PSAM 16 - Session W25 Overview

PSAM 16 Conference Session W25 Overview

Session Chair: Yong-Joon Choi (yong-joon.choi@inl.gov)

Paper 1 SK274

Lead Author: Dohun Kwon Co-author(s): Gyunyoung Heo gheo@khu.ac.kr

Development of Operator Response Time Model for DICE(Dynamic Integrated Consequence Evaluation)
DPSA (Dynamic Probabilistic Safety Assessment) which can facilitate the time dependency has been actively underway worldwide. DPSA can provide explicit consideration for dependencies among systems, components caused by plant states, or operator’s actions based on a continuous or discrete-time basis. In Kyung Hee University, DICE (Dynamic Integrated Consequence Evaluation) a dynamic reliability analysis tool using DDET (Discrete Dynamic Event Tree), was developed as a supporting tool for DPSA research. The DDET method is generally used in many DPSA frameworks. The diagnosis module, one of the modules in DICE, monitors plant status and decides the controls of the plant state by commanding components and operator action through a logical combination of physical variables. Particularly the decision by operator response time including omission or commission can greatly affect plant’s dynamics, so the diagnosis module in DICE attempts to try various approaches for operator models. Although an existing HRA (Human Reliability Assessment) works a good job for static PSA by providing HEP (Human Error Probability), the operator model, in DICE, should be able to provide operator response time to be used. There is a need for an operator model that provides operator response time based on the existing HRA method due to verifiable evidence of the model’s credibility meeting the current regulatory requirement. To develop the operator model, it is based on SPAR-H (Standardized Plant Analysis Risk Human Reliability Assessment) that is worldwide uses. To prove the credibility of the operator model with conventional SPAR-H result and the same result for operator model and SPAR-H, verification is conducted through Monte Carlo simulation. By connecting the operator model with DICE, it is possible to see dynamic plant status changes.

Development of Operator Response Time Model for DICE(Dynamic Integrated Consequence Evaluation)

DPSA (Dynamic Probabilistic Safety Assessment) which can facilitate the time dependency has been actively underway worldwide. DPSA can provide explicit consideration for dependencies among systems, components caused by plant states, or operator’s actions based on a continuous or discrete-time basis. In Kyung Hee University, DICE (Dynamic Integrated Consequence Evaluation) a dynamic reliability analysis tool using DDET (Discrete Dynamic Event Tree), was developed as a supporting tool for DPSA research. The DDET method is generally used in many DPSA frameworks. The diagnosis module, one of the modules in DICE, monitors plant status and decides the controls of the plant state by commanding components and operator action through a logical combination of physical variables. Particularly the decision by operator response time including omission or commission can greatly affect plant’s dynamics, so the diagnosis module in DICE attempts to try various approaches for operator models. Although an existing HRA (Human Reliability Assessment) works a good job for static PSA by providing HEP (Human Error Probability), the operator model, in DICE, should be able to provide operator response time to be used. There is a need for an operator model that provides operator response time based on the existing HRA method due to verifiable evidence of the model’s credibility meeting the current regulatory requirement. To develop the operator model, it is based on SPAR-H (Standardized Plant Analysis Risk Human Reliability Assessment) that is worldwide uses. To prove the credibility of the operator model with conventional SPAR-H result and the same result for operator model and SPAR-H, verification is conducted through Monte Carlo simulation. By connecting the operator model with DICE, it is possible to see dynamic plant status changes.

Paper SK274 | Download the paper file. | Download the presentation PowerPoint file.

Name: Dohun Kwon (skdmlclsrn11@gmail.com)

Bio: Dohun Kwon received a B.S. degree in nuclear engineering, from Kyung Hee University in 2021. He is currently working at Kyung Hee University for a Master's course. His research interests are thermal cycle analysis, safety analysis, and dynamic probabilistic safety assessments(DPSA) as called DPRA. He recently studies thermodynamic safety analysis for nuclear power plants and the decision of operator response time for DPSA.

