最新刊期
卷 56 ， 期 5 ， 2024
- “长江中下游崩岸预警及治理技术研究取得新进展，有望提高河道崩岸预见性和治理技术水平，为水安全保障提供科技支撑。”摘要：Significance Bank collapse is a major phenomenon in the natural evolution of alluvial river channels, prevalent in nearly all river and coastal areas worldwide. It represents a critical issue in river geomorphology evolution and its impacts. The middle and lower reaches of the Yangtze River are particularly susceptible to bank collapse. This phenomenon substantially threatens flood control, navigational channels, economic infrastructure, and human safety along rivers. The construction and operation of hydraulic projects, such as the Three Gorges Reservoir, have significantly altered the water and sediment dynamics in the middle and lower reaches of the Yangtze River. These changes have led to continuous channel erosion, localized river channel adjustments, and frequent bank collapses in recent years. Over the past two decades, more than 1000 incidents of bank collapse have been recorded, cumulatively spanning 760 km, causing various adverse impacts. Despite extensive research and remediation efforts, the complexity of influencing factors, intricate mechanisms, and the unpredictable nature of bank collapse continue to present significant early warning and prevention challenges. Therefore, conducting in-depth research on bank collapse mechanisms, monitoring, early warning technology, and governance systems in these regions has become necessary and urgent.Progress This study adopts a comprehensive approach encapsulated in the theme “Mechanism Revelation—Extensive Screening—Intervention Monitoring—Multi-scale Early Warning—Systematic Prevention and Control” to address critical scientific and technical challenges. It focuses on understanding bank collapse mechanisms under flow-solid coupling, developing intelligent hazard screening, real-time sensing of riverbank soil, bank collapse simulation, early warning technologies, and developing systematic prevention and control technologies integrating river channel control measures. The research employs data analysis, numerical simulation, and flume experiments to elucidate the relationship between river channel changes and bank collapse dynamics under continuous erosion conditions. It utilizes remote sensing, data fusion, and machine learning methods for intelligent hazard screening based on multi-source information fusion from “air, space, and ground”. In addition, the study integrates stereoscopic monitoring, data analysis, and technical integration methods to develop a comprehensive intelligent monitoring network and equipment for riverbank slopes. It develops a multi-scale universal model and a dynamic early warning technology system using coupled simulation, statistical analysis, and machine learning in prediction and early warning. The research proposes and demonstrates bank collapse prevention and control technologies coupled with river channel control measures, reflecting innovations primarily in the dynamic response relationship between river channel changes and bank collapse under discontinuous bank protection conditions; the implementation of innovative screening technologies; the integration of multi-source heterogeneous information from “air, space, and ground” for intelligent hazard screening; advancements in monitoring technologies, including the development of real-time sensing equipment; and innovations in simulation and early warning technologies, establishing a universal multi-scale dynamic early warning platform.Conclusions and Prospects This study aims to uncover the driving factors and intrinsic mechanisms of bank collapse in the Yangtze River’s middle and lower reaches under continuous erosion conditions. It seeks to develop key technologies for intelligent hazard perception, early warning, and preventing and controlling bank collapses. These efforts aim to achieve intelligent screening, multi-factor monitoring, dynamic early warning, and systematic prevention and control of bank collapse hazards. The research outcomes will guide the planning, design, and implementation of river channel management in these regions of the Yangtze River and facilitate the construction and application of a dynamic early warning platform for bank collapses in critical areas along the river. This will provide robust scientific support for decision-making by basin authorities, significantly enhance the predictability and management of bank collapse risks in the middle and lower Yangtze River, promote overall improvement in bank collapse hazard prevention and control, and advance flood control and disaster reduction technologies. The study will facilitate a shift from post-disaster rescue to pre-disaster prevention, providing technological support for river system management and water security enhancement. This approach is expected to yield significant economic, social, and ecological benefits, offering extensive application value and broad promotion prospects.关键词：riverbank collapse;mechanism;screening;monitoring and early warning;systematic management;middle and lower reaches of the Yangtze River1022|1649|0更新时间：2024-11-15
FOCUS ON STATE KEY RESEARCH DEVELOPMENT PROGRAM OF CHINA
- “在地质灾害领域，专家建立了大型危岩崩塌预警体系，为科学精准防控提供解决方案。”摘要：Large-scale rock collapse is a prevalent geological hazard in China, characterized by complex causes, wide distribution, strong concealment, sudden onset, and significant destructiveness, making early warning challenging to achieve. The instability mechanisms and warning models of rock collapse disasters vary widely. Scientific identification of damage to rock bridge structural surfaces is essential for early warning and prevention of these disasters. Conducting damage identification of rock bridge structural surfaces and comprehensive monitoring research, which includes static, dynamic, and environmental indicators (SDEI), is key to early monitoring and warning of large-scale rock collapses. The study of early warning mechanisms based on the identification of separation damage precursors is an effective means to enhance the timeliness of early warnings for such disasters. With the development of micro-electromechanical systems and cloud-edge collaboration technology, a new multivariate early warning paradigm is expected to emerge in the future. This paradigm would feature real-time linkage of dynamic stability evaluation, unstable working condition prediction, and failure time prediction models. It is also necessary to continuously enrich the early warning technology system for brittle failure disasters, such as large-scale rock collapses, to achieve real-time analysis of warning levels, stability status, unstable conditions, and timing of dangerous rock masses, effectively addressing the dual challenges of scientific and accurate prevention and control of large-scale rock collapse disasters and intelligent emergency decision-making. Finally, several development strategies and countermeasures are proposed to overcome the technical bottlenecks in current early monitoring and warning research. For instance, in the field of theoretical research on monitoring and early warning, it is crucial to investigate the causes and early warning mechanisms of brittle destruction disasters such as large-scale collapses. In terms of monitoring equipment development, promoting the development of domestic equipment that integrates dynamic indicators and SDEI is necessary. Building a monitoring and early warning index system based on the integration of multi-source information from SDEI is essential in the research on monitoring and early warning index systems. The development of intelligent monitoring and early warning technology based on cloud-edge integration should be steadily advanced. In terms of data collection for case studies of large-scale dangerous rock mass collapse disasters, comprehensively constructing the SDEI database for various large-scale dangerous rock mass collapse case monitoring samples should be undertaken, ultimately forming a virtuous development cycle of research on theoretical models of large-scale dangerous rock mass collapse disasters, development of monitoring equipment, construction of index systems, development of early warning technology, and database upgrading and improvement. These measures can provide some reference for better responding to large-scale rock collapse disasters in high-risk geological hazard areas.关键词：rock collapse;early warning;large-scale rock;brittle failure disaster;monitoring index system914|1261|0更新时间：2024-11-15
- “黄土高原填方边坡治理新思路：引入黄土-古土壤层状自稳结构，改良黄土，控水结构，为生态治理及灾变防控提供理论参考。”摘要：The increasing number of filling engineering slopes in the loess plateau area over three major projects, “building a city on leveled mountain” “consolidating the ditch and protecting the plateau” and “managing the ditch and creating land” are causing enormous potential geohazard. To provide a new solution for the treatment of the filling slopes in the loess plateau, a loess-paleosol layered self-stabilizing structure is introduced into the disaster prevention and control of loess filling slopes using quicklime and modified cellulose to improve the loess and a new water control structure of loess filling slopes is proposed based on the natural-based solution concept. Through physical model tests, the changes in the seepage field, stress field, and deformation field of filling slopes with or without water control structure are explored, and the final failure mode of slope instability is obtained. The results showed that: 1) compared to the remodeled loess, the improved loess has a better liquid limit, an increased plastic limit, an increased shear strength, a significantly decreased saturated coefficient of permeability, of which the modified cellulose improved loess coefficient of permeability decreased by 88.59% and the lime improved loess coefficient of permeability decreased by 95.18%, and a slower water migration speed in slopes with a water control structure. Besides, the low-permeability layer plays a good effect of water insulation; 2) the existence of anti-erosion surface layer improved the anti-erosion ability of the slope, and the erosion amount of the slope with water control structure was only 28.43% of that without water control structure; 3) the deformation and failure mode of the slope with water control structure was progressive shear slip instability, while the deformation and failure mode of the slope without water control structure was sudden collapse flow slip failure. The results of this study can provide a theoretical reference for ecological management and disaster prevention of filling slopes in the loess plateau.关键词：loess filling slope;loess–paleosol;natural-based solutions (NbS) concept;physical model tests;water control structure860|357|0更新时间：2024-11-15
