Facile synthesis of uniform hollow hematite sub-micro spheres with controllable shell thickness

A simple method was developed to prepare the uniform hematite hollow submicro-spheres with controllable structure and different diameter based on monodisperse poly(styrene-co-acrylic acid) [P(St-co-AA)] particles. The structure and formation mechanism of the hollow spheres were investigated in detail. The control mechanism of shell thickness was also discussed. The results indicated that the shell thickness and coarseness of the synthesized core-shell hematite hollow spheres could be tuned simply by the surface carboxyl content of the P(St-co-AA) particles. This method provided a new approach for the structure control in the preparation of hollow spheres. A Brunauer-Emmett-Teller (BET) test shows that the prepared hollow spheres have large surface areas which were decreased along with the increase of the diameter. The magnetic properties of the as-obtained hematite hollow spheres were investigated. The result showed that the coercivity and saturated magnetization were increased along with the increase of the shell thickness, and the remanent magnetization was increased along with the decrease of the diameter.     

Development of a unified viscoplasticity constitutive model based on classical plasticity theory

The traditional unified viscoplasticity constitutive model can be only applied to metal materials. The study of the unified constitutive theory for metal materials has discovered the correlation between the classical plasticity theory and the unified viscoplasticity constitutive model, thus leading to the concepts of the classic plastic potential and yield surface in the unified constitutive model. Moreover, this research has given the continuous expression of the classical plastic multiplier and presented the corresponding constructive method, which extends its physical significance and lays down a good foundation for the application of the unified constitutive theory to the material analysis in more fields. This paper also introduces the unified constitutive model for metal materials and geo-materials. The numerical simulation indicates that the construction should be both reasonable and practical. Supported by the National Natural Science Foundation of China (Grant No. 90410012)     

基于流形学习的人体动作识别

目的 提出了一个基于流形学习的动作识别框架,用来识别深度图像序列中的人体行为。方法 从Kinect设备获得的深度信息中评估出人体的关节点信息,并用相对关节点位置差作为人体特征表达。在训练阶段,利用LE（Lalpacian eigenmaps）流形学习对高维空间下的训练集进行降维,得到低维隐空间下的运动模型。在识别阶段,用最近邻差值方法将测试序列映射到低维流形空间中去,然后进行匹配计算。在匹配过程中,通过使用改进的Hausdorff距离对低维空间下测试序列和训练运动集的吻合度和相似度进行度量。结果 用Kinect设备捕获的数据进行了实验,取得了良好的效果;同时也在MSR Action3D数据库上进行了测试,结果表明在训练样本较多情况下,本文方法识别效果优于以往方法。结论 实验结果表明本文方法适用于基于深度图像序列的人体动作识别。     

基于高精度流线生成的交互流场可视化

高效准确的流线绘制一直是流场可视化的重要研究内容,流线可以对流场的重要特征进行有效的稀疏表示,但流线需要长期的粒子追踪过程及大量的积分计算,在面向大规模流场可视化时时间效率较低,需要高性能计算设备进行辅助计算.本文通过设计一种基于深度学习的高精度流线生成算法,将初始的低精度流线快速映射为稠密的高精度流线,可以在较短的时间内快速生成可靠的流线可视化结果,并在此基础上设计了交互式实时流场可视化系统,涵盖了流场的特征检测,属性关联分析,信息论分析等,帮助用户快速了解流场数据,找到自己感兴趣的区域进行后续进一步深度分析,避免了获取过多冗余数据,同时优化了分析工作的效率,满足用户对于流场结构,特征属性等多维度进行关联分析的需求.     

基于多季相Sentinel-2影像的白洋淀湿地信息提取

白洋淀湿地是华北平原上重要的浅水湖泊湿地,对雄安新区绿色发展具有重要的生态价值。对白洋淀高度异质化的景观格局进行分类,能够为白洋淀湿地资源的遥感监测提供指导意义。针对湿地季节变化的特点,对白洋淀每个季节选取一期具有代表性的Sentinel-2影像,采用分类与回归树（CART）、支持向量机（SVM）、随机森林（RF）3种常用的机器学习分类器对15种季相组合实验方案进行分类,分析不同季相遥感影像及其组合对白洋淀湿地信息提取的优劣。结果表明：相较于使用单一季相影像分类,多季相影像的组合能够显著提高分类精度,春&夏季相组合能够得到最优的分类效果,相对单季影像总体分类精度提高了10.9%~25.5%,Kappa系数提高了0.09~0.29;SVM分类器的分类表现较为稳定,能够得到最高的平均分类精度,CART分类器在处理高维特征的能力不如随机森林和SVM;不同特征类型对湿地信息提取的贡献度从高到底依次是红边光谱特征、传统光谱特征、缨帽变换特征、主成分分析特征、纹理特征。实验成果能为湿地信息的遥感识别提供依据。     

