Pub Date : 2026-01-18DOI: 10.1016/j.nucengdes.2026.114765
Anderson Alvarenga de Moura Meneses , Lenilson Moreira Araujo
The Loading Pattern (LP) design is part of the nuclear fuel management of a Nuclear Power Plant (NPP). The design of an LP includes the permutation of fuel assemblies, as well as calculations performed with reactor physics codes, aiming to producing energy with the satisfaction of constraints such as those related to safety. From a computational perspective, it is an NP-hard combinatorial problem solved with success by optimization metaheuristics. With the breakthrough of generative Artificial Intelligence (AI), the immediate question is whether LPs can be designed within such paradigm. In the present article, a methodology is proposed for training and applying Wasserstein Generative Adversarial Networks (WGANs) for automatic generation of LPs of a Pressurized Water Reactor. With the application of the methodology to the benchmark IAEA-2D, WGANs generated LPs satisfying the safety constraints and objectives proposed. Thus, WGANs can learn implicit probability distributions of nucler fuel and automatically design high-quality LPs.
{"title":"Design of nuclear fuel loading patterns for a PWR with Wasserstein generative adversarial networks","authors":"Anderson Alvarenga de Moura Meneses , Lenilson Moreira Araujo","doi":"10.1016/j.nucengdes.2026.114765","DOIUrl":"10.1016/j.nucengdes.2026.114765","url":null,"abstract":"<div><div>The Loading Pattern (LP) design is part of the nuclear fuel management of a Nuclear Power Plant (NPP). The design of an LP includes the permutation of fuel assemblies, as well as calculations performed with reactor physics codes, aiming to producing energy with the satisfaction of constraints such as those related to safety. From a computational perspective, it is an NP-hard combinatorial problem solved with success by optimization metaheuristics. With the breakthrough of generative Artificial Intelligence (AI), the immediate question is whether LPs can be designed within such paradigm. In the present article, a methodology is proposed for training and applying Wasserstein Generative Adversarial Networks (WGANs) for automatic generation of LPs of a Pressurized Water Reactor. With the application of the methodology to the benchmark IAEA-2D, WGANs generated LPs satisfying the safety constraints and objectives proposed. Thus, WGANs can learn implicit probability distributions of nucler fuel and automatically design high-quality LPs.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114765"},"PeriodicalIF":2.1,"publicationDate":"2026-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-16DOI: 10.1016/j.nucengdes.2025.114747
Seong-Su Jeon, Jungjin Bang, Sang Gyun Nam, Jehee Lee, Youngjae Park, Soon-Joon Hong
Numerous Small Modular Reactors (SMRs) are being developed worldwide, and they are equipped with various types of Passive Safety Systems (PSSs). In the Republic of Korea, SMART100 and i-SMR are representative SMRs. SMART100 includes the Passive Safety Injection System (PSIS), the Passive Residual Heat Removal System (PRHRS), and Containment Pressure and Radioactivity Suppression System (CPRSS) while i-SMR is equipped with the Passive Emergency Core Cooling System (PECCS), the Passive Containment Cooling System (PCCS), and the Passive Auxiliary Feedwater System (PAFS). These systems operate based on natural forces such as gravity and buoyancy, performing safety functions without external power or operator action. However, because their operation relies on relatively weak and time-varying driving forces, reliable modeling using system analysis codes is important. In particular, improper simulation of key thermal-hydraulic phenomena such as pressure drop, condensation, boiling, and natural circulation can lead to predictions that deviate considerably from actual performance. To address these concerns, this study reviews the modeling and simulation of PSIS, PAFS, and PCCS in Korean SMRs using various system analysis codes. Based on the authors' extensive experience, detailed modeling considerations are derived to improve the representation of key physical phenomena. Furthermore, the study discusses the importance of robustness evaluation under degraded conditions using the Best Estimate with Performance Issues (BEPI) framework. The insights provided herein are expected to support the credible and technically robust application of system analysis codes to the design and safety assessment of passive safety systems.