- “在岩土工程领域，专家建立了岩体抗剪强度参数随钻识别模型，为地质模型精准构建提供可靠方案。”摘要：Objective Accurately identifying stratigraphic structures and shear strength parameters in rock is crucial for ensuring the safety and stability of rock slopes, especially within the context of open-pit mining operations. China, known for its extensive mineral resources and diverse mining activities, places high importance on the safe production of its numerous open-pit mines. The extraction process in these mines generates large volumes of high-steep rock slopes, whose stability is essential for mining safety. Analysis of slope stability requires an understanding of the slope’s structure, the distribution of structural planes, and groundwater distribution along with the precise determination of the shear strength parameters of rock. However, the inherent complexity of natural rock makes it challenging to determine these parameters, a persistent issue in rock mechanics. Additionally, the dynamic nature of open-pit mining means that slope structures continually evolve, and controlling factors alter accordingly. Therefore, dynamically and accurately obtaining the shear strength parameters of rock at various slope locations is paramount. The measurement while drilling (MWD) technique, which installs multiple sensors on a drilling rig to record in-situ drilling parameters in real time, facilitates the inversion of lithology and shear strength parameters of strata. Daily drilling of numerous blasting holes for blasting and extraction produces extensive drilling data in mining operations. Utilizing this data to reflect the shear strength parameters of rock in real time and accurately fulfills the dynamic and in-situ requirements for these parameters, ensuring timely optimization of rock slope bench design and regionalized fine monitoring of slopes. This approach aids in developing more automated, digitized, and intelligent safe mines. However, current research on geological lithology and shear strength parameters inversion based on MWD heavily relies on manual interpretation of depth sequence downhole data from existing geological information or core drilling results, which is difficult to apply widely due to significant human influence on geological information. This reliance poses a risk of missing some weak layers and failing to meet the intelligent sensing and prediction requirements. In addition, monitoring downhole parameters such as thrust and torque relies on indirect hydraulic pressure monitoring, resulting in data that cannot be directly utilized to construct a downhole identification model for rock shear strength parameters in standard units.Methods This study establishes a model for identifying shear strength parameters during drilling using a non-dimensional drillability index. This model allows for the direct recognition of rock shear strength parameters without the need for calibration or conversion of MWD test results. The study uses unsupervised learning and an autonomously designed decision module to conduct intelligent identification and automatic smoothing of depth sequence drilling data. It intelligently identifies geological structures and quantitatively describes each rock layer’s drilling and shear strength parameters. The blasting drill holes in the Heqing Beiya open-pit mine limestone area and the F6-1 fault junction area in Yunnan were monitored employing a self-developed high-precision on-site MWD test apparatus. The comprehensive monitoring results from two representative boreholes demonstrate the application of the intelligent drilling identification method throughout the entire drilling process.Results and Discussions The intelligent drilling identification method accurately identified various rock layers with differing shear strength parameters, cohesion (C), and internal friction angle (ϕ) in the ZK1 and ZK2 boreholes. This included fractured weak rock layers, clayey limestone, and dolomitic limestone layers. Specifically, the initial geological profile of the ZK1 borehole consisted of clayey limestone (ϕ: 38.20°, C: 205.25 kPa) and dolomitic limestone (ϕ: 38.96°, C: 224.83 kPa). The intelligent drilling identification method further pinpointed fractured weak rock layers (ϕ: 35.27°～38.02°, C: 139.82～199.95 kPa), additional clayey limestone layers (ϕ: 37.78°～38.52°, C: 194.39～213.61 kPa), dolomitic limestone layers (ϕ: 38.77°～39.01°, C: 219.81～226.12 kPa), and other rock layers (ϕ: 38.99°～40.13°, C: 227.20～259.44 kPa). Similarly, the original geological profile of the ZK2 borehole included clayey limestone (ϕ: 38.67°, C: 220.84 kPa) and dolomitic limestone (ϕ: 39.22°, C: 232.80 kPa). The method also detected fractured weak rock layers (ϕ: 37.64°, C: 191.85 kPa), clayey limestone layers (ϕ: 38.67°, C: 220.00 kPa), and dolomitic limestone layers (ϕ: 39.38°, C: 236.61 kPa). This method confirmed the presence of layers noted in the original geological profiles and revealed additional rock layers with varying shear strength parameters that were not initially identified. Borehole imaging validated these findings, showing the method’s high accuracy and significant practical value in identifying critical layers absent from original geological profiles. This alignment with large-scale geological profile results provides solid validation for identifying fractured weak rock layers and their strength parameters, enhancing the accuracy of geological model construction. In addition, comparing the shear strength parameters C and ϕ from the initial geological profiles with those identified through this method shows significant variations in the ZK1 borehole, with maximum differences of 65.43 kPa for C and 2.93° for ϕ. In the ZK2 borehole, the maximum differences were 28.99 kPa for C and 1.03° for the internal friction angle. These discrepancies underscore the potential risks associated with inaccuracies in assessing rock mechanics parameters, which can undermine the stability of rock slopes.Conclusions The development and application of the intelligent drilling identification method represent a significant advancement in rock and soil engineering. This method enhances the ability to maintain slope stability in open-pit mines, augmenting the safety and sustainability of mining operations by providing a mechanism to accurately and dynamically evaluate the shear strength parameters of rock layers. Its practical deployment demonstrates the viability of this approach and underscores its potential to transform traditional practices in geological modeling and in-situ testing of shear strength parameters.关键词：rock slope;Measurement While Drilling;drillability index;shear strength parameters;Unsupervised Learning696|409|0更新时间：2024-11-15
- “在危岩体损伤识别领域，专家探究了危岩体空间三维振动特征，建立了高敏感性指标体系，为危岩体损伤识别提供解决方案。”摘要：Objective Currently, most studies on kinetic indices that reflect the damage of dangerous rock masses analyze the vibration characteristics in a single direction. Identifying the damage is challenging when the direction of structural plane evolution is difficult to determine or when the direction of evolution is multidirectional. This study aims to develop a high sensitivity index that can reflect the spatial three-dimensional damage evolution process of the dangerous rock mass. It investigates the spatial vibration characteristics of the damage evolution process, proposing a particle trajectory time-domain dynamics index with heightened sensitivity to the damage of the dangerous rock mass structural plane in various directions of damage evolution.Methods Dangerous rock mass is regarded as a three-dimensional entity. The vibration characteristics of the toppling dangerous rock mass are analyzed through numerical simulations, and formulas for calculating the amplitude and natural frequency of horizontal and vertical shimmy are derived based on the spring-proton shimmy model. The sensitivity of different vibration modes is related to the direction of structural plane evolution. The unidirectional dynamics index sensitivity is determined by the dominant vibration type in this direction, controlling the index sensitivity by the evolution direction of the structural plane. The index sensitivity is highest when the monitoring direction aligns with the normal and evolution directions of the structural plane, and it is lowest for the kinetic indices in the vertical direction to the structural plane evolution direction. Particle trajectory is applied to the damage identification of dangerous rock mass, and the spatial position during vibration is represented by a three-dimensional vector. The three-dimensional data are standardized and synthesized into one dimension, and the time-domain dynamic index algorithm of particle trajectory is defined. Finally, a collapse-similar model test is designed. Using a limestone dangerous rock mass sample, a rock similar material is configured and cast to form a toppling dangerous rock mass similar model. The damage evolution of the structural plane is simulated by deepening the trailing edge fissure through manual cutting, and the three-dimensional vibration data are recorded under two different cutting modes (vertical and oblique cutting). The change rule of the one-direction time-frequency kinetic index and particle trajectory time-domain kinetic index in the evolution process of dangerous rock mass is analyzed.Results and Discussions When comparing the vibration time course curves at the same scale, the X and Z-direction curves gradually thicken with increased cutting depth, indicating that the vibration amplitude of the dangerous rock mass increases. Conversely, the Y-direction curve does not exhibit any notable change. The standard deviation of the vibration amplitude and the kurtosis index increase significantly with the progression of damage in the dangerous rock mass, showing a strong correlation with this damage. During vertical cutting, the relative changes in the kurtosis indices for the X, Y, and Z-directions are 55.66, 2.34, and 8.63, respectively. The kurtosis indices in the X and Z-directions increase exponentially with the deepening of the cracks, whereas the Y-direction shows only a slight fluctuation. The relative changes in the indices for the X, Y, and Z-directions are 0.21, 0.02, and 0.21, respectively. The change in the main frequency in the Y-direction is much smaller than in the X and Z-directions, indicating a considerable difference in the sensitivity of the main frequency among the different directions. The sensitivity of different directions to the main frequency is much smaller than X and Z-directions. The unidirectional kinetic indices are strongly correlated with the damage evolution of the structural plane and show directional differences in the damage sensitivity of the dangerous rock mass. The maximum difference in the sensitivity of the indices across different directions can range from 2.50 to 23.79 times. When the structural plane evolves obliquely along the Y–Z direction, the relative changes in the standard deviation of the Y-direction amplitude and the kurtosis index are 0.57 and 6.45, respectively. The sensitivity of the Y-direction index increases due to the simultaneous decrease in the Y–Z direction constraints during oblique cutting. The time-domain indices of the particle trajectories of the dangerous rock mass under the two cutting methods have been calculated. In Test 1, the standard deviation of the particle trajectory and the kurtosis index shows a clear positive correlation with the damage of the structural plane, with the index increasing faster as the damage deepens. The relative changes in the standard deviation of the particle trajectory and the kurtosis index are 2.85 and 8.62, respectively, while in Test 2, these changes are 2.12 and 68.50.Conclusions The particle trajectory dynamics indices are highly sensitive to the damage of the structural plane of the dangerous rock mass when evolving in different directions. The sensitivity of the particle trajectory index is significantly higher than that of the non-sensitive direction time-domain index, effectively addressing the issue that the unidirectional dynamics index is only sensitive to a specific evolution direction of the structural plane. This characteristic makes it suitable for the damage identification of dangerous rock masses, especially when it is challenging to determine the evolution direction of the structural plane or when the direction is multidirectional.关键词：rock collapse;toppling dangerous rock mass;damage identification;spatial vibration characteristics;kinetic index715|492|0更新时间：2024-11-15