Country: KOR
Company: Kyun Hee University
Job Title: Master's Candidate

Paper 2 ST93

Lead Author: Steve Prescott Co-author(s): Zhegng Ma (zhegang.ma@inl.gov) Svetlana Lawrence (svetlana.lawrence@inl.gov) Robby Christian (robby.christian@inl.gov) Daniel Nevius (daniel.nevius@inl.gov)

Using EMRALD to Simplify and Perform Dynamic Analysis with MAAP
Event Modeling Risk Assessment using Linked Diagrams (EMRALD) is a software tool developed at Idaho National Laboratory for researching the capabilities of dynamic probabilistic risk assessment (PRA). It provides a simple interface to represent complex interactions often seen when developing dynamic models. EMRALD can also interface with other applications by modifying inputs, running, and using their results within EMRALD for dynamic and integrated assessment. Linking with external codes was formerly done by user-defined scripts, requiring users to be familiar with both the scripting and the details of the application they wanted to use. A recent, enhanced feature to EMRALD provides a library capability to add custom forms for developing simple interfaces to run specific applications, making it so the user does not need to write scripts or have extensive training in the tool they want to pull data from. This feature is especially useful for thermal hydraulic applications such as Modular Accident Analysis Program (MAAP), in which after the plant model is developed, an inhouse MAAP expert is still needed to make any modifications for an analysis. This report outlines the custom form process used to develop an EMRALD user interface for MAAP and demonstrates how it can be used to enable a typical analyst to easily perform a custom thermal hydraulic analysis with existing MAAP general model.

Using EMRALD to Simplify and Perform Dynamic Analysis with MAAP

Event Modeling Risk Assessment using Linked Diagrams (EMRALD) is a software tool developed at Idaho National Laboratory for researching the capabilities of dynamic probabilistic risk assessment (PRA). It provides a simple interface to represent complex interactions often seen when developing dynamic models. EMRALD can also interface with other applications by modifying inputs, running, and using their results within EMRALD for dynamic and integrated assessment. Linking with external codes was formerly done by user-defined scripts, requiring users to be familiar with both the scripting and the details of the application they wanted to use. A recent, enhanced feature to EMRALD provides a library capability to add custom forms for developing simple interfaces to run specific applications, making it so the user does not need to write scripts or have extensive training in the tool they want to pull data from. This feature is especially useful for thermal hydraulic applications such as Modular Accident Analysis Program (MAAP), in which after the plant model is developed, an inhouse MAAP expert is still needed to make any modifications for an analysis. This report outlines the custom form process used to develop an EMRALD user interface for MAAP and demonstrates how it can be used to enable a typical analyst to easily perform a custom thermal hydraulic analysis with existing MAAP general model.

Paper ST93 | Download the paper file. | Download the presentation PowerPoint file.

Name: Steve Prescott (Steven.Prescott@inl.gov)

Bio: Steve Prescott is a software engineer at the Idaho National Laboratory. He started working on the PRA software EMRALD as an intern and has never stopped working on risk analysis software since. Now his primary focus is on dynamic PRA and incorporating other tools such as fire, flood, and physical security. For fun and to have a place to live, over the last few years, he designed and built a net zero house that is solar powered and solar heated.

Country: ---
Company: Idaho National Labratory
Job Title: Software Engineer

Paper 3 SA202

Lead Author: Gulcin Sarici Turkmen Co-author(s): Alper Yilmaz yilmaz.15@osu.edu Tunc Aldemir aldemir.1@osu.edu