{"title":"Modeling considerations for passive safety systems in Korean SMRs using system analysis codes","authors":"Seong-Su Jeon, Jungjin Bang, Sang Gyun Nam, Jehee Lee, Youngjae Park, Soon-Joon Hong","doi":"10.1016/j.nucengdes.2025.114747","DOIUrl":"10.1016/j.nucengdes.2025.114747","url":null,"abstract":"<div><div>Numerous Small Modular Reactors (SMRs) are being developed worldwide, and they are equipped with various types of Passive Safety Systems (PSSs). In the Republic of Korea, SMART100 and i-SMR are representative SMRs. SMART100 includes the Passive Safety Injection System (PSIS), the Passive Residual Heat Removal System (PRHRS), and Containment Pressure and Radioactivity Suppression System (CPRSS) while i-SMR is equipped with the Passive Emergency Core Cooling System (PECCS), the Passive Containment Cooling System (PCCS), and the Passive Auxiliary Feedwater System (PAFS). These systems operate based on natural forces such as gravity and buoyancy, performing safety functions without external power or operator action. However, because their operation relies on relatively weak and time-varying driving forces, reliable modeling using system analysis codes is important. In particular, improper simulation of key thermal-hydraulic phenomena such as pressure drop, condensation, boiling, and natural circulation can lead to predictions that deviate considerably from actual performance. To address these concerns, this study reviews the modeling and simulation of PSIS, PAFS, and PCCS in Korean SMRs using various system analysis codes. Based on the authors' extensive experience, detailed modeling considerations are derived to improve the representation of key physical phenomena. Furthermore, the study discusses the importance of robustness evaluation under degraded conditions using the Best Estimate with Performance Issues (BEPI) framework. The insights provided herein are expected to support the credible and technically robust application of system analysis codes to the design and safety assessment of passive safety systems.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114747"},"PeriodicalIF":2.1,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The high-temperature gas-cooled reactor pebble-bed module (HTR-PM) in China represents the world's first demonstration power plant of its kind. As one of the Generation IV reactors, it is designed with inherent safety characteristics and offers potential for optimization in the structural integrity management of its reactor pressure vessel (RPV). Key advantages of the HTR-PM vessel, such as lower fast neutron flux than conventional pressurized water reactors, significantly slow thermal transients and reversed thermal gradients, collectively contribute to reducing both the risk and consequence of its failure. Using probabilistic fracture mechanics (PFM), a quantitative model has been developed to assess flaw initiation, crack propagation, and fracture of the RPV under typical HTR-PM transient and test conditions. This model enables assessment of the RPV failure probability, thereby establishing a basis for risk-informed safety evaluation and for optimization of both manufacturing and in-service inspection (ISI) requirements. Simulation results show that the RPV failure probability of HTR-PM is extremely low in the conservative case of no ISIs during its entire service life. Furthermore, flaws in circumferential weld regions contribute only minimally to the overall failure probability, suggesting potential for further extending inspection intervals or even exempting these areas from periodic inspection.
{"title":"Risk-informed structural integrity assessment of the HTR-PM reactor pressure vessel","authors":"Bowen Li, Haitao Wang, Yanhua Zheng, Zhengming Zhang, Heng Peng","doi":"10.1016/j.nucengdes.2026.114764","DOIUrl":"10.1016/j.nucengdes.2026.114764","url":null,"abstract":"<div><div>The high-temperature gas-cooled reactor pebble-bed module (HTR-PM) in China represents the world's first demonstration power plant of its kind. As one of the Generation IV reactors, it is designed with inherent safety characteristics and offers potential for optimization in the structural integrity management of its reactor pressure vessel (RPV). Key advantages of the HTR-PM vessel, such as lower fast neutron flux than conventional pressurized water reactors, significantly slow thermal transients and reversed thermal gradients, collectively contribute to reducing both the risk and consequence of its failure. Using probabilistic fracture mechanics (PFM), a quantitative model has been developed to assess flaw initiation, crack propagation, and fracture of the RPV under typical HTR-PM transient and test conditions. This model enables assessment of the RPV failure probability, thereby establishing a basis for risk-informed safety evaluation and for optimization of both manufacturing and in-service inspection (ISI) requirements. Simulation results show that the RPV failure probability of HTR-PM is extremely low in the conservative case of no ISIs during its entire service life. Furthermore, flaws in circumferential weld regions contribute only minimally to the overall failure probability, suggesting potential for further extending inspection intervals or even exempting these areas from periodic inspection.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114764"},"PeriodicalIF":2.1,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Manual extraction of foreign objects from the lower head of a pressurizer poses significant challenges, including high personnel radiation exposure and operational complexity. This paper offers a comprehensive overview of the design and structure of a foreign object extraction robot for the lower head. It delves into the specific designs of key components like the clamping mechanism, waist and wrist rotating joints, shoulder and elbow swinging joints, and foreign object extraction actuators. Additionally, it provides an in-depth analysis of the robot's reliability. By enabling remote-controlled operation, the robot can extract foreign objects from the lower head without requiring personnel to enter the pressurizer. The findings of this study hold great significance for addressing similar foreign objects related incidents in pressurizers at other power plants.