- “在地质灾害领域，专家通过土工离心试验和图像分析技术，建立了斜坡变形模式与宏观形变特征的耦合关系，为反倾层状岩质边坡变形识别和防控提供技术支撑。”摘要：Due to the continuous development of construction projects in China's southwest region, geological disaster issues have become increasingly prominent. Many scholars have employed InSAR remote sensing technology to conduct early identification research in the region. The results are fruitful; however, the research is primarily focused on identifying and determining surface macroscopic deformation. The corresponding relationship between slope internal rupture and surface macroscopic deformation remains unclear. Given the complex geological structure in southwest China, many highly concealed and unpredictable anti-dip layered rock slopes have emerged, posing another challenge for the early identification capabilities of InSAR remote sensing. Hence, this study analyzes the bending-toppling deformation stage of slopes by conducting large-scale geotechnical centrifuge tests. It establishes a coupling relationship between the slope toppling deformation mode and the surface macroscopic deformation characteristics by integrating image analysis technology. The InSAR interpretation and field survey results of the prototype slope are utilized to infer the slope stage reversely, visualizing the slope deformation stage to provide technical support for identifying, preventing, and controlling such slopes. The research shows that: 1) The bending and toppling failure of anti-dip layered rock slopes mainly occurs in four stages: initial creep, toppling-bending, progressive deformation, and fracture zone penetration. Damage typically initiates at the foot of the slope and progressively develops upward. With the development of the stepped fracture surface, the slope experiences comprehensive destruction. 2) Analysis of test image recognition yields rate changes in the deep and surface layers, establishing a corresponding coupling relationship between the two. In different stages of toppling deformation, the large deformation area (surface key cracks) adheres to a developmental law progressing from bottom to top, specifically from the toe of the slope to the middle of the slope to the steep and gentle junction. Two significant peaks in the rate curve of the rock plate at the steep and gentle junction are observed during the onset of toppling deformation and the penetration of the internal fracture zone. 3) The InSAR interpretation results of the Zhayong deformation body are compared to the experimental conclusions, and through on-site reviews, it is preliminarily concluded that the deformation body is in the progressive stage of deformation.关键词：anti-dip stratiform rocky slope;centrifuge model test;early identification;macroscopic deformation characteristics714|599|0更新时间：2024-11-15
NEW TECHNOLOGY FOR SENSING ROCK SLOPE DISASTERS
- “在超高压大电流领域，专家构建了气体绝缘母线多场耦合模型，探究了其温升特性，为设备设计及监测提供参考。”摘要：Objective This study focuses on the temperature rise characteristics of ultra-high voltage, high-current gas-insulated busbars. It investigates the temperature and flow rate distribution of the busbar under 1.1 times the rated current (8 800 A) and 0.3 MPa gas pressure. Additionally, a high-current temperature rise test platform is built to verify the model's validity. Based on this, the study examines the influence of the through-current level, the SF6 gas pressure, and the contact state on the temperature distribution of the busbar.Methods In this study, the 550 kV gas-insulated busbar is taken as the research subject. The contact resistance is calculated through direct current (DC) low resistance measurement. Electrical and thermal conductivities are the determined using the resistance formula and the Wiedemann-Franz law. These calculations are incorporated into the development of a three-dimensional electromagnetic-temperature-fluid multi-field coupling model of the busbar. The temperature and flow velocity distribution under 1.1 times the rated current (8 800 A) and 0.3 MPa air pressure are calculated, considering factors such as contact resistance, skin effect, natural convection, and thermal radiation. The temperature differences between the busbar conductor, the maximum hot spot, and the shell are analyzed and compared with the temperature difference at the axial section where the basin convex and concave contacts are located. A high-current temperature rise test platform is built to validate the model’s results, and the test measurements align closely with the simulation results. Based on the reliable model verified by the test, the influence of the through-current level, SF6 air pressure, and contact state on the temperature distribution of the busbar is further investigated. Additonally, a variation diagram showing the maximum hot spot temperature of the busbar with each factor is obtained.Results and Discussions To calculate the bus temperature distribution, it is first necessary to determine its loss size. Considering the skin depth of the conductive rod and basin insert, the total loss of the bus is calculated to be 695.786 W. The primary heat sources are the conductor rod and electrical contact components. The maximum magnetic induction intensity occurs at the basin insert edge, and the current density is mainly concentrated on the surface of the conductor. This loss is mapped to the temperature field to calculate the bus temperature and gas flow rate distribution. The rise in bus temperature is mainly concentrated in the through-flow conductor components. The temperatures at both ends of the contact are significantly higher than at the middle of the guide rod. The highest temperature of the bus, approximately 54 °C, is located at the concave surface of the basin at the top of the electrical contact part. This temperature is about 15 °C higher than the SF6 temperature at the end of the contact. The temperature rise in the SF6 is mainly concentrated in the upper part of the cavity. Additionally, the radial temperature gradient of the contact at the concave surface of the busbar is significantly higher than that at the convex surface. Hot air buildup around the shell at the concave surface of the basin is noticeable, showing a radial pattern of higher temperature at the top and lower temperatures at the bottom, with a left–right symmetrical distribution pattern. The location of the busbar conductor hot spot corresponds to the shell, with a clear difference in the temperature gradient at different x-axis locations. The temperature gradient difference between the conductor and the heat near the shell causes the internal SF6 gas and air to move upward. The external air flow velocity is approximately 0.23 ms−1, creating a circulation along the boundary of the solution domain due to non-isothermal natural convection. The influence of different factors on the bus temperature distribution law is further explored through simulation. As the load current increases, the highest temperature trend for the conductor and electrical contact is almost identical and significantly increases, while the shell temperature increase is much smaller than that of the through-flow conductor. The three highest temperatures rise with the current increase nonlinearly; SF6 gas pressure per 0.05 MPa shows a roughly linear decreasing trend for both bus conductor and shell temperatures, with the conductor temperature being more sensitive to changes in SF6 gas pressure than the shell. The abnormal electrical contact temperature increases approximately linearly with the resistance value. The normal electrical contact temperature rise is about 15 ℃. Even when the contact resistance value reaches approximately 70 μΩ the highest shell temperature exceeds the normal temperature of the electrical contact. Finally, the high-current temperature rise test platform validates the constructed model, providing data support for the temperature rise characterization discussed in this study.Conclusions This research delineates the temperature rise characteristics of the electromagnetic-thermal-fluid multi-field coupling model of a 550 kV gas-insulated busbar, considering the influence of contact resistance on the thermal characteristics. It analyzes the temperature rise characteristics under high-current conditions and investigates the influence of current level, SF6 pressure, and contact state on the temperature distribution of the busbar. These findings have considerable engineering implications, offering valuable references for product design and condition monitoring of gas-insulated equipment, and helping to reduce the occurrence of overheating faults.关键词：gas-insulated busbar;multi-field coupling;temperature rise characteristics;contact resistance;influencing factors148|608|0更新时间：2024-11-15
- “在环保气体断路器领域，专家基于40.5kV瓷柱式断路器样机，探究了C4F7N/CO2混合气体的温升特性，为新型环保绝缘气体替代SF6提供理论依据。”摘要：Objective Due to the significant greenhouse effect of SF6 circuit breakers, research on circuit breakers using new environmentally friendly gases has garnered considerable attention. Existing studies on the C4F7N/CO2 gas mixture primarily focus on arc extinguishing and insulation performance, yet investigations into its temperature rise characteristics remain insufficient. However, these characteristics are crucial for designing the current-carrying capacity and monitoring the operational condition of the circuit breaker. Therefore, this study investigates the temperature rise characteristics of environmentally friendly gas circuit breakers based on a 40.5 kV porcelain column circuit breaker prototype.Methods The internal structure of the interrupter chamber in the porcelain column circuit breaker is disassembled to analyze the current path and the mechanisms of heat generation and transfer. The thermal process primarily occurs inside the interrupter chamber, with the primary heat sources being the conductor circuit and the contact points between the movable and static contacts. Based on this, a temperature rise experimental platform is constructed. The arrangement of temperature sensors and the testing scheme are established, providing a data foundation for further studies on temperature rise characteristics. This research develops a simulation model of electromagnetic-thermal-fluid multi-physical field coupling based on the experimental prototype at a 1∶1 scale to explore the temperature rise mechanism. It employs the finite element method to calculate the temperature rise and fluid field distribution inside the interrupter chamber. The simulation’s material parameters and boundary conditions are consistent with those in the experiment.Results and Discussions The experimental and simulation temperature rise curves are closely aligned, with the temperature rise error at designated measurement points, such as the primary contact, arc contact, and nozzle, not exceeding 10%. Thus, the experiment validates the simulation model. This simulation provides a method for studying the overall temperature rise and flow field distribution of the circuit breaker, addressing the limitation of experimental measurements to only certain parts of the circuit breaker. By integrating experiments and simulations, this study also identifies the significant heating at the contact surface between the primary and arc contacts due to high current and power density. This is attributable to the small cross-section of the load current and high film resistance at the contact surface. This study indicates the temperature rise characteristics and flow field distribution of the new environmentally friendly gas circuit breaker using two analytical methods: experimental testing and simulation calculation. It also analyzes the temperature rise field and flow field distribution characteristics of the 40.5 kV porcelain column circuit breaker regarding current magnitude, insulating gas type, and gas mixture components.Conclusions The results showed that the temperature rise within the circuit breaker interrupter is symmetrical, exhibiting a step-like distribution with higher temperatures at the top and lower temperatures at the bottom. The heat is primarily concentrated in the conductor and the top of the internal insulating gas. The flow field distribution inside the interrupter remains relatively stable, with the gas near the conductor experiencing natural convection due to heat. The gas flow rate increases with the temperature, reaching its maximum at the nozzle. Under certain conditions, the temperature rise in the conductor circuit of the porcelain column circuit breaker, the porcelain casing shell, the nozzle, and other typical measurement points increases with the rise in load current. The temperature rise rate at the nozzle and conductor is significantly higher than at the porcelain casing shell. The temperature rise field of the porcelain column circuit breaker, when utilizing different gases, follows a similar distribution law. The C4F7N/CO2 gas mixture shows good heat dissipation properties. As the proportion of C4F7N gas in the mixture increases, the temperature rise of the conductor gradually decreases. The flow field distribution inside the interrupter chamber is also similar across various gas environments. The gas flow velocity is closely related to the interrupter’s temperature and the gas’s viscosity. The results of this study provide a theoretical basis for applying the new environmentally friendly gas C4F7N in porcelain column circuit breakers. This is of great significance for replacing SF6 with new environmentally friendly insulating gases in managing the temperature rise of high-voltage circuit breakers.关键词：temperature rise characteristics;C4F7N/CO2 mixed gas;Porcelain column circuit breakers;finite element simulation.114|568|0更新时间：2024-11-15
