Progress in Nuclear Energy最新文献

The article deals with the problems of creating composite materials for neutron shielding. A method for creating a composite material is proposed and described based on polyethylene and B₄C boron carbide. For the first time, the possibility of synthesizing composites of the proposed composition using cryogenic grinding under low-temperature processing conditions was established in this article. The main physical-mechanical, thermal and neutron-protective properties of the composite material were studied depending on the content of B₄C. A composite containing 30% boron carbide synthesized using cryogenic grinding has the following mechanical characteristics: flexural strength - 5.96 MPa, flexural strain at flexural strength 8.56 %, Vickers microhardness, 7.7 ± 0.8 HV₂₀₀.

The utilization of swirling flow in multiphase flow devices is prevalent for purposes such as mixing, separation, stabilization, and heat transfer enhancement, primarily owing to its characteristic of inducing low-pressure drop. In the nuclear industry, for example, two-phase swirling flow is applied in the nuclear gas generator to improve gas quality. In this study, an experimental investigation was conducted on the decaying swirling flow of gas-liquid in a horizontal pipe equipped with a vane-type swirler. The flow patterns were visually examined, and the pressure gradients along the test pipe and across the swirler were measured. The findings suggest the presence of four distinct swirling flow patterns at the swirler outlet (z/D = 0), namely chain flow, swirling gas column flow, swirling intermittent flow, and swirling annular flow. Because of swirl decay, these swirling flows recover their original pattern approximately 70D downstream from the swirler, with the exception of the swirling gas column flow. The flow regime maps at z/D = 10, 40, 70 and 100 are proposed and the pattern-based pressure gradient characteristics are analyzed. It is shown that the pressure gradient rises as both gas superficial velocity (j_g) and liquid superficial velocity (j_l) increase. The largest pressure gradient occurs within the swirler section, while the lowest is found upstream of the swirler. Near the swirler outlet (z/D = 0–33), the pressure gradient is approximately 1.5–2.3 times higher than at z/D = 33–67. Further downstream, at z/D = 67–100, it is 2.2–3.5 times greater, depending on the flow patterns.

The liquid droplet radiator (LDR) system can achieve a maximum heat transfer capacity of ten times that of traditional heat pipe heat exchangers per unit mass, which is very advantageous for the use of future space equipment. LDR mainly consists of a droplet generator, droplet layer, droplet collector, circulation pump, and corresponding pipelines, among which the droplet generator is the key equipment for generating the droplet layer. By using numerical and computational fluid dynamics methods, a droplet flow calculation model is developed to simulate the generation of droplets through high-frequency disturbance jet breakup in a thin atmosphere, without considering gravity. This model aims to replicate the operation of a droplet generator. After verification, the calculation result has an error of about 5%. Analyze the working fluids of silicone oil and liquid metal lithium, examining the parameters of droplet diameter d, droplet spacing s, and droplet uniformity generated by the jet under varying jet velocities, disturbance frequencies, and other working conditions. Additionally, assess the impact of each condition on the results. The results show that the diameter d of the droplet is directly proportional to the surface tension and the initial velocity of the jet, and inversely proportional to the disturbance frequency. The droplet spacing s and droplet polydispersity index λ are inversely proportional to the disturbance frequency and directly proportional to the jet velocity. The increase in surface tension and disturbance frequency can improve the uniformity of droplet flow, and the minimum droplet polydispersity index can be reduced to around 1%. Priority should be given to adjusting the jet velocity to obtain more uniform droplets to determine the working fluid. After leaving the nozzle, the jet will generate vortices, which will affect the subsequent droplet velocity. The average velocity of the droplet is about 80% of the undisturbed velocity of the jet. The analysis of jet motion characteristics and the variation patterns of droplet diameter and uniformity can provide a basis for the design of droplet generators.

Accurately predicting the mass flux, pressure profile, and velocity profile of the critical flow in a slit is essential for analyzing the breaking process of the liquid phase and calculating the aerosol source term for leak-before-break (LBB) monitoring and Loss of Coolant Accident (LOCA) risk analysis. A new critical flow model combining Two-fluid Model (TFM) and Delayed Equilibrium Model (DEM) is built to get accurate profiles while avoiding the same phase velocity in DEM and the arbitrary critical flow criterion in TFM. The new model is verified using past experiments of the critical flow in a slit. It proves to be accurate in mass flux but not in critical pressure, with maximum relative errors of around 25% in mass flux and around 80% in critical pressure. The new model is optimized for higher accuracy in critical pressure. The empirical equation of saturated phase mass flow rate fraction gradient is optimized by conducting approximate pressure profile calculation and regression analysis. The maximum relative error decreases little while the ratio of critical pressure relative errors lying in the range of ±40% increases after optimization. In contrast, the difference in the average abstract relative error of pressure between original TFM-DEM and DEM is much larger, for the maximum relative error of mass flux and critical pressure are around 25% and 110%. The comparison between the original and optimized TFM-DEM proves that the new critical flow model is accurate in mass flux and can be optimized to raise pressure calculation accuracy. The comparison between the original TFM-DEM and DEM proves that the phase velocity difference is the major source of accuracy improvement in the pressure profile.