{"title":"Design of a foreign object retrieval robot for the internal lower head of a nuclear power plant pressurizer","authors":"Fenghan Ran , Shiqi Chen , Shuai Huang , Xingli Zhu","doi":"10.1016/j.nucengdes.2026.114785","DOIUrl":"10.1016/j.nucengdes.2026.114785","url":null,"abstract":"<div><div>Manual extraction of foreign objects from the lower head of a pressurizer poses significant challenges, including high personnel radiation exposure and operational complexity. This paper offers a comprehensive overview of the design and structure of a foreign object extraction robot for the lower head. It delves into the specific designs of key components like the clamping mechanism, waist and wrist rotating joints, shoulder and elbow swinging joints, and foreign object extraction actuators. Additionally, it provides an in-depth analysis of the robot's reliability. By enabling remote-controlled operation, the robot can extract foreign objects from the lower head without requiring personnel to enter the pressurizer. The findings of this study hold great significance for addressing similar foreign objects related incidents in pressurizers at other power plants.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114785"},"PeriodicalIF":2.1,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-15DOI: 10.1016/j.nucengdes.2026.114771
Taizo Kanai , Miki Saito
Pool scrubbing is a key mitigation process in nuclear severe accidents (SAs), in which aerosols and gaseous fission products (FPs) are injected into water and removed through gas–liquid interactions. Although the decontamination factor (DF) depends strongly on particle size and hydrodynamic conditions, systematic DF datasets obtained under high-flow, non-condensing conditions—together with corresponding detailed two-phase flow-field measurements—remain scarce. This study addresses this gap by providing a high-accuracy, size-resolved DF dataset consistent with independently measured flow-field data. A 0.5-m-diameter, 8-m-high test facility was used to reproduce the characteristic severe-accident flow evolution from large globules to bubble breakup and the formation of a fine-bubble swarm. Using Welas, SMPS, and ELPI+, size-resolved DF values for soluble CsI aerosols were obtained over a wide aerodynamic-diameter range (sub-0.05 μm to 10 μm). The mixture water level was varied continuously from 0 to approximately 5 m, enabling DF measurements across evolving flow structures. The DF exhibited clear dependencies on both particle size and water level, and the transition from the injection/breakup region to the swarm region was directly reflected in the DF behavior. These trends were consistent with detailed axial bubble-size evolution measured previously in the same facility.
An empirical DF correlation was developed as a function of mixture water level and aerodynamic diameter. Comparison of this correlation with the DF measurements, flow-field data, and the mechanistic MELCOR/SPARC90 model showed that, while the major hydrodynamic transitions were consistent with the model, the experimentally observed particle-size dependence and gas-flow-rate independence revealed characteristics associated with strongly turbulent, multidimensional bubble motion in the swarm region. Furthermore, SMPS measurements demonstrated a pronounced DF increase for ultrafine particles (<0.05 μm) due to Brownian diffusion—providing new experimental evidence in a particle-size range for which reliable DF data have been largely unavailable.