-
Influence Mechanism of Micro Grain-boundary Parameters to the Electrical Properties of ZnO Varistor AI导读
“在电力设备保护领域，研究人员基于Voronoi网络及改进的晶界分区模型，研究了ZnO压敏电阻微观晶界特性对其宏观电气性能的影响规律，为高性能ZnO压敏电阻研制提供重要参考。”摘要：ZnO varistors exhibit excellent nonlinear voltage–current characteristics and substantial energy capacity, enabling them to absorb impulse energy during power system overvoltages and thus protect the electrical equipment. The nonlinear behavior of ZnO varistors is attributed to the double Schottky barrier structure at their grain boundaries. The parameters of these micro-grain boundaries play a decisive role in determining the macroscopic electrical performance of ZnO varistors. However, few studies have investigated the influence of micro grain-boundary structural parameters on the macroscopic electrical properties of ZnO varistors. This study utilizes a Voronoi network and an improved grain boundary partitioning model to simulate and calculate how micro grain-boundary parameters, such as grain donor density, grain boundary surface state density, and grain boundary partitioning parameters, affect the macroscopic electrical properties of ZnO varistors. The development process of high-performance ZnO varistors is considered a multi-variable and multi-objective problem. The optimization targets and variables are classified based on whether any optimization variable has similar influences on the optimization targets and whether the variable affects both types of optimization targets equally. This classification reveals the influence mechanism of micro grain boundaries on electrical properties. By reasonably classifying optimization variables and objectives, this study simplifies complex multi-variable and multi-objective problems. It formulates step-by-step optimization strategies based on the characteristics of classified variables and objectives, thus improving the performance of ZnO varistors. This approach considerably enhances varistor performance at the microscopic physical layer, which is crucial for the development of high-performance varistors.关键词：ZnO varistor;materials computation;double Schottky barrier;grain boundary partitioning;influence mechanism98|461|0更新时间：2024-11-15 - “最新研究揭示C6F12O/CO2混合气体与EPDM材料的长期相容性，为环保绝缘气体在电气设备中的应用提供重要参考。”摘要：Objective SF6 is widely used in electrically insulated equipment due to its excellent insulating properties. However, its global warming potential is 23 500 times greater than that of CO2, with an atmospheric lifespan of up to 3200 years. As the severity of global warming intensifies, finding a new environmentally friendly insulating gas to reduce the using of SF6 has become a hot research topic in recent years. C6F12O has both environmental benign characteristics and insulating properties, which has attracted the attention of scholars. However, the liquefaction temperature of pure C6F12O gas is 49 ℃ at atmospheric pressure, which is difficult to use as an insulating medium in practical engineering scenarios. It must be mixed with low-boiling point buffer gases, including CO2 and N2. It has been shown that a gas mixture comprising 4% C6F12O/CO2, when pressurized above 0.14 MPa, can be used as an insulating gas in 10 kV switchgear. For new eco-friendly insulation gas, its compatibility with the internal sealing materials of the equipment is also need to be considered.Method The experimental study on the compatibility between the C6F12O/CO2 gas mixture and the sealing rubber EPDM was carried out by building a thermally accelerated compatibility experimental platform in this paper. The effect of trace O2 on the compatibility of C6F12O gas mixture and EPDM was investigated by adding a small amount of O2 into the gas mixture. The changes in the surface morphology, the elements of rubber samples, the gas mixture components, and the mechanical properties were analyzed before and after the experiments. The interaction between C6F12O/CO2/O2 gas mixture and rubber was also calculated based on the MD theory.Results and Discussions The results from scanning electron microscopy showed that the scaly structures appeared on the surface of EPDM when mixed with C6F12O/CO2 at 110 °C. A flocculent structure appeared when EPDM was mixed with C6F12O/CO2/O2 at only 90 °C. This observation suggested that the inclusion of O2 promotes the interaction between the C6F12O/CO2 gas mixture and EPDM, which in turn leads to the corrosion of EPDM. The results of X-ray photoelectron spectroscopy showed that a decrease in the carbon content alongside increases in oxygen and fluorine within the C6F12O/CO2 mixed EPDM with increasing temperature. This trend indicated that C6F12O/CO2 reacts with EPDM. In group of C6F12O/CO2/O2, the changes were more significant, which indicate that O2 promoted the reaction between the C6F12O/CO2 mixture and EPDM. Post-experimental analyses identified new spectral peaks at 284.5 eV, 285.8 eV and 289.2 eV, corresponding to C—C, C—H and —COO— ester bonds, respectively. Three characteristic peaks of elemental F were detected on the EPDM surface for metal fluoride at 683 eV, organofluoride at 684.5 eV, and —CF2CH2— at 687 eV. The intensity of C characteristic peak was significantly lower, while the intensity of O and F were higher in the C6F12O/CO2/O2 gas mixture after testing. These findings indicated that O2 actively participates in the reaction between the C6F12O/CO2 mixture and the EPDM, fostering an intensified interaction. The results of tensile properties showed that the mechanical properties of EPDM decreased significantly with increasing temperature. The mechanical property of EPDM in C6F12O/CO2 atmosphere was slightly worse than those of the original samples. The mechanical property of EPDM under C6F12O/CO2/O2 was worse than that of C6F12O/CO2. The reaction between the C6F12O/CO2 mixture and EPDM was more intensive after adding O2, leading to a significant decrease in the mechanical properties of EPDM. At the end of the experiment, a small amount of the resultant gas was collected and subjected to Gas chromatography–mass spectrometry for further analysis. The small amount of H2O remaining inside the gas tank was involved in the decomposition reaction of the C6F12O gas mixture. Throughout the reaction,
${\rm H}^{\,\cdot} $ ${\rm OH}^{\,\cdot} $ ${\rm CF}_3^{\,\cdot} $ ${\rm C_{3}F}_7^{\,\cdot} $ ${\rm F}^{\,\cdot} $ $\chi $ $\chi $ $\chi $ $\chi $ $\chi $ $\chi $ 关键词：C6F12O gas mixture;EPDM rubber;compatibility;molecular dynamics simulation112|355|0更新时间：2024-11-15 - “在直流电压下气-固界面电荷积聚现象研究中，专家提出了改进计算方法，并通过实验验证了其准确性，为直流输电设备设计提供理论支持。”摘要：Free charge accumulation on the surface of solid insulation under direct current (DC) voltage distorts the electric field distribution on the insulator’s surface, affecting discharge behavior, such as creeping flashover. Numerous experiments and simulation studies have been conducted by scholars worldwide. Current research shows that there are three primary sources of surface charge on gas–solid insulation: bulk current on the solid side, ionic current on the gas side, and surface current on the gas–solid interface. In GIS, GIL, and other power equipment, the field strength of each critical component is low, and no partial discharge occurs during normal operation; hence, the source of gas-side charge is mainly the weak ionization of gas molecules. However, oscillating solutions occur in the discrete form of the convection-diffusion equation in the gas field, which leads to non-convergence of iteration. This study proposes, for the first time, an improved saturated current density calculation method for charge accumulation at the gas–solid interface of insulators. The surface charge accumulation characteristics of post-insulators in air under DC voltage at normal temperature and pressure are measured. The rationality and accuracy of the simulation calculation method are verified by comparing and analyzing the experimental results. The results indicate that there are two different polarity potential peaks and one homopolar potential peak on the surface of the post-insulator under DC voltage. The position of the potential peak remains unchanged with the increase in voltage application time. At the lower part of the sample, the gas-side current is less than the solid-side current density, resulting in the accumulation of negative charges on the surface. In contrast, at the upper part of the sample, the gas-side current exceeds the solid-side current density, and the surface also accumulates negative charges. In the middle part of the sample, the insulator surface area gathers positive charges, driven by the solid-side charge. This research can provide theoretical support for the design of DC transmission equipment.关键词：surface charge;saturation current;HVDC systems;post insulator;optimization calculation166|454|0更新时间：2024-11-15
- “C4F7N作为环保绝缘气体，替代SF6，研究O2对C4F7N/N2混合气体击穿特性的影响，为环保绝缘气体研究提供新思路。”摘要：Objective In the field of electrical equipment insulation, the application of gas insulation is becoming increasingly prevalent. SF6 gas has been extensively utilised due to its stable chemical properties and its excellent insulation and arc-quenching capabilities. Nevertheless, SF6 is a powerful greenhouse gas with the potential to persist in the atmosphere, underscoring the urgency of identifying a suitable replacement. C4F7N, with its superior insulation properties and environmental benefits, has the potential to be developed as an eco-friendly insulating gas that could replace SF6. Owing to the elevated liquefaction temperature of C4F7N, it is frequently employed in engineering applications in conjunction with other gases including CO2 and N2. At present, CO2 is predominantly employed as a buffer gas for C4F7N, typically constituting in excess of 90% of the mixture. Nevertheless, as CO2 remains a greenhouse gas with a substantial greenhouse effect, the utilisation of N2 as a buffer gas is egarded as a means of markedly enhancing the environmental benignity of the mixture. Nevertheless, the C4F7N/N2 mixture may undergo decomposition during discharge, which could result in the formation of solid carbon deposits on metal electrodes. This phenomenon has the potential to significantly impair the normal operation of gas-insulated equipment. Some researchers have put forth the suggestion of incorporating oxygen as a second buffer gas into C4F7N mixtures with the aim of enhancing the arc-quenching performance of binary mixtures. The incorporation of O2 can facilitate the breakdown characteristics of C4F7N/N2 binary mixtures, while simultaneously inhibiting the formation of decomposition products. Nevertheless, the microscopic precise mechanism by which O2 exerts its influence on the breakdown and decomposition characteristics of C4F7N/N2 mixtures remains unclear. The objective of this study is to investigate the microscopic mechanism by which oxygen influences the breakdown and decomposition characteristics of C4F7N/N2 binary mixtures.Methods This study employs density functional theory to determine the electrostatic potentials of the gas molecules present in C4F7N/N2/O2 mixtures. The electron affinity and electronegativity of O2 are calculated, as are those of the C4F7N/N2 mixture. Molecular dynamics methods