针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器...针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器人专为提升清洁效率和自动化水平设计,配备了远程控制、视觉识别、智能抓取、重量检测及状态显示等功能。采用英伟达Jetson Nano作为核心处理器,结合Intel D415深度相机和基于FloW数据集训练的YOLOv8算法,实现水面漂浮垃圾的实时检测与精确定位。系统通过STM32微控制器解析视觉数据并控制机械臂完成精准抓取。为提高移动性能,机器人采用麦克纳姆轮实现全向运动,当内置称重传感器检测到收集装置满载时,系统可自主返回基地卸载垃圾。此外,系统集成HC-05蓝牙模块实现远程无线控制,并通过OLED显示屏实时显示工作状态。综合应用了AIoT、自动化控制及视觉识别技术,突破了传统清洁方式的局限,显著提升了沙滩水域清洁工作的效率和便捷性,为环保行动提供了强有力的工具。展开更多
Vacuum glazing is highly regarded for its ability to transmit light while providing heat preservation,sound insulation,lightweight characteristics,and resistance to condensation.Scholars have made significant strides ...Vacuum glazing is highly regarded for its ability to transmit light while providing heat preservation,sound insulation,lightweight characteristics,and resistance to condensation.Scholars have made significant strides in the study of vacuum glazing through their notable efforts.This study systematically reviewed vacuum glazing and its composite structures,including material selection,fabrication techniques,research methods,and performance evaluation.This review initially presented fundamental techniques for preparing vacuum glazing,with a focus on edge seal and support pillar arrangements,and introduced common composite structures such as hybrid and tinted vacuum glazing.Furthermore,this review summarized the analytical,numerical,and experimental methodologies used to assess the thermal performance of vacuum glazing.This study also outlined heat transfer coefficients associated with various vacuum glazing structures,investigated the influence of different parameters on their heat transfer coefficients,and evaluated their potential for energy conservation across diverse climatic regions.Finally,the research delineated future trends in the advancement of vacuum glazing to provide guidance for both theoretical studies and practical applications in industry.This research serves as a valuable resource for both theoretical exploration and practical integration of vacuum glazing,facilitating its improvement and optimization to advance sustainable low-carbon building practices.展开更多
Laser directed energy deposition(LDED)is an emerging branch of metal-based additive manufacturing(AM)processes,offering unprecedented capabilities for high-performance fabrication with complex geometries and near-net ...Laser directed energy deposition(LDED)is an emerging branch of metal-based additive manufacturing(AM)processes,offering unprecedented capabilities for high-performance fabrication with complex geometries and near-net shapes.This technology is gathering increasing attention from industries such as biomedical,automotive,and aerospace.However,achieving consistent part quality and desired material properties is challenging due to intricate processing parameters and potential process defects such as dynamic melt-pool behavior and localized heat accumulation.This paper reviews recent advances in on-line quality control,focusing on in-situ measurement and closed-loop control for efficient assurance of LDED-fabricated parts.The quality principles,encompassing accuracy and material performance,are summarized to lay a foundation for understanding the mechanisms of quality defects and influencing factors.This review explores and thoroughly compares advancements in indirect process measurements,such as optical,thermal,and acoustic monitoring with direct quality measurements,including laser-line scanning and operando synchrotron X-ray imaging.Depending on the sensing techniques,this paper highlights a hierarchical control strategy for adaptive parameter regulation on intra-layer and inter-layer scales.The requirements and performance of various state-of-the-art controllers are critically compared to indicate their suitable applications.The importance of machine learning in detecting process anomalies and predicting build quality based on sensory signals is also outlined.Future directions are proposed towards adaptive,automated,and intelligent quality control,with a focus on multi-modal monitoring,physics-informed neural networks for interpretable analysis,and multi-objective control applications.展开更多
