumentation and control (DI&C) and its probabilistic safety assessment (PSA) application. Recently, He is also interested in STAMP(Systems-Theoretic Accident Model and Processes)/STPA(Systems-Theoretic Process Analysis) application for analyzing the risk factors of DI&C.

Country: KOR
Company: Korea Atomic Energy Research Institute
Job Title: Senior Researcher