based on the principle of energy optimisation are employed to derive the structural formulas of N2/O2 and C4F7N with varying molecular quantities (1, 2, 3, and 4 molecules). The interaction energy and charge transfer between varying quantities of N2 and O2 with C4F7N gas are calculated. The Fukui function of the C4F7N molecule is calculated based on Hirshfeld atomic charges, and the bond dissociation energies of C4F7N/N2/O2 gas molecules are determined. The quantum chemistry calculations employ the B3LYP functional and the 6–G(d,p) basis set, while the molecular dynamics simulations utilise the COMPASS force field. The results of the calculations are used to analyze the radical reaction processes involved in the discharge decomposition of C4F7N with O2. This analysis investigates the impact of O2 on the breakdown and decomposition characteristics of C4F7N/N2 mixtures from a microscopic perspective.Results and Discussions The findings demonstrate that O2 exhibits a greater electron affinity and electronegativity in comparison to N2. The incorporation of O2 results in a reduction of free electrons within the gaseous environment, thereby diminishing the likelihood of collision ionisation during the discharge process. This in turn, gives rise to an enhancement in the breakdown voltage of C4F7N/N2 mixtures. In the case of N2/O2 gases, it can be observed that as the number of molecules present increases, the interaction energy also increases approximately in proportion. The interaction energy between O2 and C4F7N is approximately five times that between N2 and C4F7N. In the C4F7N/O2 molecular system, the C4F7N molecule is observed to a negative charge, indicating that electrons are gained by C4F7N within the system. As the number of molecules increases, C4F7N gains more electrons, resulting in an overall increase in charge transfer. In the C4F7N/N2 system, the C4F7N molecule is observed to carry a positive charge, indicating that electrons are lost by C4F7N within the system. As the number of molecules increases, C4F7N loses electrons in greater quantities. In the absence of discharge decomposition, the interactions between the buffer gas and C4F7N molecules are predominantly physical in nature. During the discharge process, O2 provides a substantial number of oxygen radicals. Once the chemical bonds of C4F7N have been broken, forming small molecular radicals, these react with
${\rm O}^{\,\cdot}$ ${\rm CF}_2^{\,\cdot} $ ${\rm C}^{\,\cdot} $ 关键词：C4F7N/N2 mixed gas;density functional theory;O2;breakdown characteristics;decomposition characteristics71|313|0更新时间：2024-11-15
ENVIRONMENTALLY FRIENDLY POWER EQUIPMENTS AND ADVANCED ELECTRICAL MATERIALS
-
Analysis of Parameters Influencing Swimming Ability of Crucian Carp Based on Numerical Simulation AI导读
“最新研究发现，鲫鱼游泳能力受摆动频率、摆幅系数、厚长比和体长影响，壁面效应也会影响游泳速度，为鱼类仿生学研究提供参考。”摘要：The mechanisms of fish swimming behavior and the factors influencing swimming ability are pivotal in current research on fish bionomics. The Crucian carp, a typical Carangiform swimmer, is selected as the subject of this study. A numerical simulation method, dynamic grid technology, and User–Defined Functions (UDF) are employed to investigate the effects of motion parameters, fish body size, and swimming environment (wall boundary) on the swimming ability of Crucian carp. This study measures the swimming ability by the time it takes for the bionic carp to achieve a dynamic and stable cruising speed. The findings indicate that when the fluctuation frequency ranges from 0.07 to 5.50, and the amplitude factor is between 0.03 and 0.06, an increase in frequency and amplitude results in a reduction in the time required for the bionic carp to reach a dynamic stable cruising velocity, which also increases the cruising velocity. Dimensionless numerical calculations are performed on Crucian carp with varying thickness-to-length ratios and body lengths. It is determined that body size is inversely proportional to swimming velocity when swimming forward with consistent swing amplitude and frequency. A higher cruising velocity is achieved when the thickness-to-length ratio is less than 0.166. Regarding body length L, juvenile carp can reach the cruising stage more rapidly but exhibit a lower cruising velocity of about 0.6L s–1. The speed of the bionic young fish is more pronounced during the acceleration stage, and the cruising stage mirrors the dynamic stable cruising velocity of the bionic adult fish, which is approximately 1L s–1. The presence of walls near which the fish swim can diminish the swimming velocity. The impact of bilateral walls on the swimming velocity of bionic carp is more substantial than that of unilateral walls; when the fish body is 0.5L from a unilateral wall and 0.8L from bilateral walls, it is not affected by the wall. These results can provide a valuable reference for research in fish bionomics.关键词：crucian carp;self-propelled swimming;influencing parameters;wall effect;numerical simulation118|755|0更新时间：2024-11-15 - “在城市化进程中，空气污染问题日益严重。专家基于低秩矩阵分解方法，提出了一种融合时空特征的空气质量推断模型，有效提升了无站点区域空气质量的推断性能。”摘要：With rapid urbanization, air pollution has become increasingly severe, making the provision of a spatio-temporal fine-grained air quality distribution essential to support outdoor planning and promote good health. However, the sparseness of air quality stations, the incompleteness of related feature data, and the nonlinear variation of air quality across locations and times pose substantial challenges for accurately inferring air quality in unobserved areas. This study proposes a matrix factorization-based approach to infer air quality by analyzing a real air quality dataset and discovering the low-rank structure of the air quality matrix. This approach fuses knowledge from the low-rank structure, air quality measurements, and various spatio-temporal features. Unlike existing works that address feature recovery, feature extraction, and air quality inference separately, this study unifies these three tasks into a single model. Such integration allows for improved inference performance through the collaborative training and supervision of different tasks. In this model, spatial and temporal feature matrices and the air quality matrix are constructed and collaboratively factorized into spatial and temporal feature representations. By sharing spatio-temporal matrix factors with the air quality matrix, the similarity knowledge of spatial and temporal features is transferred into air quality inference to enhance its performance. The proposed model is evaluated using real data sources obtained in Beijing city. Comparison results with baseline models demonstrate that the proposed model surpasses these models in various metrics, such as inference error and standard deviation, and achieves a better FAC2 result. Additionally, the model effectively reveals the principal spatial and temporal features to a certain extent.关键词：Spatial-temporal feature;matrix factorization;air quality inference;low-rank structure120|366|0更新时间：2024-11-15
- “双腿轮式机器人研究取得新进展，提出了分布式动力学建模策略和全身力矩控制框架，为机器人运动控制研究提供新理论框架和实践指导。”摘要：Objective The two-legged wheeled robot is a new type of composite ground mobile robot constituted by designing wheels at the end of the legs of footed robots, which combines the improved maneuverability and flexibility of traditional wheeled mobile platforms and footed robotic structures and has enhanced application scenarios and research value. However, the motion control of the two-legged wheeled robot is highly dependent on an accurate dynamics model. At the same time, it has control difficulties such as underdrive, strong coupling, and nonlinearity. These challenges lead to difficulties in achieving effective motion control and self-balancing of the robot. Therefore, proposing an efficient method for the motion control of a two-legged wheeled robot is of great practical significance.Methods This study adopts the idea of a distributed model to establish the dynamics model of the leg-wheel and torso subsystems and address the issue with the huge structure of the overall dynamics model of the two-legged wheeled robot, which is not conducive to characterization and the construction of the controller. These models retain all the dynamics characteristics of the robot and connect them to the inter-module motion/force transfer relationship to complete the whole-body dynamics with the wheel-leg-torso interaction force as the end output force model. Thereafter, the joint moment solver, with the end output force as the task space, is constructed based on this model. It includes feed-forward compensation of the rod inertia force caused by the initial state quantity performed by the observation signal. Then, a distributed control framework with torso position as the task space is proposed to plan the torso joint force and joint moment hierarchically, and the walking motion control architecture based on the whole-body moment is constructed. Based on this walking motion control architecture, an adaptive planning method for the longitudinal trajectory of the torso in jumping motion is proposed, giving the torso the longitudinal velocity required for jumping by planning the longitudinal motion of the torso. To ensure the smoothness of the velocity and acceleration of the jumping motion trajectory, a cubic polynomial is utilized to plan the torso height trajectory of the jumping support phase, the torso height trajectory of the jumping airborne phase is obtained by integrating the velocities, and the variation of its torso height trajectory is presented. Furthermore, the control method of the airborne phase based on the virtual model and the momentum moment theorem is proposed when the torso reaches the jumping speed by controlling the contraction of the legs to detach from the contact between the wheel and the ground and enter into the free-fall state, and the whole jumping process is presented. Finally, the effectiveness of the whole-body moment control system is proved through simulation experiments as well as prototype tests.Results and Discussions A simulation experiment platform is developed using MATLAB to demonstrate the feasibility of the distributed dynamics modeling and whole-body moment control framework for the two-legged wheeled robot. Two walking experiment scenarios, including straight-line and circular motions, are established, along with three jumping simulation experiments at different heights. The velocity following the curve for straight-line motion is shown, with a maximum error of no more than 0.09 m/s. The pitch angle and height following curves of the robot are displayed, indicating height errors ranging from −0.011 m to 0.002 m and pitch angle errors from −0.001 rad to 0.006 rad. For circular motion, the velocity following the curve is displayed with a maximum error of 0.071 m/s. The pitch angle and height following curves for circular motion indicate a maximum height error of 8.633 mm and a maximum pitch angle error of 0.006 rad. The jumping heights are 0.55 m, 0.50 m, and 0.45 m, where the jumping results show a maximum jumping height error of no more than 0.02 m, confirming the feasibility of the distributed dynamics modeling and whole-body moment control framework proposed in this study. A prototype of the two-legged wheeled robot is developed, and multiple motion mode experiments are conducted. The pitch angle and height following curves for the prototype indicate a trunk height error of no more than 3.38 mm and a pitch angle error of no more than 0.04 rad. Finally, by comparing and analyzing the simulation experiment results and prototype experiments, it is found that coordinated wheel and leg movements can achieve higher motion following accuracy while maintaining dynamic balance based on the trunk as the task space. This confirms the correctness of the distributed dynamics modeling and whole-body moment control framework proposed in this study, providing an effective reference for the study of motion control of two-legged wheeled robots.Conclusions This study proposes a distributed whole-body dynamics modeling method to establish the dynamics models of the torso subsystem and the leg-wheel subsystem and realize the complete mapping of the robot’s individual joint moments to the end output force to preserve the dynamic characteristics of the two-legged wheeled robot. A whole-body moment control framework is proposed based on the distributed whole-body dynamics model to achieve motion control and dynamic balance targeting the two-legged wheeled robot with the characteristics of instability, strong coupling, and nonlinearity. A multi-motion mode planning method is proposed based on the whole-body moment control framework to improve the motion performance of the two-legged wheeled robot. Finally, a simulation experiment platform is built to carry out the simulation experiments of jumping and walking motion modes to verify the feasibility of the whole-body moment control system based on the distributed dynamics model. A prototype of a two-legged wheeled robot based on synchronous belt-driven tandem legs and a carbon fiber plate frame is developed and tested in multiple motion modes to prove the correctness of the distributed dynamics model and the whole-body moment control framework of the two-legged wheeled robot.关键词：Two-legged wheeled robot;Distributed modelling;Torque control;motion control336|1026|0更新时间：2024-11-15