The Molten Salt Reactor (MSR) idea is increasingly being recognized in the nuclear field due to its potential safety, sustainability, and economic efficiency advantages. The Molten Salt Fast Reactor (MSFR) benchmark, introduced in 2019, highlighted variations in results tied to different neutron cross-section libraries. This study investigates the impact of utilizing the ENDF/B-VIII.0 and JEFF-3.3 cross-section libraries for MSFR benchmark assessment compared to the ENDF/B-VII.1 database. Monte Carlo based open source code OpenMC is used for the analyses. Rigorous sensitivity analyses assess the influence of individual components, including the cross-section database, resonance elastic scattering, and Thermal Scattering Law (TSL). Beyond the criticality assessments, parameters such as delayed neutron fraction, temperature coefficient of reactivity, and neutron spectrum are compared for different cross-section libraries. Our analyses reveal that incorporating new evaluations for ²³³U (n,γ) and fission cross-sections in ENDF/B-VIII.0 significantly alters criticality results, i.e., more than 1700 pcm difference is seen between libraries. Similarly, critical concentration using ENDF/B-VII.1 and JEFF-3.3 is over-predicted by approximately 3%. The variations in Thermal Scattering Law (TSL) files do not yield substantial differences in outcomes due to the fast spectrum of the reactor. In some cases, the treatment of resonance elastic scattering leads to reactivity differences greater than 50 pcm. The benchmark compares ²³³U-started and Minor Actinide (MA)-started core. From the reactor physics point of view, the MA-started core leads to a 29% higher (n, γ) reaction rate than the ²³³U-started core. A 3–4% smaller value of thermal reactivity coefficient is obtained using the ENDF/B-VIII.0 library compared to the ENDF/B-VII.1 value. Using the ENDF/B-VIII.0 for the MSFR benchmark signifies using newer and better data for the GEN-IV reactors neutron physics calculations.

Safety, efficiency and economic benefits cannot be ignored in the development of nuclear energy. As a type of widely used fuel in nuclear reactors, solid fuel has limited potential, long investment return cycle of new nuclear reactors and great construction resistance. Addressing these challenges, two effective approaches involve the utilization of new fuel cladding materials, specifically Accident Tolerant Fuel (ATF), and the incorporation of novel fuel pellet structures to improve economic viability and safety. In this paper, an ATF of U₃Si₂-FeCrAl system with annular structure is analyzed based on a fuel behavior analysis code CAMPUS-ANNULAR. The assessment encompasses fuel performance under typical normal operating conditions and accident scenarios such as Loss of Coolant Accident (LOCA) and Reactivity Initiated Accident (RIA). By employing the solid fuel performance analysis code CAMPUS, a comparative work is conducted to evaluate the performance of the solid U₃Si₂-FeCrAl system under both normal and accident conditions. Results indicate that, during normal operation, the annular U₃Si₂-FeCrAl system with equivalent power density reduces peaking fuel temperatures by about 70 K–150 K in comparison to the solid U₃Si₂-FeCrAl system. This reduction enhances the temperature margin under accident conditions, subsequently lowering the risk of fuel meltdown. However, the annular U₃Si₂-FeCrAl system increases the risk of Pellet Cladding Mechanical Interaction (PCMI) failure under RIA condition.

This study aims to enhance the fuel cycle and fissile breeding performance of a sodium-cooled fast breeder reactor (FBR) by utilizing minor actinides (MAs) as a means of reactivity control alongside partially-inserted control rods. Choosing the PFBR-500 as the reference design, four core models, designated as Cases A, B, C, and D, utilizing various proportions of minor actinides (MAs) were built and simulated using OpenMC. The MA concentrations were optimized to compensate for the withdrawal of control rods and ensure the same initial reactivity for all cores. Burnup analysis over 365 EFPDs revealed a significant increase in cycle length and burnup for the modified cores along with a modest rise in breeding ratio. Notably, Case C, employing 3.45 wt.% MAs in the 88 inner-core fuel subassemblies achieved an extra 62.25 EFPDs cycle length, a 33.74% rise in single-cycle burnup, and a 3.86% increase in breeding gain compared to the reference. Loading MAs into the inner-core region proved to be more effective in enhancing both fertile-to-fissile conversion (thus extending the cycle length and fuel burnup) and transmutation than utilizing MAs throughout the core due to greater neutron flux at the core center. While Case D utilizing 2.2 wt.% MAs both in the inner and outer core fuel subassemblies had the highest overall MA loading, it demonstrated lower increments in cycle length, burnup, and breeding gain compared to Case C. Case C also exhibited the highest overall destruction rate (approximately 24%/y) for Np and Am isotopes, and successfully transmuted around 24.08 kg ²³⁷Np, 9.33 kg ²⁴¹Am and 3.82 kg ²⁴³Am over the course of a single year. The addition of MAs also achieved a slight flattening of the axial and radial flux profile and a decrease in the flux peaking factor. However, it slightly lowered the beta-effective, Doppler constant, and control rod assembly worth, and shifted the coolant void reactivity worth to the positive side.

