蛋类礼盒包装结构的缓冲性能研究

目的研究蛋类礼盒包装结构的缓冲性能。方法以蛋类尺寸为基础,建立可发性聚乙烯(expandable polyethylene. EPE)缓冲单元结构和组合结构,进行静态仿真分析和实验验证,比较结构在形状、叠合层数、组合形式等参数变化时的载荷与位移。结果结构层数变化相同时,单元结构的极限载荷从143N增加到236 N,组合结构的极限载荷从224 N增加到476 N,均呈近线性增长。结论蛋类礼盒包装中,组合结构的承载能力优于单元结构,通过单元结构的组合,可满足不同蛋类的包装要求。     

Preparation of Silver Nanowires via a Rapid,Scalable and Green Pathway

Rapid synthesis of silver nanowires(Ag NWs) with high quality and a broad processing window is challenging because of the low selectivity of the formation of multiply twinned particles at the nucleation stage for subsequent Ag NWs growth.Herein we report a systematic study of the water-involved heterogeneous nucleation of Ag NWs with high rate(less than 20 min) in a simple and scalable preparation method.Using glycerol as a reducing agent and a solvent with a high boiling point,the reaction is rapidly heated to 210 ℃ in air to synthesize Ag NWs with a very high yield in gram level.It is noted that the addition of a small dose of water plays a key role for obtaining highly pure Ag NWs in high yield,and the optimal water/glycerol ratio is0.25%.After investigating a series of forming factors including reaction temperature and dose of catalysts,the formation kinetics and mechanism of the Ag NWs are proposed.Compared to other preparation methods,our strategy is simple and reproducible.These Ag NWs show a strong Raman enhancement effect for organic molecules on their surface.     

GOS+MU:一种查询对象选择新方法

在分析单一MU(Most Uncertainty)采样缺陷的基础上,提出一种"全局最优搜寻"方法 GOS(Global Optimal Search),并结合MU共同完成查询选择。GOS+MU方法中,GOS着眼全局寻找目标,在应用环境能提供的训练样本数量有限、分类器受训不充分时,该方法选择的对象学习价值高,能快速推进分类器学习进程;MU则能够在GOS采样失效情形下,利用分类器当前训练成果,选择查询不确定性最强的样本补充训练集。通过对网络商品的用户评论进行分类仿真,并比较其他采样学习方法的效果,证明了GOS+MU方法在压缩学习成本、提高训练效率方面的有效性。