地空频率域电磁系统在小煤窑勘查中的应用

为实现金谷煤矿小煤窑采掘边界圈定，采用地空频率域电磁系统开展小煤窑采空积水区的勘查试验工作。实际勘探中设计124条南北方向测线，利用无人机搭载探测天线进行电磁数据的连续采集。通过数据处理与解释，获得了测区内地下50~300 m范围内的视电阻率剖面。根据相对低阻异常的分布情况，圈围出两个煤窑采空积水区疑似位置。经过与实际煤层采掘资料对比，圈定的1处低阻异常区域下方存在高度充水的采空区。研究结果证明了地空频率域电磁系统在小煤窑勘查工作中的可行性及有效性。     

A地区SⅢ油层组下部含泥含钙储层有效厚度及夹层划分标准

有效厚度标准是产能预测和储量计算的重要参数。分析不同含油产状岩心长度占比及产油能力，确定含油性下限为油斑。分析储层岩性与含油性关系并结合含油性下限，确定岩性下限为泥质粉砂岩和钙质粉砂岩；分析岩性、含油性与物性关系并结合岩性、含油性下限，确定物性下限：φ≥9.0%,K≥0.1×10^-3μm²;根据试油资料，采用AC-RLLD交会图版，确定有效厚度电性标准：AC≥260μs/m,RLLD≥7.8Ω·m; 240μs/m﹤AC﹤260μs/m,RLLD≥-82.09ln(AC)+464.92。分析研究区泥质和钙质夹层测井响应特征，优选MSFL、RMN、RMG、HAC作为泥质和钙质夹层敏感曲线；确定泥质夹层划分标准：MSFL回返程度≥42%、RMN回返程度≥42%、RMG回返程度≥41%,钙质夹层划分标准：MSFL/MSFL_S1D≥12、RMN/RMN_S1D≥4.2、HAC/HAC_S1D≤0.67。研究区5口取心井储层有效厚度解释与对比，表明建立的有效厚度及夹层划分标准适用于研究区目的...     

Microfluidic-Generated Seeds for Gold Nanotriangle Synthesis in Three or Two Steps

Nanoparticle synthesis has drawn great attention in the last decades. The study of crystal growth mechanisms and optimization of the existing methods lead to the increasing accessibility of nanomaterials, such as gold nanotriangles which have great potential in the fields of plasmonics and catalysis. To form such structures, a careful balance of reaction parameters has to be maintained. Herein, a novel synthesis of gold nanotriangles from seeds derived with a micromixer, which provides a highly efficient mixing and simple parameter control is reported. The impact of the implemented reactor on the primary seed characteristics is investigated. The following growth steps are studied to reveal the phenomena affecting the shape yield. The use of microfluidic seeds led to the formation of well-defined triangles with a narrower size distribution compared to the entirely conventional batch synthesis. A shortened two-step procedure for the formation of triangles directly from primary seeds, granting an express but robust synthesis is further described. Moreover, the need for a thorough study of seed crystallinity depending on the synthesis conditions, which – together with additional parameter optimization – will bring a new perspective to the use of micromixers which are promising for scaling up nanomaterial production is highlighted.     

Nanoscale Oxide Formation at α-Al2O3–Nb Interfaces

Parts for metallurgical applications made from refractory metal–ceramic composites offer improved thermal shock resistance due to their capability for resistive heating compared to ones made solely from ceramics such as Al₂O₃. The combination of Al₂O₃ and Nb is intriguing as both show similar thermal expansion behavior over a wide temperature range. The high affinity of Nb for O to form nonprotective oxides, however, hampers its use in oxidative environments. Formation of such phases at the ceramic–metal interface can have detrimental effects on the cohesion of the composites. For this work, nanocrystalline Nb films are deposited on sapphire substrates by magnetron sputtering to study diffusion of O and high-temperature phase formation at a refractory metal–ceramic interface during heat treatment under Ar at 1600 °C. A combined approach of atom probe tomography and transmission electron microscopy for compositional and crystallographic analyses reveals that at triple junctions of the sapphire–Nb interface with Nb grain boundaries, heterogeneous nucleation of nanoscale NbO₂ occurs, which further reacts with Al₂O₃ to form AlNbO₄, while the Nb film itself remains metallic. Fast O transport through grain boundaries leads to internal oxidation at the interface, whereas regions further away from Nb grain boundaries remain unchanged.     