INTELLIGENCE INTERDISCIPLINARY SCIENCE AND ENGINEERING
- “在磷矿采空区失稳分析领域，专家构建了随机场模型，实现了失稳风险的定量评价，为矿柱稳定性研究提供了新方法。”摘要：Goaf instability accidents such as roof falling, rib spalling, or pillar spalling often occur during the room-pillar mining process in layered phosphate deposits, severely affecting mine production safety. Considering that the spatial variability of rock mass mechanical parameters is rarely incorporated into goaf instability analysis methods, a database for rock mass matrix cohesion (c) and internal friction angle (ϕj) is first established in the Shanshuya phosphate deposit using the displacement inversion method. Then, the spatial variability characteristic parameters of c and ϕj are fitted based on geostatistical principles. Random fields are constructed using the Karhunen–Loeve (K–L) series expansion method and assigned to the FLAC3D grid model through element traversal to represent the autocorrelation, cross-correlation, and non-Gaussian properties of c and ϕj during numerical calculations. The limit state equation for rock mass failure is then established, using layered rock mass’s failure approach index (FAI) as the evaluation index. The Hermite random polynomial coefficients between the input field of c and ϕj and the output field of the FAI are solved by the probability collocation method, establishing an explicit function expression between the probability collocation point and the FAI. Finally, the FAI for each element is calculated using the cross-correlation standard random variables instead of the probabilistic collocation points. Hence, the element failure probability of the phosphate mine goaf is determined by the Monte Carlo method, which reveals a quantitative evaluation of the goaf instability risk. The results showed that the random field assigned to the FLAC3D grid model effectively characterizes the spatial variability of rock mass mechanical parameters. The high instability risk area of the goaf (failure probability ≥ 70%) is located in the middle of the pillar within 2 meters, consistent with the actual pillar failure location and depth. Compared to conventional numerical calculations, the mechanical parameters of layered rock mass in this study exhibit objective spatial variability characteristics such as autocorrelation, cross-correlation, and non-Gaussian properties, and the calculated failure probability achieves a more specific and comprehensive quantitative evaluation of the instability risk of phosphate mine goaf. This addresses the difficulty in determining whether the phosphate mine goaf is unstable due to the varying distribution of maximum displacement, maximum and minimum principal stresses, and the FAI. The research results can provide a technical reference for stability analysis and support optimization of similar underground engineering projects in layered rock mass.关键词：layered rock mass;phosphate deposit;goaf;risk of instability;quantitative evaluation91|396|0更新时间：2024-11-15
- “在地下洞室群地应力场研究领域，专家提出了融合逐步多元线性回归和神经网络的联合反演方法，为预测地下洞室群区域地应力场提供解决方案。”摘要：Deep valleys and large structural planes considerably affect the distribution of the in-situ stress field in large underground caverns. Accurate prediction of the in-situ stress field in underground caverns is crucial for engineering construction. The inversion of the in-situ stress field based on measured data is the primary method for predicting this field. Existing inversion methods are predominantly based on multiple linear regression or neural network approaches, which mainly consider factors such as gravity and tectonic movements that influence the in-situ stress field. The multivariate regression method offers simplicity, rapidity, and a unique solution; however, it occasionally fails to account for the physical significance of tectonic movement factors. This method may present issues such as negative regression coefficients for factors, strong collinearity among factors, or insignificant effects. The neural network method improves the fitting degree through its nonlinear relationship between the input and output layers, but determining the reference values for each factor can be challenging. In order to address these issues, this study proposes a joint inversion approach that combines stepwise multiple linear regression and neural networks. This approach obtains reasonable tectonic movement factors and their reference values through stepwise multiple regression, subsequently improving inversion results using neural network inversion. The study also introduces an inversion method for the in-situ stress field of underground cavern groups under complex geological conditions. The method includes analyzing measured in-situ stress data, establishing a three-dimensional geological model, performing joint inversion using stepwise multiple linear regression and BP artificial neural network, and verifying results based on the stress-type failure characteristics of the surrounding rock. Starting from the historical tectonic movements, this method constrains the multiple regression factors and their coefficients, addresses excessive shear stress deviation, and improves the accuracy of in-situ stress inversion results through multi-source constraints and verification. This includes historical geological structure analysis, measured in-situ stress analysis, and the stress-type failure characteristics of the cavern group. The method is applied to the underground cavern group of Yebatan Hydropower Station. The study results indicated that the maximum principal stress of the cavern group is primarily between 25 and 30 MPa, oriented NWW–EW, with a gently inclined valley. Upon excavation, the upstream side spandrel and downstream side wall foot are susceptible to stress-type failure. Large structural planes, such as faults, significantly impact the distribution of the in-situ stress. Near faults, the maximum principal stress value decreases, with the direction aligning closely with the vertical fault. The stress value rapidly increases to the original rock stress within 5～10 m from the fault, with orientation gradually changing to near parallel with the fault, approaching the direction of the macroscopic principal stress. Densely developed faults are the main reason for the significant variability in principal stress within the underground cavern group of the Yebatan Hydropower Station. This study clarifies the distribution of the in-situ stress field in underground caverns under the combined influence of deep valleys and multiple structural planes, especially the variation characteristics near faults. It provides the basis for predicting stress concentration areas and potential stress-type disaster sites in underground caverns.关键词：in-situ stress inversion;large underground cavern groups;numerical simulation; intelligent inversion;high geo-stress;deep valley;faults142|570|0更新时间：2024-11-15
DEEP UNDERGROUND SCIENCE AND ENGINEERING
- “在中小型边坡治理领域，专家建立了考虑土体塑性变形的锚索微型桩群计算模型，为控制桩身变形与内力提供解决方案。”摘要：Anchor cable micro-pile groups are increasingly utilized in small and medium-sized slope treatment and slope emergency rescue engineering. However, existing calculation models rarely consider the plastic deformation of the soil in front of the embedded section of the pile and the continuity of the deformation of the soil on the side of the pile. To address this issue, the characteristics of the anchor cable micro-pile group in strengthening soil slopes are examined, and the micro-pile group is divided into a loaded section and an embedded section based on the location of the potential sliding surface. For the loaded section, the soil between piles is simplified as a horizontal soil spring, which effectively accounts for the transfer of soil force between piles. For the embedded section, the ideal elastic–plastic p–y curve is employed to analyze the interaction between the pile and the soil. The prestressed anchor cable at the crown beam is simplified as an elastic support. Its force is transferred to the pile top of the back row of piles and then redistributed to the pile tops of each row of piles by the crown beam. This arrangement achieves the constraint effect on pile top displacement. Finally, based on the Pasternak two-parameter foundation beam method, a calculation model for the anchor cable micro-pile group is established, considering the elastoplastic deformation of soil in front of the embedded pile and the continuity of soil deformation around the pile. The finite difference method is utilized to solve the calculation model, and the horizontal displacement and internal forces of the structure are obtained. The calculation model is utilized to evaluate examples of soil slope reinforcement with micro-pile groups and anchor cable micro-pile groups, respectively. The study results are compared to both experimental and numerical simulation ones to verify the model's rationality. The effects of soil shear stiffness, the number of anchor cables, and pile diameter on the horizontal displacement and internal forces of the micro-pile group are discussed. The research demonstrates that the plastic deformation of the soil before the embedded section of the pile and the continuity of the deformation of the soil on the side of the pile significantly influence the displacement and internal forces of the micro-pile. As the number of anchor cable bundles increases, the maximum positive bending moment and horizontal displacement of the pile body continuously decrease. The goal of controlling pile deformation and internal forces can be achieved by adding anchor cable bundles.关键词：slope;micro-pile;anchor cable;internal force and deformation;ideal elastic-plastic p–y curve133|423|0更新时间：2024-11-15
