A comparison of finite elements for nonlinear beams: the absolute nodal coordinate and geometrically exact formulations

Two of the most popular finite element formulations for solving nonlinear beams are the absolute nodal coordinate and the geometrically exact approaches. Both can be applied to problems with very large deformations and strains, but they differ substantially at the continuous and the discrete levels. In addition, implementation and run-time computational costs also vary significantly. In the current work, we summarize the main features of the two formulations, highlighting their differences and similarities, and perform numerical benchmarks to assess their accuracy and robustness. The article concludes with recommendations for the choice of one formulation over the other.     

负压洁净室相对负压和绝对压差的换算

对测量与室外不直接相通房间的绝对压差，不必直接测量室内外压差，可以通过换算来实现。本文就压差换算提出一种方法并举出一个实例。     

一类时滞直接控制系统绝对稳定的充要条件

讨论了一类时滞直接控制系统ｘ（ｔ）＝Ａｘ（ｔ）＋Ｂｘ（ｔ－τ）＋ｂｆ（σ（ｔ））,σ（ｔ）＝ｃＴｘ（ｔ）的绝对稳定性,获得了其一阶形式绝对稳定的充要条件的代数判据,对ｎ阶系统作了相应的推广．     

Quantitative PCR for the specific quantification of Lactococcus lactis and Lactobacillus paracasei and its interest for Lactococcus lactis in cheese samples

The first objective of this work was to develop real-time quantitative PCR (qPCR) assays to quantify two species of mesophilic lactic acid bacteria technologically active in food fermentation, including cheese making: Lactococcus lactis and Lactobacillus paracasei. The second objective was to compare qPCR and plate counts of these two species in cheese samples. Newly designed primers efficiently amplified a region of the tuf gene from the target species. Sixty-three DNA samples from twenty different bacterial species, phylogenetically related or commonly found in raw milk and dairy products, were selected as positive and negative controls. Target DNA was successfully amplified showing a single peak on the amplicon melting curve; non-target DNA was not amplified. Quantification was linear over 5 log units (R² > 0.990), down to 22 gene copies/μL per well for Lc. lactis and 73 gene copies/μL per well for Lb. paracasei. qPCR efficiency ranged from 82.9% to 93.7% for Lc. lactis and from 81.1% to 99.5% for Lb. paracasei. At two stages of growth, Lc. lactis was quantified in 12 soft cheeses and Lb. paracasei in 24 hard cooked cheeses. qPCR proved to be useful for quantifying Lc. lactis, but not Lb. paracasei.     

Stereology of tubular structures

The morphology of a tubular structure can be characterized, in at least some of its important respects, through stereological methods. We study the geometric meaning of standard stereological quantities when applied to tubular structures, with particular regard to their curvature or tortuosity. Measures defined specifically in relation to tubular structures are also introduced for practical use. The ideal smooth bent cylinder, used here, is not realistic but provides general principles from which a more specialized investigation should be developed. The emphasis is placed on simple counting methods of measurement, specifically the tangent counts introduced by DeHoff, and counts of section profiles exhibiting a particular character (such as ‘figure-eights’). These measures convey information about the tortuosity of tubular structures, whereas the standard indices V_v, S_v, M_v give no information about tortuosity. Some data from human testicular tubules are discussed.     

采用法布里-珀罗腔的绝对距离干涉系统的方案分析

论述了线性调频绝对距离干涉测量方案的基本原理以及拍频测量中存在的问题;分析表明,在激光器调频范围确定的情况与,可以通过测量起点与终点拍频信号相位差的方法,实现１０＾－５的相对测量精度;简述了采用Ｆ－Ｐ腔实现高精度频率基准的方法,并给出了测量系统的原理图。     

基于标记适应的人脸年龄识别优化算法

缺乏完整和足够的人脸年龄标记数据集是当前人脸年龄识别问题中最突出的挑战之一。由于相近年龄的面部具有相似性,因此在年龄识别中可以学习并利用相邻年龄的面部信息,将每张人脸图像看作相关年龄的标记分布(LD),从而有效缓解了训练和测试数据集不足的问题。但是在不同年龄阶段,人的面部衰老变化速度显著不同,如儿童和老年时期面部变化较快,而中年时期面部变化较平缓,当前常用的LD方法存在模式单一的缺点,难以适应不同年龄阶段人脸特征的变化规律。为提高人脸年龄识别算法的通用性,提出一种深度学习框架下基于标记适应的人脸年龄识别优化算法IFAR-LA。引入标记适应机制,能够更好地学习特征表示,挖掘人脸图像数据中丰富的语义信息,从而有效提取不同年龄阶段面部变化的特征,大幅提升表示学习能力和泛化能力。改进后的标记适应算法能够适应不同年龄阶段人脸变化规律,使每幅人脸图像在学习其真实年龄和相关年龄的时候都能发挥作用,缓解训练数据不足的问题,同时,提升了算法普适性,能够适应不同年龄阶段人的面部衰老变化速度。在公开的人脸数据集MORPH和FG-NET上的实验结果表明,IFAR-LA算法相比改进前的人脸年龄识别算法平均绝对误差分别降低了6.5%和11.5%。     

高维数据挖掘中特征选择的稳健方法

针对高维数据的特点,即数据中变量个数往往大于样本观测数目,并且数据往往具有异质性特点,基于众数回归分析和变量选择降维技术,提出了一种稳健有效的特征选择方法,利用局部二次逼近算法(LQA)和最大期望(EM)算法,给出估计算法和最优调节参数的选取方法。通过实验的模拟数据分析表明,所提出的特征提取选择方法整体优于基于最小二乘和中位数的正则化估计方法,特别当误差是非正态分布时,与已有方法相比具有较高的预测能力和稳健性。     

使用群体兴趣偏好度的协同过滤推荐

推荐系统是电子商务系统中最重要的技术之一,用户相似性度量方法是影响推荐算法准确率高低的关键因素。针对用户评分数据极端稀疏情况下传统相似性度量方法的不足,提出了一种基于群体兴趣偏好度的协同过滤推荐算法,根据群体兴趣偏好度来预测用户对未评分项目的评分,在此基础上再采用传统的相似性度量方法计算目标用户的最近邻居。实验结果表明,该算法可以有效解决用户评分数据极端稀疏情况下传统相似性度量方法存在的问题,显著提高推荐系统的推荐质量。     

基于H.264码率控制的模型参数更新算法

为获得精确的模型参数,从率点选择和模型参数估计2个方面对H.264码率控制算法加以改进,利用一种基于曼哈顿和二维滑动窗口机制相结合的策略,实现高效的率点选择算法,采用加权的线性回归技术,实现有效的模型参数预测。实验结果表明,改进后的算法能改善视频编码质量,提高输出码率的控制精度。