A comparison with insoluble BaSO₄ aerosols generated under identical conditions showed consistently lower DF values for BaSO₄ despite nearly identical particle density, indicating that aerosol solubility has a significant influence on removal efficiency.
The systematic DF dataset obtained in this study elucidates the governing mechanisms of particle-size-dependent aerosol removal under realistic pool-scrubbing flow regimes. Combined with detailed flow-field measurements from the same facility, these results provide a robust benchmark for improving mechanistic DF models, particularly under high-flow conditions relevant to severe-accident source-term assessments.
{"title":"Experimental study of decontamination factor in pool scrubbing with two-phase flow evolution","authors":"Taizo Kanai , Miki Saito","doi":"10.1016/j.nucengdes.2026.114771","DOIUrl":"10.1016/j.nucengdes.2026.114771","url":null,"abstract":"<div><div>Pool scrubbing is a key mitigation process in nuclear severe accidents (SAs), in which aerosols and gaseous fission products (FPs) are injected into water and removed through gas–liquid interactions. Although the decontamination factor (DF) depends strongly on particle size and hydrodynamic conditions, systematic DF datasets obtained under high-flow, non-condensing conditions—together with corresponding detailed two-phase flow-field measurements—remain scarce. This study addresses this gap by providing a high-accuracy, size-resolved DF dataset consistent with independently measured flow-field data. A 0.5-m-diameter, 8-m-high test facility was used to reproduce the characteristic severe-accident flow evolution from large globules to bubble breakup and the formation of a fine-bubble swarm. Using Welas, SMPS, and ELPI+, size-resolved DF values for soluble CsI aerosols were obtained over a wide aerodynamic-diameter range (sub-0.05 μm to 10 μm). The mixture water level was varied continuously from 0 to approximately 5 m, enabling DF measurements across evolving flow structures. The DF exhibited clear dependencies on both particle size and water level, and the transition from the injection/breakup region to the swarm region was directly reflected in the DF behavior. These trends were consistent with detailed axial bubble-size evolution measured previously in the same facility.</div><div>An empirical DF correlation was developed as a function of mixture water level and aerodynamic diameter. Comparison of this correlation with the DF measurements, flow-field data, and the mechanistic MELCOR/SPARC90 model showed that, while the major hydrodynamic transitions were consistent with the model, the experimentally observed particle-size dependence and gas-flow-rate independence revealed characteristics associated with strongly turbulent, multidimensional bubble motion in the swarm region. Furthermore, SMPS measurements demonstrated a pronounced DF increase for ultrafine particles (<0.05 μm) due to Brownian diffusion—providing new experimental evidence in a particle-size range for which reliable DF data have been largely unavailable.</div><div>A comparison with insoluble BaSO₄ aerosols generated under identical conditions showed consistently lower DF values for BaSO₄ despite nearly identical particle density, indicating that aerosol solubility has a significant influence on removal efficiency.</div><div>The systematic DF dataset obtained in this study elucidates the governing mechanisms of particle-size-dependent aerosol removal under realistic pool-scrubbing flow regimes. Combined with detailed flow-field measurements from the same facility, these results provide a robust benchmark for improving mechanistic DF models, particularly under high-flow conditions relevant to severe-accident source-term assessments.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114771"},"PeriodicalIF":2.1,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-14DOI: 10.1016/j.nucengdes.2026.114755
Jeehee Lee , Seong-Su Jeon , Ju-Yeop Park , Hyoung Kyu Cho
The purpose of this study is to develop a robustness assessment methodology for performance evaluation considering the performance characteristics of passive safety systems being introduced in light water reactors and to propose safety analysis guidelines for passive safety systems by evaluating their impact on various performance degradation factors. To develop the methodology, the concerns with the introduction of passive safety systems and the current technical standards for passive safety systems from regulatory bodies around the world were analyzed. Since passive safety systems have less existing operating experience, there is uncertainty about the performance of the system, and it is necessary to prove the applicability of existing system analysis codes. In addition, since a passive safety system does not use devices such as pumps, it is more likely than a conventional safety system that the performance of the system will be degraded by changes in the internal or external environment. Therefore, this study developed a robustness assessment methodology consisting of seven steps to evaluate the impact of issues on the introduction of a passive safety system and to demonstrate the ability of the passive system to perform safety functions.