In recent years,cold-formed steel(CFS)built-up sections have gained a lot of attention in construction.This is mainly because of their structural efficiency and the design advantages they offer.They provide better loa...In recent years,cold-formed steel(CFS)built-up sections have gained a lot of attention in construction.This is mainly because of their structural efficiency and the design advantages they offer.They provide better loadbearing strength and show greater resistance to elastic instability.This study looks at both experimental and numerical analysis of built-up CFS columns.The columns were formed by joining two C-sections in different ways:back-to-back,face-to-face,and box arrangements.Each type was tested with different slenderness ratios.For the experiments,the back-to-back and box sections were connected using two rows of rivets.The face-to-face sections,on the other hand,were joined by welding.In order to improve axial strength and overall stability,all column samples were filled with ordinary concrete,conforming to class C25/30.The numerical modeling was done in ABAQUS to study themechanical behavior of the columns.This helped in understanding how different joining methods affect their axial compression performance.Analytical checkswere also carried out using Eurocode 3 for hollowsections and Eurocode 4 for concretefilled sections.The role of concrete confinement was examined as well,following American Concrete Institute(ACI)guidelines,for both face-to-face and box-shaped columns.Thenumerical results matched closely with the experimental findings,with variations of less than 5%.The study identified key failure modes such as local buckling and distortional buckling.It highlighted how section shape,type of connection,and concrete infill all play amajor role in improving the strength of built-up CFS columns.展开更多
文摘目的评估时空电阻抗断层成像(spatiotemporal electrical impedance tomography,ST-EIT)在言语发声任务下,能否有效捕捉并区分腭裂(cleft palate,CP)患者与正常对照(normal control,NC)的言语呼吸功能特征。方法本研究纳入75名受试者(CP组37例,NC组38例)。在标准化发声任务中,同步采集电阻抗断层成像(electrical impedance tomography,EIT)图像与口鼻气流信号,构建涵盖时间、气流与空间维度的三域特征,采用曼-惠特尼U检验(MannWhitney U test)比较组间差异。基于肺量计法(spirometry)、鼻音计(nasometry)及ST-EIT等多源数据,分别训练极端梯度提升(extreme gradient boosting,XGBoost)分类模型,采用5折交叉验证评估性能,并引入Shapley加性解释(Shapley additive explanations,SHAP)方法进行特征贡献分析。结果CP组呈现显著的呼吸表型差异。时间域中,吸/呼相位时长均显著缩短(P<0.001),吸/呼时间比显著升高;气流域中,呼气期平均气流与峰值气流显著增强,吸气期无明显差异;空间域中,感兴趣区(region of interest,ROI)1和4的潮气阻抗变化(tidal impedance variation,TIV)显著升高,ROI2显著降低,全局不均一性(global inhomogeneity,GI)较低,通气中心(center of ventilation,CoV)呈轻度升高。ST-EIT模型分类性能最佳,曲线下面积(area under the curve,AUC)达0.915,准确率优于单一肺功能检测或鼻音计检测。SHAP结果表明,时空特征对分类决策贡献最大。结论ST-EIT技术能有效揭示CP患者言语呼吸功能在时间-气流-空间三域的特异性改变,为床旁筛查、康复评估及随访监测提供了优于常规检测的客观量化依据。
文摘针对沙滩水域环境复杂且难以有效清洁的问题,综合应用人工智能物联网(Artificial Internet of Things, AIoT)和视觉识别等先进技术,开发了一款具有远程控制、视觉识别、智能抓取、重量检测及智能显示等功能的多功能清洁机器人。该机器人专为提升清洁效率和自动化水平设计,配备了远程控制、视觉识别、智能抓取、重量检测及状态显示等功能。采用英伟达Jetson Nano作为核心处理器,结合Intel D415深度相机和基于FloW数据集训练的YOLOv8算法,实现水面漂浮垃圾的实时检测与精确定位。系统通过STM32微控制器解析视觉数据并控制机械臂完成精准抓取。为提高移动性能,机器人采用麦克纳姆轮实现全向运动,当内置称重传感器检测到收集装置满载时,系统可自主返回基地卸载垃圾。此外,系统集成HC-05蓝牙模块实现远程无线控制,并通过OLED显示屏实时显示工作状态。综合应用了AIoT、自动化控制及视觉识别技术,突破了传统清洁方式的局限,显著提升了沙滩水域清洁工作的效率和便捷性,为环保行动提供了强有力的工具。
基金supported by the National Key R&D Program of China(2023YFC3806202)the National Natural Science Foundation of China(52308093)+3 种基金the Natural Science Foundation of Hunan Province(2023JJ40154)the Science and Technology Innovation Leading Talent Program of Hunan Province(2023RC1042)the Natural Science Foundation of Changsha(kq2208032)the China Postdoctoral Science Foundation(2023M741132 and 2024T170263)。
文摘Vacuum glazing is highly regarded for its ability to transmit light while providing heat preservation,sound insulation,lightweight characteristics,and resistance to condensation.Scholars have made significant strides in the study of vacuum glazing through their notable efforts.This study systematically reviewed vacuum glazing and its composite structures,including material selection,fabrication techniques,research methods,and performance evaluation.This review initially presented fundamental techniques for preparing vacuum glazing,with a focus on edge seal and support pillar arrangements,and introduced common composite structures such as hybrid and tinted vacuum glazing.Furthermore,this review summarized the analytical,numerical,and