高分子材料导热性能影响因素研究进展

介绍了高分子材料导热性能影响因素研究进展，重点阐释了聚合物基体的结构特性（链结构、分子间相互作用、取向、结晶度等）、导热填料（种类、含量、形态、尺寸等）以及制备方法等对高分子材料导热性能的影响。     

From enfant terrible to problem-solver? Tracing the competing discourse to explain blockchain-related technological diffusion

Since the inception of blockchain-related technologies over a decade ago, investors’ uptake of the technologies has grown rapidly. But even with the advancement in standing from conceptual beginnings to real-world experimentation, mainstream adoption of the technologies in either organizational or individual contexts has yet to eventuate. To resolve this paradox, we examine progression of the technologies’ diffusion instead, and in the process, seek to uncover the underlying dynamics of sentiments at play. Diffusion involves multiple actions by different parties, with social discourses via diverse media playing an essential role. In demonstrating the interplay of support to resistance via media framing and the diffusion of innovation theory, findings from our paper contribute to the technology adoption literature. Understanding resistant behavior toward a new technology area could help resolve potential issues that may arise. This would hopefully lead to better technological implementation outcomes in future.     

Lake charr (Salvelinus namaycush) clotting response may act as a plasma biomarker of sea lamprey (Petromyzon marinus) parasitism: Implications for management and wound assessment

Sea lampreys (Petromyzon marinus) have decimated the Great Lakes fisheries over the past century, and their control is central to protecting native fish populations. Wounding data collected from host fish, including classifying wounds as Type A (penetrating the integument) or Type B (superficial), is an integral part of sea lamprey monitoring efforts and helps inform management decisions; however, wound assessment is subjective and error-prone. This study aimed to determine if protein biomarkers of parasitism could be quantified in host fish plasma to serve as a potential objective aide in current wound assessment practices. Male siscowet lake charr (Salvelinus namaycush) were parasitized in a lab setting for four days, after which the sea lamprey was removed, host blood was collected, and the wound Type recorded. A second blood sample was collected from host fish with Type A wounds 7 months later. The plasma proteome was quantified using iTRAQ, and the relative abundances of 169 proteins were compared between parasitized and non-parasitized control fish. Three functional classes of proteins were modified by sea lamprey parasitism: immune response, lipid transport, and blood coagulation. A major finding was evidence of a concerted anticoagulation response in fish with Type A wounds, including changes in protein components and regulators of fibrin clot formation, some of which did not fully recover within 7 months. A modified clotting assay yielded a smaller thrombin-induced fibrin clot from parasitized fish, supporting the proteomic results. Therefore, measuring blood clottability could improve sea lamprey damage estimates by providing a more objective and quantitative index of parasitism than is offered by wounding data alone.     

对大中型钢结构企业生产管理体系的研究分析

以某大型钢结构制造企业为例,通过对产量分布、质量控制、劳动效率等基本数据的分析,研判外协生产管理体系方面存在的问题,从组织形态、管理模式、职能管理等多个维度提出了可行性建议,可为同类型钢结构制造企业提供经验借鉴。     

以职业教育推动传统建筑工艺传承的探讨

传统建筑工艺是建筑遗产中不可或缺的重要组成部分，然而辉煌背后，传统建筑工艺却面临着断代或失传的风险。职业教育倡导的“教学做”一体化教学模式，强调培养学生的职业技能、动手能力的特点与传统建筑工艺传承所需的千锤百炼相吻合，并在实践教学中已取得了一定的成果，但同时显现了不足，文章将针对这些不足之处，以广州城市职业学院城市建设工程系为例，探讨职业院校开展传统建筑工艺的教学模式。     

光纤有线电视在乡村振兴中的作用

2018年初,我国提出“乡村振兴”战略,此战略目的是增强农村农业发展的新活力,从而加速推进其现代化进程。光纤有线电视因具有损耗小、传输距离远、频带宽、传输容量大、抗电磁干扰等实用优势,并在贯彻及落实着“社会效益优先,两个效益统一”的原则下在社会、经济及文化等方面助力着乡村的振兴。     

金属增材制造技术研究热点聚类分析

为深入分析金属增材制造技术分类和发展状况,采集中国期刊全文数据库(CNKI)收录的核心期刊上的768篇科技文献,借助文献分析可视化软件CiteSpace,对关键词聚类进行了全景式描绘,构建金属增材制造技术知识图谱,揭示该技术研究分类以及演化趋势。     

基于条件信息卷积生成对抗网络的图像识别

传统的图像识别方法需要大量有标签样本进行训练，且模型训练难以达到稳定。针对这些问题，结合条件生成网络和信息最大化生成网络的结构优势建立了条件信息卷积生成网络（C-Info-DCGAN）。模型增加图像的类别信息和潜在信息作为输入数据，然后利用Q网络去更好地发挥类别信息和潜在信息对训练的引导作用，并且利用深度卷积网络来加强对图像特征的提取能力。实验结果表明，该方法能够加快模型训练收敛速度，并有效提高图像识别的准确率。     

结合边缘特征先验引导的深度卷积显著性检测

针对当前基于深度学习的显著性检测算法缺少利用先验特征和边缘信息，且在复杂场景中难以检测出鲁棒性强的显著性区域的问题，提出了一种结合边缘特征，利用先验信息引导的全卷积神经网络显著性检测算法。该算法利用三种被经常用到的先验知识结合边缘信息形成先验图，通过注意力机制将提取的先验特征与深度特征有效融合，最终通过提出的循环卷积反馈优化策略迭代地学习改进显著性区域，从而产生更可靠的最终显著图预测。经过实验定性定量分析，对比证明了算法的可靠性。