{"title":"Development of robustness assessment methodology for passive safety system against potential performance issue","authors":"Jeehee Lee , Seong-Su Jeon , Ju-Yeop Park , Hyoung Kyu Cho","doi":"10.1016/j.nucengdes.2026.114755","DOIUrl":"10.1016/j.nucengdes.2026.114755","url":null,"abstract":"<div><div>The purpose of this study is to develop a robustness assessment methodology for performance evaluation considering the performance characteristics of passive safety systems being introduced in light water reactors and to propose safety analysis guidelines for passive safety systems by evaluating their impact on various performance degradation factors. To develop the methodology, the concerns with the introduction of passive safety systems and the current technical standards for passive safety systems from regulatory bodies around the world were analyzed. Since passive safety systems have less existing operating experience, there is uncertainty about the performance of the system, and it is necessary to prove the applicability of existing system analysis codes. In addition, since a passive safety system does not use devices such as pumps, it is more likely than a conventional safety system that the performance of the system will be degraded by changes in the internal or external environment. Therefore, this study developed a robustness assessment methodology consisting of seven steps to evaluate the impact of issues on the introduction of a passive safety system and to demonstrate the ability of the passive system to perform safety functions.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114755"},"PeriodicalIF":2.1,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-14DOI: 10.1016/j.nucengdes.2026.114762
Yohan Kim, Hyungdae Kim
The bent heat pipe, which connects a compact reactor core to a relatively large power conversion system, is a critical component in the design of kilowatt-scale heat pipe-cooled reactors for space nuclear applications. Conventional screen wicks (SWs) used in straight heat pipes are prone to structural damage when bent, resulting in a loss of capillary performance. To address this issue, bendable braided-wire wicks (BWWs) have recently been proposed as a promising alternative due to their ability to maintain capillary functionality even under bending conditions. However, comprehensive studies on the capillary performance of BWWs in bent configurations remain limited. This study experimentally investigates the capillary flow characteristics of BWWs in both straight and bent geometries. The porosity was calculated through a geometric analysis of a unit cell in the wick and found to be 0.215. Capillary rise in a vertically oriented straight wick was visualized and quantitatively assessed using infrared thermography. The effective pore radius and permeability were determined by fitting the experimental data of the sample, yielding values were 67.3 μm and m2, respectively. Subsequently, a series of rate-of-rise experiments was conducted on bent BWWs to evaluate the impact of bending on their capillary behavior. The experimental results demonstrated that the BWW maintains its intrinsic capillary properties irrespective of geometric configuration. Finally, theoretical models were proposed to predict the porosity, permeability, and effective pore radius of the BWW structure. The predicted values agreed with the experimental measurements within a margin of approximately 20%.