基于高分辨质谱和网络药理学的可可茶多酚降血糖活性研究

目的:探究可可茶茶多酚的组成特性及功能活性。方法:以可可茶多酚粗提物为研究材料，采用高分辨质谱分析并鉴定其组分；基于网络药理学预测其潜在功能。结果:经酶抑制试验证实可可茶具有显著抑制α-葡萄糖苷酶和α-淀粉酶活性的体外降血糖活性；通过GO功能分析和KEGG通路富集预测其作用机制，涉及GAPDH、AKT等6个关键靶点和PI3K-AKT等多条通路，通过构建“组分—靶点—通路”网络，结合各组分与淀粉水解酶结合的分子对接筛选出杨梅素、花青素等10种核心降血糖活性组分。结论:可可茶多酚粗提物通过多组分、多靶点、多通路可以共同发挥降血糖功效。     

基于多样卷积单元高效人体姿态估计

设计兼具准确率和轻量化的人体姿态估计网络模型成为了人机交互领域的迫切需求。为了满足这一需求,结合HRNet模型的高分辨率设计模式,提出了一种多样化高效卷积单元的高分辨率网络模型DU-HRNet。为了探索并增强来自不同感受野大小层的多尺度信息,鼓励卷积层间信息更加多样化,模型允许并行分支中的每一分支拥有不同类型的高效卷积单元。为了改善模型的非线性,在高效卷积单元中使用通道注意力ECANet。在MS COCO关键点检测数据集和MPII数据集中验证了模型的有效性。模型在参数量等于7.6 M、GFLOPs为2.66,没有经过任何后期处理的条件下,在COCO val2017数据集上达到了71.1 mAP(mean Average Precision,平均精度均值)分数,在COCO test-dev2017数据集上达到71.8 mAP分数。通过消融实验验证了模型整体和组成部分的有效性。     

基于扩展目标先验的贝叶斯压缩感知成像

已有的基于压缩感知理论的雷达成像技术通常是基于待重构目标散射点自身非常“稀疏”的前提下。然而实际情形中，针对大型刚体目标成像，如飞机、舰船等，其距离维及方位维通常存在一定的扩展特性，此时成像场景目标空间域的稀疏性相对较差，如果仍采用传统方法进行目标反演，所获得的目标重构性能通常并不理想。据此，基于扩展目标的先验信息，提出了一种改进的贝叶斯压缩感知成像方法。仿真试验验证了所提方法的有效性。     

UPLC–Q–Orbitrap HRMS结合全碳标记稳定同位素内标法精确测定粮油中黄曲霉毒素

建立了全自动免疫亲和柱净化–超高效液相色谱–四级杆/静电场轨道阱高分辨质谱（UPLC-Q-Orbitrap HRMS）结合全碳标记稳定同位素内标法对粮油中四种黄曲霉毒素精确定量方法。样品准确称量后加入全碳标记稳定同位素内标，经84/16(V/V)的乙腈/水提取、离心、稀释后，通过全自动固相萃取仪采用黄曲霉毒素免疫亲和柱进行全自动净化，净化液在UPLC-Q-Orbitrap HRMS上进行检测后采用内标法进行定量分析。结果表明：黄曲霉毒素的加标回收率为84.7%～104.6%，日内相对标准偏差为1.7%～9.4%，日间精密度为0.7%～8.4%。采用本方法对国家标准物质的检测结果在赋值范围内，能够最大限度地降低实验人员工作量，提高工作效率，避免因人员操作失误导致的结果偏差，适用于粮油中黄曲霉毒素的精准测定，可满足日常检测和监测需求。