- “在隧道工程领域，专家基于能量方法，提出了考虑地层损失率差异的双线下穿开挖作用下既有隧道沉降的计算方法，为评估既有隧道受新建隧道穿越施工影响时的纵向变形响应提供理论支持。”摘要：A study is conducted on the complex impact of twin tunneling. Utilizing the energy method, a calculation method is developed to assess the settlement of an existing tunnel induced by the excavation of twin tunnels below, considering variations in ground loss. This method integrates the modification factor for soil displacement above the twin tunnel construction, as proposed by Hunt, to account for the influence of ground loss differences based on the classical theories of Peck and Mair. The method combined the Rayleigh–Ritz method and elastic foundation beam theory to derive several expressions for energy and work within the entire system, and these are employed to establish the total potential energy equation. The method for calculating the settlement of the existing tunnel due to the twin tunneling underneath is obtained by solving the total governing equation using the variational method. The results are then compared to the measured data from an engineering case published in the literature. In an example where twin tunnels undercross existing tunnels, a parameter analysis is conducted, including the center-to-center distance between the twin tunnels, the diameter of the new tunnels, and the apparent distance between the new and existing tunnels. The research results showed that the calculated values generally align with the measured data in terms of overall trends, confirming the validity of the proposed method. The asynchronous construction of twin tunnels significantly impacts the settlement of the existing tunnel. As the center-to-center distance between the twin tunnels and the apparent distance between the new and existing tunnels increases, the settlement of the existing tunnel decreases. In contrast, as the diameter of the new tunnel increases, the settlement rises, and the impact of the subsequent undercrossing construction becomes more pronounced. The proposed method provides a quick and accurate evaluation of the longitudinal response of existing tunnels to new tunnel crossings, offering a theoretical foundation and practical guidance for related twin tunnel construction projects.关键词：twin tunnels construction;tunnel settlement;energy method;simplified calculation method;Pasternak foundation96|306|0更新时间：2024-11-15
- “在非饱和黄土力学特性研究领域，专家采用颗粒流方法进行三维离散元仿真模拟分析，建立了软胶结接触模型，准确复现了非饱和黄土侧限压缩和快速增湿试验中的力学响应行为，为黄土力学特性研究提供了新方向。”摘要：A three-dimensional discrete element simulation analysis is conducted using the particle flow method to explore unsaturated loess's macroscopic and microscopic characteristics under the effects of confined compression and rapid wetting. A soft cementation model accounts for the chemical cementation between particles, and inter-particle van der Waals and capillary forces are introduced. The study focuses on the variations in microscopic parameters by selecting suitable contact model parameters. The results showed that the soft cementation contact model accurately reproduces the mechanical properties of unsaturated loess during lateral compression and rapid humidification tests; the micromechanical coordination number increases with an increase in vertical pressure, whereas the inter-particle contact force and contact moment gradually decreases with an increase in water content. However, saturated specimens exhibited an inverse increase. In addition, the proportion of the damage state contact between particles increases with vertical pressure. As vertical pressure gradually grows, strain energy, sliding energy, and damping energy also rise, with strain energy significantly exceeding both sliding and damping energy values. The growth rate of strain energy accelerates with an increase in vertical pressure, peaking at approximately 200 kPa. In addition, strain energy grows gradually with an increase in vertical pressure until it reaches a maximum of around 200 kPa; higher water content correlates with lower final values of strain energy, while the final values of sliding energy and damping energy fluctuate within a small range.关键词：loess;discrete element method;confined compression test;wetting test;microscopic analysis91|385|0更新时间：2024-11-15
- “在建筑结构抗震领域，研究者提出了一种新型高强钢梁柱节点，有效解决了阻尼器受压屈曲问题，具有震后易修复等优势。”摘要：A high-strength steel beam-column joint with flexural-shear type replaceable components is proposed to solve the problems of damper buckling under compression and unstable energy dissipation capacity under certain practical working conditions. This joint offers the advantages of minimal plastic residual deformation, stable energy dissipation capacity, and convenient repair after an earthquake. Five specimens are designed based on the mechanical characteristics of the flexural-shear energy dissipators in the joint. The energy dissipators’ collaborative working mechanism and failure modes are investigated through quasi-static tests. The influences of various design parameters on the bearing capacity, stiffness degradation, and energy dissipation capacity were analyzed. In addition, the applicability of calculation equations for bearing capacity and stiffness is discussed, leading to the proposal of appropriate design advice. Test results indicated that the flexural-shear energy dissipators exhibited superior synergistic performance between the struts. Damage is primarily concentrated at the ends of the energy-dissipating struts, where cracks develop from the interior toward the surface, causing fractures. Then, seismic energy is dissipated. The hysteretic curves of the specimens are relatively full and demonstrate good energy dissipation capacity. Increasing the height-width ratio or the width-thickness ratio of the struts results in decreased initial stiffness and bearing capacity of the energy dissipators. The trend toward out-of-plane deformation becomes evident, reducing the equivalent viscous damping coefficient and worsening the energy dissipation capacity. The strength equation, which considers the effects of flexure and shear strength, accurately predicts the yield and ultimate load of the energy dissipators. The average error between the predicted and test results is less than 5%. However, when the struts’ height-width ratio (HT·B–1) exceeds 5, the predicted results from the stiffness equation are too high, indicating that the softening effect of flexural deformation should be considered at this stage. After a comprehensive comparison of the seismic performance indicators of each specimen, it is recommended that the value of HT·B–1 be approximately 3.5, B·t–1 be 1.5, and n be 10.关键词：energy dissipators;cyclic loads;failure modes;hysteresis loops;bearing capacity;stiffness136|326|0更新时间：2024-11-15
- “在地基处理技术领域，专家研究了透水管桩复合地基固结特性，为透水管桩技术应用提供理论依据。”摘要：Composite foundation is an effective technology for treating soft ground and is widely used in engineering practices. Permeable pipe piles combine the advantages of the permeability of granular piles with the high bearing capacity of adhesive columns, offering a new option for foundation reinforcement. The soil consolidation differential equations and definite solution conditions are established based on the modified equal strain assumption to explore the consolidation characteristics of a composite foundation treated by permeable pipe piles under external loads. This assumption considers the penetration depth of the pile into the cushion and underlying layer, where the permeable pipe pile-soil interface is regarded as an impeded boundary. Various forms of surcharge stress distribution and the disturbance effect are also considered. Then, the finite sine Fourier transform derives analytical solutions for the excess pore-water pressure in the soil element, settlement, and consolidation degree of the composite foundation. The validity of these solutions is confirmed by comparison to the degenerate case, and numerical solutions are calculated using the finite difference technique. Finally, the influences of factors such as the opening ratio, opening radius, distribution form of surcharge stress in the foundation, disturbance effect, and the compression modulus of the cushion and underlying layer on the consolidation characteristics of the composite foundation are investigated through parametric analyses. Compared to conventional pipe piles, the results showed that permeable pipe piles significantly accelerate the dissipation of excess pore-water pressure in the soil around the piles; the higher the opening ratio and the smaller the opening radius, the quicker the dissipation occurs. The fastest consolidation rate occurs when the surcharge stress is distributed in an inverted triangle, and the disturbance effect is not considered. In addition, the consolidation rate of the foundation increases with the increment of the compression modulus of the cushion and underlying layer. The results of this research can provide a theoretical basis for the advancement and engineering design of permeable pipe pile technology.关键词：permeable pipe pile;composite foundation;consolidation;piercing deformation;disturbance effect;analytical solution86|362|0更新时间：2024-11-15
- “粗粒土渗透特性研究取得新进展，专家建立了渗透系数换算模型，为准确预测原级配土渗透系数提供解决方案。”摘要：Coarse-grained soil is extensively utilized in soil and rock dam engineering as well as in projects addressing weak foundations. The permeability characteristics of coarse-grained soil are critical to the safety of engineering designs. Due to indoor instrument size limitations, only a scaled-down version of the original coarse-grained soil can be tested. However, reducing the original grain size alters the pore size between soil particles and the connectivity of pore channels, leading to variations in the permeability coefficient between the scaled-down and original grain-sized soils. This variation is crucial for accurately determining the permeability characteristics of the original grain-sized soil. Therefore, accurately predicting the permeability coefficient of the original grain-sized soil through tests on scaled-down samples has become essential in evaluating the permeability characteristics of coarse-grained soil at engineering sites. Utilizing the continuous grading equation, 16 different grading samples are designed, using the maximum particle diameter dmax and grading area S as variables. The permeability of each sample is quantitatively analyzed using a self-made large constant head permeameter, and the variations in the permeability coefficient of scaled-down grading are examined along with a method for converting the permeability coefficient between coarse-grained soils with varying gradations. The results indicated that when the grading area S remains constant, the permeability coefficient k increases with the increase in the maximum particle diameter dmax. This trend is more pronounced when the grading area is smaller. Conversely, when the maximum particle diameter dmax remains constant, the permeability coefficient k decreases with the increase in the grading area S. The impacts of dmax and S on k can be quantitatively described using specific functions. A permeability coefficient conversion model between different gradations of coarse-grained soil is established based on the relationship among dmax, S, and k. According to the literature, the prediction errors of the permeability coefficient values for different grades of coarse-grained soil obtained using this model are within 6% of the measured values. Additionally, the original permeability coefficient values are successfully estimated using the scaled-down sample test results, with an error of only 2.5%, verifying the applicability of this permeability coefficient calculation model to coarse-grained soil.关键词：coarse-grained soil;reduced scale gradation;permeability coefficient;maximum particle size;gradation area113|458|0更新时间：2024-11-15
CIVIL ENGINEERING
- “陶瓷练泥机仿生设计取得新进展，专家通过优化机头结构，有效提高泥料挤出效率和质量。”摘要：A ceramic extruder is a key piece of equipment for ceramic plastic forming, which plays a critical role in extruding mud with high density, strong plasticity, and uniform composition. However, due to the resistance to adhesion between the extruded clay and the barrel’s internal contact surface, the extruded clay often exhibits non-uniform velocity distribution, delamination, cracking, and other defects that seriously affect the surface quality and forming performance. Inspired by the surface structure of the dung beetle’s body, which reduces viscosity and resistance, this study analyzes the movement status of mud in the ceramic extruder’s head and researches a bionic design with a non-smooth, pit-shaped structure. Using surface modification technology, a pit-shaped bionic structure is implemented on the inner wall of the ceramic extruder’s head, disrupting the original layered order of the mud and altering the state of the water film. Furthermore, a geometric model of the dung beetlex’s pit-shaped bionic structure is established for the extruder’s head. Building on this, the structure is further optimized using the numerical calculation methods of COMSOL multiphysical field and orthogonal experimental design. Subsequent extruding experiments verified the effectiveness of the proposed method. The results showed that the geometric factors affecting the comprehensive performance of the bionic machine head, in descending order of impact, are the arrangement of non-smooth bodies, the spacing between adjacent non-smooth bodies, and the cross-sectional radius of non-smooth bodies. Moreover, when the pits were uniformly distributed on the head’s surface of the ceramic extruder in a parallel arrangement, with a pit radius of 1.5 mm and a distance between the centers of two adjacent pits of 4.5 mm, the drag reduction rate (