USE OF MACHINE LEARNING TECHNIQUES TO REDUCE THE COMPUTATIONAL EFFORT FOR DYNAMIC PROBABILISTIC RISK ASSESSMENT
Dynamic probabilistic risk assessment (DPRA) is an important approach for assessing safety of nuclear power plant (NPP) operation. Since NPPs are highly complex systems, it is necessary to produce large amounts of data that represent different possible situations during NPP evolution following an accident in order to carry out the DPRA comprehensively. In addition to the fact that it may take months to produce such data with NPP accident analysis codes (e.g, RELAP5, MELCOR) and DPRA software developed for such a task (e.g., ADAPT), the task also requires the use of significant amount of computer and human resources. The effectiveness of models that have been developed using Machine Learning (ML) techniques in recent years, especially those that can represent time-dependent data and make predictions for the consequences of possible initiating events, have proven to be useful in reducing the computational effort for such a task. Recurrent Neural Network (RNN) approach represent some efficient ML methods that can be used in modeling the development of accidents and predicting potential consequences. This study is aimed at using DPRA data sets for two different NPPs to train RNN model to make predictions of possible NPP behavior under accident conditions as the accident evolves. The data sets to be used have been obtained from previous studies. The first data set consists of about 10,000 scenarios generated for a 4-loop pressurized water reactor (PWR) with station blackout (SBO) as the initiating event using RELAP5-3D/RAVEN, and the second data set consists of 2,656 scenarios generated for a 3-loop PWR under SBO using MELCOR/ADAPT. After developing deep learning (DL) model trained with the data for one power plant, the same DL model will be retrained with the data of the second power plant to apply the Transfer learning (TL) approach. An important advantage of TL is that it would save time and resources for a more comprehensive DPRA, as well as making make more accurate predictions under accident conditions.

USE OF MACHINE LEARNING TECHNIQUES TO REDUCE THE COMPUTATIONAL EFFORT FOR DYNAMIC PROBABILISTIC RISK ASSESSMENT

Dynamic probabilistic risk assessment (DPRA) is an important approach for assessing safety of nuclear power plant (NPP) operation. Since NPPs are highly complex systems, it is necessary to produce large amounts of data that represent different possible situations during NPP evolution following an accident in order to carry out the DPRA comprehensively. In addition to the fact that it may take months to produce such data with NPP accident analysis codes (e.g, RELAP5, MELCOR) and DPRA software developed for such a task (e.g., ADAPT), the task also requires the use of significant amount of computer and human resources. The effectiveness of models that have been developed using Machine Learning (ML) techniques in recent years, especially those that can represent time-dependent data and make predictions for the consequences of possible initiating events, have proven to be useful in reducing the computational effort for such a task. Recurrent Neural Network (RNN) approach represent some efficient ML methods that can be used in modeling the development of accidents and predicting potential consequences. This study is aimed at using DPRA data sets for two different NPPs to train RNN model to make predictions of possible NPP behavior under accident conditions as the accident evolves. The data sets to be used have been obtained from previous studies. The first data set consists of about 10,000 scenarios generated for a 4-loop pressurized water reactor (PWR) with station blackout (SBO) as the initiating event using RELAP5-3D/RAVEN, and the second data set consists of 2,656 scenarios generated for a 3-loop PWR under SBO using MELCOR/ADAPT. After developing deep learning (DL) model trained with the data for one power plant, the same DL model will be retrained with the data of the second power plant to apply the Transfer learning (TL) approach. An important advantage of TL is that it would save time and resources for a more comprehensive DPRA, as well as making make more accurate predictions under accident conditions.

Paper SA202 | | Download the presentation pdf file.

Name: Gulcin Sarici Turkmen (sariciturkmen.1@osu.edu)

Bio: Gulcin Sarici Turkmen got her Bachelor of Science and Master of Science degrees from the Department of Nuclear Engineering at Hacettepe University in Turkey. In Master of Science, she focused on modelling “Effect of Thermal-Neutronic Coupling on the Cross-Sections of Nuclear Fuel”. She also worked as a project manager in a private company in Turkey and worked the system modelling of VVER-1200. She was awarded a doctoral scholarship by the Turkish government and now she is a PhD student at the Ohio State University and she is doing her research on multi-unit dynamic probabilistic risk assessment for small modular reactors with Prof. Tunc Aldemir.

Country: USA
Company: The Ohio State University
Job Title: Graduate Research Associate