{"title":"Characterization of braided-wire wicks for bent heat pipe applications: Experiments and modeling","authors":"Yohan Kim, Hyungdae Kim","doi":"10.1016/j.nucengdes.2026.114762","DOIUrl":"10.1016/j.nucengdes.2026.114762","url":null,"abstract":"<div><div>The bent heat pipe, which connects a compact reactor core to a relatively large power conversion system, is a critical component in the design of kilowatt-scale heat pipe-cooled reactors for space nuclear applications. Conventional screen wicks (SWs) used in straight heat pipes are prone to structural damage when bent, resulting in a loss of capillary performance. To address this issue, bendable braided-wire wicks (BWWs) have recently been proposed as a promising alternative due to their ability to maintain capillary functionality even under bending conditions. However, comprehensive studies on the capillary performance of BWWs in bent configurations remain limited. This study experimentally investigates the capillary flow characteristics of BWWs in both straight and bent geometries. The porosity was calculated through a geometric analysis of a unit cell in the wick and found to be 0.215. Capillary rise in a vertically oriented straight wick was visualized and quantitatively assessed using infrared thermography. The effective pore radius and permeability were determined by fitting the experimental data of the sample, yielding values were 67.3 μm and <span><math><mn>1.8</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>11</mn></mrow></msup></math></span> m<sup>2</sup>, respectively. Subsequently, a series of rate-of-rise experiments was conducted on bent BWWs to evaluate the impact of bending on their capillary behavior. The experimental results demonstrated that the BWW maintains its intrinsic capillary properties irrespective of geometric configuration. Finally, theoretical models were proposed to predict the porosity, permeability, and effective pore radius of the BWW structure. The predicted values agreed with the experimental measurements within a margin of approximately 20%.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114762"},"PeriodicalIF":2.1,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-13DOI: 10.1016/j.nucengdes.2026.114756
Rui Hou , Shaowei Tang , Jingliang Zhang , Yi Lei , Bin Zhang , Jianqiang Shan
Under Loss-of-Flow (LOF) accident conditions, the boiling behavior of the coolant has a decisive impact on core integrity. A multi-bubble slug model has been implemented and integrated within the integrated severe accident analysis code ISAA-Na to dynamically simulate the two-phase flow and heat transfer behavior in sodium-cooled fast reactors (SFRs) under LOF conditions. To validate the model's effectiveness, experiments BI1, E8, and EFM1 from the French CABRI facility were selected as benchmarks. The predictive capability of the model was assessed through a systematic comparison of the calculation results from ISAA-Na with experimental data and published results from other mainstream codes. The assessment indicates that the model accurately captures key physical phenomena, at boiling inception, the saturation temperature is predicted with a relative error of 0.12%. For the subsequent two-phase conditions, the model captures boiling initiation times with an absolute error of less than 0.3 s and axial locations within a relative error of 6%, while also accurately reproducing interface propagation and flow oscillations. This confirms the reliability of the model's implementation in ISAA-Na for SFR safety analysis, providing a robust basis for predicting subsequent accident progression.
{"title":"Validation of the ISAA-Na two-phase model for sodium-cooled fast reactors: An assessment against CABRI LOF experiments","authors":"Rui Hou , Shaowei Tang , Jingliang Zhang , Yi Lei , Bin Zhang , Jianqiang Shan","doi":"10.1016/j.nucengdes.2026.114756","DOIUrl":"10.1016/j.nucengdes.2026.114756","url":null,"abstract":"<div><div>Under Loss-of-Flow (LOF) accident conditions, the boiling behavior of the coolant has a decisive impact on core integrity. A multi-bubble slug model has been implemented and integrated within the integrated severe accident analysis code ISAA-Na to dynamically simulate the two-phase flow and heat transfer behavior in sodium-cooled fast reactors (SFRs) under LOF conditions. To validate the model's effectiveness, experiments BI1, E8, and EFM1 from the French CABRI facility were selected as benchmarks. The predictive capability of the model was assessed through a systematic comparison of the calculation results from ISAA-Na with experimental data and published results from other mainstream codes. The assessment indicates that the model accurately captures key physical phenomena, at boiling inception, the saturation temperature is predicted with a relative error of 0.12%. For the subsequent two-phase conditions, the model captures boiling initiation times with an absolute error of less than 0.3 s and axial locations within a relative error of 6%, while also accurately reproducing interface propagation and flow oscillations. This confirms the reliability of the model's implementation in ISAA-Na for SFR safety analysis, providing a robust basis for predicting subsequent accident progression.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114756"},"PeriodicalIF":2.1,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-13DOI: 10.1016/j.nucengdes.2026.114763
J. Schäfer , S. Taş , U. Hampel
This work presents a novel training approach for machine learning based instance segmentation, without the need for manual annotated datasets. With applications in experimental investigations in bubbly flows in reactor safety research. This semi-supervised process consists of two different neural networks, a conditional generative adversarial network used for data generation and a U-net style convolutional neural network for instance segmentation. We validated the approach using a fully automated experimental setup creating layered bubble curtains, which enables an evaluation of the bubble size distribution with and without overlapping bubbles. The final neural network is able to measure the average bubble size as well as to recreate the bubble size distribution accurately.