experimental methodologies used to assess the thermal performance of vacuum glazing.This study also outlined heat transfer coefficients associated with various vacuum glazing structures,investigated the influence of different parameters on their heat transfer coefficients,and evaluated their potential for energy conservation across diverse climatic regions.Finally,the research delineated future trends in the advancement of vacuum glazing to provide guidance for both theoretical studies and practical applications in industry.This research serves as a valuable resource for both theoretical exploration and practical integration of vacuum glazing,facilitating its improvement and optimization to advance sustainable low-carbon building practices.
基金supported by Royal Academy of Engineering(IF2223B-125)Royal Society(IECR3213107)。
文摘Laser directed energy deposition(LDED)is an emerging branch of metal-based additive manufacturing(AM)processes,offering unprecedented capabilities for high-performance fabrication with complex geometries and near-net shapes.This technology is gathering increasing attention from industries such as biomedical,automotive,and aerospace.However,achieving consistent part quality and desired material properties is challenging due to intricate processing parameters and potential process defects such as dynamic melt-pool behavior and localized heat accumulation.This paper reviews recent advances in on-line quality control,focusing on in-situ measurement and closed-loop control for efficient assurance of LDED-fabricated parts.The quality principles,encompassing accuracy and material performance,are summarized to lay a foundation for understanding the mechanisms of quality defects and influencing factors.This review explores and thoroughly compares advancements in indirect process measurements,such as optical,thermal,and acoustic monitoring with direct quality measurements,including laser-line scanning and operando synchrotron X-ray imaging.Depending on the sensing techniques,this paper highlights a hierarchical control strategy for adaptive parameter regulation on intra-layer and inter-layer scales.The requirements and performance of various state-of-the-art controllers are critically compared to indicate their suitable applications.The importance of machine learning in detecting process anomalies and predicting build quality based on sensory signals is also outlined.Future directions are proposed towards adaptive,automated,and intelligent quality control,with a focus on multi-modal monitoring,physics-informed neural networks for interpretable analysis,and multi-objective control applications.
文摘In recent years,cold-formed steel(CFS)built-up sections have gained a lot of attention in construction.This is mainly because of their structural efficiency and the design advantages they offer.They provide better loadbearing strength and show greater resistance to elastic instability.This study looks at both experimental and numerical analysis of built-up CFS columns.The columns were formed by joining two C-sections in different ways:back-to-back,face-to-face,and box arrangements.Each type was tested with different slenderness ratios.For the experiments,the back-to-back and box sections were connected using two rows of rivets.The face-to-face sections,on the other hand,were joined by welding.In order to improve axial strength and overall stability,all column samples were filled with ordinary concrete,conforming to class C25/30.The numerical modeling was done in ABAQUS to study themechanical behavior of the columns.This helped in understanding how different joining methods affect their axial compression performance.Analytical checkswere also carried out using Eurocode 3 for hollowsections and Eurocode 4 for concretefilled sections.The role of concrete confinement was examined as well,following American Concrete Institute(ACI)guidelines,for both face-to-face and box-shaped columns.Thenumerical results matched closely with the experimental findings,with variations of less than 5%.The study identified key failure modes such as local buckling and distortional buckling.It highlighted how section shape,type of connection,and concrete infill all play amajor role in improving the strength of built-up CFS columns.