$\lambda_{\mathrm{f}} $ $\lambda_{\mathrm{s}}^2 $ 关键词：ceramic extruder;clay defects;bionic pits;viscosity and resistance reduction696|407|0更新时间：2024-11-15 - “在柔性可穿戴器件领域，研究人员设计了球冠式自供能压缩传感结构，实现了压缩应变监测，为可穿戴设备提供新思路。”摘要：This study designs a spherical crown self-powered compressive strain sensing structure to meet the demand for flexible wearable devices capable of compressive strain sensing. Nylon cloth and silicone serve as positive and negative friction materials, respectively, based on the principle of frictional electricity generation of the Triboelectric Nanogenerator (TENG). The design strategy is utilized to achieve variable compression in the friction contact area by employing heterogeneous silicone materials. The principle of self-powered compressive strain sensing is elucidated. Three different mass ratios of printing ink composition materials are developed. The multi-nozzle silicone direct writing 3D printing technology is employed to fabricate this self-powered compressive strain sensing structure. The microscopic morphology of the spherical crown friction structure section is analyzed using a scanning electron microscope. The impact of spherical crown size and compressive strain on the output electrical performance of the sensing structure is investigated through a specially developed compressive motion experimental device. The experimental results indicated that the output signal of the maximum compressive strain increases with the height of the supporting spherical crown. Moreover, the open-circuit voltage, short-circuit transferred charge, and short-circuit current all positively correlate with the compressive strain. As the compressive strain increases from 7% to 35%, the peak open-circuit voltage escalates from 5.3 V to 22.2 V. Concurrently, the sensing structure exhibits robust durability and stability. Ultimately, a self-powered sensing insole utilizing the spherical crown compressive strain sensing structure is designed, demonstrating effective application in monitoring gait and plantar pressure distribution. This study offers novel insights and approaches for the design and production of multi-material wearable flexible self-powered compressive strain sensing structures.关键词：Flexible sensing;3D printing;Self-powered;compressive sensing;wearable device432|438|0更新时间：2024-11-15
- “在海底气举扬矿领域，专家建立了气-液-固混合流体压差模型，验证了理论模型的可靠性，为装置优化设计提供技术支撑。”摘要：An accurate model of a seabed airlift device is difficult to propose using the traditional mechanical equilibrium method because the forces among gas, liquid, and solid phases are very complex. In order to improve lifting performance, a new model is proposed for this device, considering the pressure drop in the gas–liquid–solid mixture without taking into account the forces among these three phases. The flow unit of the gas–liquid–solid three-phase mixture in this device is considered, and a pressure model is established by analyzing the friction between this unit and the pipe. Subsequently, the pressure model is utilized to develop an airlift theoretical model by combining inlet and outlet pressure constraints. Finally, an experiment is conducted to validate this new model; the results showed that: 1) The new model aligns well with the experimental data with a maximum error of 17.6%. 2) A high gas flow rate increases the pressure drop but reduces the solid fraction. The liquid friction loss is substantial under high gas flow rates, suggesting that a flow drag reduction strategy could be proposed for enhanced lifting performance. 3) The solid superficial velocity initially increases and then decreases with the gas superficial velocity, and an optimal gas flow rate exists for achieving maximum ore flow rate. The overall research provides a theoretical basis and technical support for the analysis of multiphase flow and the optimal design of seabed airlift devices.关键词：Seabed mining;Airlift;gas–liquid–solid;Pressure drop651|289|0更新时间：2024-11-15
- “在液晶显示器生产领域，专家建立了芯片互连电阻仿真体系，探究了最优工艺参数，为提高封装效率提供解决方案。”摘要：The chip-on-glass process is a core technology in the production of liquid crystal displays, primarily utilizing thermosetting curing glue and various conductive particles within anisotropic conductive adhesive film to achieve interconnection. The resistance of the chip after interconnection serves as a crucial evaluation index for material selection and process parameters. This study aims to establish a simulation system for chip interconnect resistance and to explore the optimal process parameters for chip packaging with a consistent distribution of conductive particles. In this regard, numerous experiments are conducted to determine the bump interconnection resistance of the chip using CP6530ID anisotropic conductive adhesive film under various process parameters and the number of conductive particles captured by the bump, with the experimental data undergoing statistical analysis. Subsequently, the sequential thermal-mechanical coupling method is utilized to simulate the coupling method of binding temperature and pressure. After determining the deformation variables of conductive particles in CP6530ID anisotropic conductive adhesive film under a thermodynamic coupling field, the chip interconnection resistance is calculated using the established mathematical model of conductive particle interconnection resistance, and the interconnection resistance calculation results are compared with experimental results to verify the validity of the simulation results and the guiding significance of the experiment. The optimization starting point is based on the packaging process parameter range recommended by Sony Chemical. By randomizing the distribution of conductive particles, the study explored the optimal process parameters for chip packaging, using the simulation results as a reference. The findings indicate that under effective chip packaging conditions, the chip interconnect resistance is minimized at a binding temperature of 190 ℃ and a binding pressure of 100 MPa. During the chip packaging process, pressure is a decisive factor in the interconnection resistance, while the influence of temperature on the interconnection resistance is relatively weak. In terms of pressure factors, the greater the pressure, the smaller the interconnection resistance, without considering the fracturing of conductive particles. Conversely, higher temperatures lead to more intense thermal expansion of conductive particles, increasing the interconnection resistance.关键词：liquid crystal display;chip-on-glass process;numerical simulation;interconnect resistance;process parameters598|722|0更新时间：2024-11-15
MECHANICAL ENGINEERING
- “在数字广告领域，专家提出了自注意力深度域嵌入因子分解机模型，通过精化嵌入向量和特征交互，实现点击率预测，性能优于主流模型。”摘要：Objective Click-through rate (CTR) prediction realizes accurate recommendation of digital advertisements by predicting the user’s click probability on advertisements or commodities. However, current CTR prediction models have the following key issues. First, the raw embedding vectors have not been fully refined. Second, the corresponding feature interaction method is too simple. As a result, the performance of the models is heavily restricted. To alleviate these issues, a novel CTR model named self-attention deep field-embedded factorization machine (Self-AtDFEFM) is proposed. Methods First, a well-known multi-head self-attention mechanism is employed to capture the implicit information of the raw embedding vectors on different sub-spaces, and the corresponding weight is calculated to further refine the key low-level features. Second, a novel field-embedded factorization machine (FEFM) is designed to strengthen the interaction intensity between different feature fields by the field pair symmetric matrix. The key low-order feature combinations are fully optimized by the FEFM module for the subsequent high-order feature interaction. Third, a deep neural network (DNN) is built based on the low-order feature combinations to complete implicit high-order feature interaction. Finally, both the explicit and implicit feature interactions are combined together to implement CTR prediction. Results and Discussions Extensive experiments have been performed on the two public available datasets, namely Criteo and Avazu. First, the proposed Self-AtDFEFM is compared with numerous state-of-the-art baselines on the AUC (area under curve) and LogLoss metrics. Second, all parameters of Self-AtDFEFM was tuned, and the parameters included the number of the explicit high-order feature interaction layers, the number of the attention heads, the embedding dimension, and the number of the implicit high-order feature interaction layers. Further, ablation experiments of our model were completed. The results of the experiments showed that: the Self-AtDFEFM model outperformed mainstream baseline models on the AUC and LogLoss metrics; all parameters of Self-AtDFEFM have been adjusted to their optimal values; each module form a kind of joint force to improve the final CTR prediction performance. Notably, the explicit high-order feature interaction layer plays the most important role in Self-AtDFEFM. Conclusions Each module of Self-AtDFEFM is plug-and-play, that is, the Self-AtDFEFM is easier to build and deploy. Hence, Self-AtDFEFM achieves a good trade-off between prediction performance and model complexity, making it highly practical.关键词：click-through rate prediction;multi-head self-attention;feature interaction;field-embedded factorization machine;deep neural network139|466|0更新时间：2024-11-15
- “RBNet算法在阴影检测领域取得突破，专家验证了其在提高检测准确率和推理速度方面的优越性，为移动端设备应用提供新方案。”摘要：Since the number of pixels along shadow boundaries is often significantly smaller than that of pixles within shadow regions, accurately detecting shadow boundary areas is more challenging than detecting pixels within the shadow interior. To improve detection accuracy at shadow boundaries, a novel and efficient lightweight boundary-aware shadow detection algorithm called RBNet was proposed in this paper. During the supervised training phase, the input image was divided into shadow and non-shadow regions, and the distance transforms were applied to decouple the shadow and non-shadow regions into a boundary part and a body part respectively. Secondly, the boundary features of shadow regions were learned and the impact of shadow boundaries and interior pixels were balanced in the loss function of RBNet. Additionally, a multi-branch fusion structured re-parameterization module named RepConv was designed in RBNet. Through re-parameterization, the model parameters and computational cost were reduced, and the inference speed was improved. A series of shadow detection comparison experiments and algorithm model comparison experiments between the proposed RBNet and other common shadow detection algorithms is conducted in the paper. Experimental results demonstrate that the proposed shadow detection algorithm, RBNet, not only has the smallest model size but also achieves the fastest inference speed, while outperforming existing shadow detection algorithms in terms of BER performance. RBNet is highly applicable to mobile devices. When combined with shadow removal algorithms, it can significantly enhance the accuracy of object detection or segmentation tasks.关键词：Shadow detection;deep learning;mask decoupling;convolutional network;reparameterization155|346|0更新时间：2024-11-15
INFORMATION ENGINEERING
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0