Nuclear power plants (NPPs) are becoming increasingly interesting for future energy supply. Nowadays, most of the modern NPPs, such as Generation III+ and Small Modular Reactors (SMRs), offer an even higher safety standard than their predecessors, often relying on passive systems. Due to the continuous improvement of design and safety solutions in the nuclear field, it is essential to simultaneously increase modelling capabilities. This can be achieved by the advancement of modelling tools and by increasing the experience of the analyst. This work focuses on code validation to potentially allow users to gain modelling experience and to provide insights for further code development. Due to the complexity of severe accidents, it may prove to be challenging to model passive systems under such conditions, thus the validation is especially important for numerical codes, such as MELCOR.

Codes used for the simulation of severe accidents are simplified in order to be capable of capturing all occurring phenomena in a realistic computational time frame. Thus, it is not trivial if these codes are capable of modelling the combination of passive safety systems within the new integrated features present in many NPP designs. For this reason, this work aims to investigate how such assessments should be performed as well as to consider the severe accidents code MELCOR with respect to the simulation of a passive isolation condenser at the large-scale experimental facility PANDA with and without the presence of non-condensable gases.

Our work summarises the present ideas with regards to validation and verification of nuclear codes and highlights the fact that the severe accident code MELCOR is capable of simulating passive safety systems, such as the passive isolation condenser. Improvements can be made when modelling condensation in the presence of non-condensable gases and thus suggestions were made for the improved modelling.

In the event of a loss-of-coolant accident at a nuclear power plant, a large amount of steam enters the main pump, causing the pump's pressurization to deteriorate or even fail. To reveal the deterioration mechanism of the pump performance, the gas-liquid distribution characteristics in the centrifugal pump were studied by using structured grids and the Eulerian-Eulerian model. Based on the dimensional analysis method, a predictive correlation for bubble size was established, which included factors such as inlet gas volume fraction (IGVF), rotational speed, liquid flow rate, and impeller geometric parameters. When the predictive correlation is applied to the numerical simulation, the numerical two-phase pressurization agrees well with that obtained from the experiment. As the IGVF increases, the gas begins to accumulate at the impeller inlet under the effect of the pressure gradient force. Due to the large increase in liquid velocity, the gas begins to accumulate from the middle of the diffuser flow channel. The area occupied by the gas pocket in the impeller loses its pressurization capability. The pressure vortex formed at the inlet of the channel causes the diffuser to lose its pressurization capacity. An increase in rotational speed and a decrease in liquid flow rate can effectively prevent the formation and development of gas pockets in the impeller.

Transients are events that cause nuclear power plants (NPPs) to transition from a normal state to an abnormal state, which can lead to severe accidents if not properly handled. Transient identification is crucial for NPPs’ safety and operation. In this paper, we propose a novel multimodal text-time series learning framework(MTTL), the first work to apply a large language model for transient identification. The MTTL consists of self-supervised learning pre-training and zero-shot classification for transient identification. During pre-training, the framework utilizes a large language model(LLM) and a time-series(TS) encoder to fully exploit the rich multimodal information available in NPPs, i.e., to obtain the embeddings of both text data and TS data. The LLM is used to capture the transient knowledge of the NPPs by learning from the text data, and the TS encoder is used to capture the temporal dependencies of the transients by encoding the TS data. Both the LLM and the TS encoder have a linear projection head to map the embeddings into a common space. The similarity between the embeddings of the text and TS data is calculated to minimize the contrastive learning loss and obtain a pre-trained model with rich transient knowledge. During the zero-shot classification, the framework utilizes a pre-trained model to effectively identify real-world NPP transients where the data is different from the pre-trained simulated data. The proposed framework is evaluated on the High-Temperature Reactor-Pebblebed Modules (HTR-PM) plant, and the results demonstrate that the MTTL outperforms several baseline methods, including Transformer, LSTM, and CNN1D. The better zero-shot transient identification capability makes it possible to perform better in real-world NPPs.