{"title":"Gas bubble detection and segmentation using a machine learning approach leveraging semi-supervised training","authors":"J. Schäfer , S. Taş , U. Hampel","doi":"10.1016/j.nucengdes.2026.114763","DOIUrl":"10.1016/j.nucengdes.2026.114763","url":null,"abstract":"<div><div>This work presents a novel training approach for machine learning based instance segmentation, without the need for manual annotated datasets. With applications in experimental investigations in bubbly flows in reactor safety research. This semi-supervised process consists of two different neural networks, a conditional generative adversarial network used for data generation and a U-net style convolutional neural network for instance segmentation. We validated the approach using a fully automated experimental setup creating layered bubble curtains, which enables an evaluation of the bubble size distribution with and without overlapping bubbles. The final neural network is able to measure the average bubble size as well as to recreate the bubble size distribution accurately.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114763"},"PeriodicalIF":2.1,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-12DOI: 10.1016/j.nucengdes.2026.114769
Björn Engström, Weimin Ma
A simplified neutron-kinetics model for recriticality calculations has been developed and implemented within the severe accident code MELCOR. The model partitions the core into active and inactive regions and determines neutron flux using factorization with adiabatic and prompt-jump approximations. Homogenization of pre-calculated core cell multiplication factors, together with Doppler and xenon reactivity defects calculated by perturbation theory, avoids explicit treatment of cross-sections. This approach allows recriticality analysis to be performed seamlessly within MELCOR at standard time-steps and computational cost. Comparisons with SARA project simulations show reasonable agreement. Integrated with MELCOR, the model extends its capabilities to predict reactivity, recriticality timing, fission power, fuel temperatures, and xenon transients. The model was applied to station blackout scenarios in a Nordic BWR with varying low-pressure safety injection timing and two decay-heat curves. Simulations suggest that the time window for recriticality may be substantial if the core periphery remains intact and coolant flow through a breached vessel is limited. In cases where recriticality occurred, fission power evolution was irregular, and even relatively low fission power accelerated containment heat-up and pressurization. These results demonstrate that the simplified model provides an efficient tool for investigating recriticality phenomena, their impact on severe accident progression, and the effectiveness of mitigation strategies under uncertainty and sensitivity analyses.
{"title":"Development and application of a simplified neutron-kinetics model for severe accident recriticality assessment","authors":"Björn Engström, Weimin Ma","doi":"10.1016/j.nucengdes.2026.114769","DOIUrl":"10.1016/j.nucengdes.2026.114769","url":null,"abstract":"<div><div>A simplified neutron-kinetics model for recriticality calculations has been developed and implemented within the severe accident code MELCOR. The model partitions the core into active and inactive regions and determines neutron flux using factorization with adiabatic and prompt-jump approximations. Homogenization of pre-calculated core cell multiplication factors, together with Doppler and xenon reactivity defects calculated by perturbation theory, avoids explicit treatment of cross-sections. This approach allows recriticality analysis to be performed seamlessly within MELCOR at standard time-steps and computational cost. Comparisons with SARA project simulations show reasonable agreement. Integrated with MELCOR, the model extends its capabilities to predict reactivity, recriticality timing, fission power, fuel temperatures, and xenon transients. The model was applied to station blackout scenarios in a Nordic BWR with varying low-pressure safety injection timing and two decay-heat curves. Simulations suggest that the time window for recriticality may be substantial if the core periphery remains intact and coolant flow through a breached vessel is limited. In cases where recriticality occurred, fission power evolution was irregular, and even relatively low fission power accelerated containment heat-up and pressurization. These results demonstrate that the simplified model provides an efficient tool for investigating recriticality phenomena, their impact on severe accident progression, and the effectiveness of mitigation strategies under uncertainty and sensitivity analyses.</div></div>","PeriodicalId":19170,"journal":{"name":"Nuclear Engineering and Design","volume":"449 ","pages":"Article 114769"},"PeriodicalIF":2.1,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145950224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号