振动磨简体内部的能量传递规律

在对大量摄像实验反复观测和研究后发现:磨筒内的磨介存在抛射、整体回转和本身自转3种运动状态;磨介与筒壁的抛离与接触是沿着振动方向发生的,此过程是周期性的,其周期与振动周期相同．在此基础上,建立了粉磨介质冲量传递模型,探讨了振动磨简体内部的冲量传递、能量传输和能量耗散的分布规律,揭示了振动磨的粉碎机理．以此为理论依据,提出了减小磨筒内的惰性区、提高粉磨效率的途径．     

X射线荧光光谱法测定化探样品中的主次痕量元素

采用低压聚乙烯镶边垫底的粉末样品压片制样,用ZSX primus Ⅱ X射线荧光光谱仪测定化探样品中C、N、Na、Mg、A1、Si、P、S、C1、K、Ca、Sc、Ti、V、Cr、Mn、Fe、Co、Ni、Cu、Zn、Ga、As、Br 、Rb、Sr、Y、Zr、Nb、Mo、Sn、Ba、La、Ce、Hf、U、Th、Pb和Nd,共39个组分,C、N的分析采用RX45、RX61专用分析晶体.使用经验系数法和康普顿散射线作内标校正基体效应,经标准物质检验的分析结果与标样值吻合.用GBW07308国家一级标准物质作精密度试验的统计结果,除U的RSD(n=12)为23.8%,Cl、Sc、Mo、Sn的RSD＜13.0%以外,其余各组分均小于6.0%.     

飞行器三轴姿态测量方法

介绍了到目前为止常用飞行器三轴姿态测量敏感器.详述了这些姿态敏感器的工作原理.根据这些原理分析了它们的优缺点,并比较了它们的性能指标.给出了具有指导意义的结论.     

电感耦合等离子质谱法测定地质样品中的稀贵金属元素

研究建立了直接用王水消解地质勘探样品，利用电感耦合等离子质谱仪(ICP—MS测定其中稀贵金属元素Au、Pt、Pd、Ir、Ru、Rh的分析方法,建立校正方程及引入内标元素Cs，有效克服了质谱干扰及由于基体效应产生的非质谱干扰．方法的测定下限为0．7—2．0ng/g，精密度为3．3％-7．4％，加标回收率为91％-113％，适于大量地质样品中稀贵金属元素的及时分析。     

张峰水库输水工程枢纽地下水对环境的腐蚀性评价与分析

从离子浓度的大小可以判定地下水与地表水对混凝土、钢筋是否有腐蚀性，以便为材料的选择提供依据。对张峰水库输水工程枢纽部分的地下水样进行了腐蚀性分析测试，并就测试结果进行了评价，指出地下水对混凝土没有腐蚀性，对钢有弱腐蚀，选择钢筋时应考虑。     

基于核最小二乘回归子空间分割的高维小样本数据聚类

针对基于线性表示理论的子空间分割方法没有考虑高维小样本数据的非线性性质,借鉴核理论,提出核最小二乘回归子空间分割方法,使子空间分割方法适合高维小样本数据的非线性性质.经6个基因表达数据集和4个图像数据集上的实验,表明该方法是有效的.     

基于改进的卷积神经网络的钢号识别

为获取样本的多样性特征,提出了一种改进的卷积神经网络结构。该网络中引入多层递归神经网络,利用卷积神经网络提取输入图像的浅层特征,同时利用卷积神经网络和递归神经网络并行提取高层特征,最后将两种网络学习到的特征进行融合输入到分类器中分类。利用迁移学习理论解决小样本集数据训练不足的问题,并将这种卷积神经网络结构应用于石油物资管线钢号识别中。实验探究了递归神经网络个数与卷积核个数对网络性能的影响,实验结果表明,改进的网络结构与其它网络进行对比,错误率降低了 3% 。     

改进的核极限学习机定位算法

针对神经网络无线定位方法,存在训练耗时长,定位结果易受噪声干扰的问题,提出了一种改进的核极限学习机无线定位算法。采取在同一位置进行多次测量的方法得到训练数据;把同一位置测得的数据划分为一个样本子空间并提取样本子空间的特征,以样本子空间的特征代替原来的训练数据;利用矩阵近似及矩阵扩展的相关理论改进核极限学习机算法;将处理过的训练数据利用改进的核极限学习机进行训练,得到定位预测模型。仿真结果表明,在相同数据集下,改进的核极限学习机训练用时短、定位速度快;在相同噪声干扰情况下,此算法定位预测误差小。经验证,该算法不但能提高网络的训练速度、定位速度,还能有效地降低噪声的干扰,提高定位精度。     

Native agarose gel electrophoresis and electroelution: A fast and cost‐effective method to separate the small and large hepatitis B capsids

Hepatitis B core antigen (HBcAg) expressed in Escherichia coli is able to self‐assemble into large and small capsids comprising 240 (triangulation number T = 4) and 180 (triangulation number T = 3) subunits, respectively. Conventionally, sucrose density gradient ultracentrifugation and SEC have been used to separate these capsids. However, good separation of the large and small particles with these methods is never achieved. In the present study, we employed a simple, fast, and cost‐effective method to separate the T = 3 and T = 4 HBcAg capsids by using native agarose gel electrophoresis followed by an electroelution method (NAGE‐EE). This is a direct, fast, and economic method for isolating the large and small HBcAg particles homogenously based on the hydrodynamic radius of the spherical particles. Dynamic light scattering analysis demonstrated that the T = 3 and T = 4 HBcAg capsids prepared using the NAGE‐EE method are monodisperse with polydispersity values of ～15% and ～13%, respectively. ELISA proved that the antigenicity of the capsids was not affected in the purification process. Overall, NAGE‐EE produced T = 3 and T = 4 capsids with a purity above 90%, and the recovery was 34% and 50%, respectively (total recovery of HBcAg is ～84%), and the operation time is 15 and 4 times lesser than that of the sucrose density gradient ultracentrifugation and SEC, respectively.     

Determination of organochlorine pesticides (OCPs) in breast milk from Rio Grande do Sul,Brazil, using a modified QuEChERS method and gas chromatography-negative chemical ionisation-mass spectrometry

ABSTRACT

The food chain is the main source of exposure to humans by organochlorine pesticides (OCPs) due to the bioaccumulation. Breast milk can accumulate OCPs, so this matrix is often used as an environmental bioindicator. The currently available methods for the determination of several OCPs and metabolites in breast milk involve, in general, multi-step sample preparation and quantification techniques with low selectivity, high cost and much time and labour. Thus, a fast and efficient method based on sample preparation using the quick, easy, cheap, effective, rugged and safe (QuEChERS) method combined with gas chromatography coupled to mass spectrometry with negative chemical ionisation (GC-NCI-MS) was developed, validated and applied for determination of 16 OCPs and metabolites in breast milk samples. The extract was cleaned by dispersive solid-phase extraction (d-SPE) using MgSO₄ and C18, evaporated in a Turbovap® system, redissolved and analysed by GC-NCI-MS. The method was validated showing acceptable recoveries (72–118%) and precision (RSD <19%). Method limits of detection (LODs) and quantification (LOQs) ranged from 0.75 to 7.5 ng g^?1 and from 2.5 to 25 ng g^?1 lipid, respectively. The method was successfully applied to 20 samples of breast milk from different regions of the Rio Grande do Sul state, Brazil, of which 75% contained residues below the LOQs.