Multi-objective robust design optimization of a novel negative Poisson’s ratio bumper system

Negative Poisson’s ratio (NPR) structure has outstanding performances in lightweight and energy absorption, and it can be widely applied in automotive industries. By combining the front anti-collision beam, crash box and NPR structure, a novel NPR bumper system for improving the crashworthiness is first proposed in the work. The performances of the NPR bumper system are detailed studied by comparing to traditional bumper system and aluminum foam filled bumper system. To achieve the rapid design while considering perturbation induced by parameter uncertainties, a multi-objective robust design optimization method of the NPR bumper system is also proposed. The parametric model of the bumper system is constructed by combining the full parametric model of the traditional bumper system and the parametric model of the NPR structure. Optimal Latin hypercube sampling technique and dual response surface method are combined to construct the surrogate models. The multi-objective robust optimization results of the NPR bumper system are then obtained by applying the multi-objective particle swarm optimization algorithm and six sigma criteria. The results yielded from the optimizations indicate that the energy absorption capacity is improved significantly by the NPR bumper system and its performances are further optimized efficiently by the multi-objective robust design optimization method.     

Machine-learning-aided precise prediction of deletions with next-generation sequencing

When detecting deletions in complex human genomes,split-read approaches using short reads generated with next-generation sequencing still face the challenge that either false discovery rate is high,or sensitivity is low.To address the problem,an integrated strategy is proposed.It organically combines the fundamental theories of the three mainstream methods(read-pair approaches,split-read technologies and read-depth analysis) with modern machine learning algorithms,using the recipe of feature extraction as a bridge.Compared with the state-of-art split-read methods for deletion detection in both low and high sequence coverage,the machine-learning-aided strategy shows great ability in intelligently balancing sensitivity and false discovery rate and getting a both more sensitive and more precise call set at single-base-pair resolution.Thus,users do not need to rely on former experience to make an unnecessary trade-off beforehand and adjust parameters over and over again any more.It should be noted that modern machine learning models can play an important role in the field of structural variation prediction.     

基于Pushover方法的大跨度悬挑钢梁结构的弹塑性分析

本文以某实际工程为例,采用静力弹塑性分析方法,通过SAP2000进行Pushover分析,研究大跨度悬挑梁在地震作用下的结构抗震性能。通过在多遇和罕遇地震荷载作用下对结构的顶点位移、层间位移角以及塑性铰发展过程的研究,分析其抗震性能。同时研究不同悬挑长度对结构抗震性能的影响规律并提出相应的加固措施,可为今后类似结构的设计与抗震分析提供参考。     

Microstructure and properties of cement foams prepared by magnesium oxychloride cement

Microstructural features including pore size distribution, cell walls and phase compositions of magnesium oxychloride cement foams (MOCF) with various MgO powders and water mixture ratios were studied. Their influences on compressive strength, water absorption and resistance of MOCF were also discussed in detail. The experimental results indicated that moderate and slight excess MgO powders (MgO/MgCl₂ molar ratios from 5.1 to 7) were beneficial to the formation of excellent microstructure of MOCF, but increasing water contents (H₂O/MgO mass ratios from 0.9 to 1.29) might result in opposite conclusions. The microstructure of MOCF produced with moderate and slight excess MgO powders could enhance the compressive strength, while serious excess MgO powders addition (MgO/MgCl₂ molar ratios = 9) would destroy the cell wall structures, and therefore decrease the strength of the system. Although MOCF produced with excess MgO powders could decrease the water absorption, its softening coefficient was lower than that of the material produced with moderate MgO powders. This might be due to the instability of phase 5, the volume expansion and cracking of cell walls as immersed the sample into water.     

聚羧酸助磨剂合成及其性能

利用水泥粉磨过程中的机械力促使水解聚马来酸酐（HPMA）与聚乙二醇单甲醚（MPEG）进行酯化反应形成聚羧酸助磨剂。试验研究了HPMA、MPEG、HPMA与MPEG的混合物、HPMA与MPEG的反应合成物对水泥助磨效果以及对砂浆性能的影响。结果表明,HPMA与MPEG都有一定的助磨性,HPMA与MPEG的混合物和反应合成物具有更佳的助磨性能,且可以改善水泥砂浆的流动性能和显著提高水泥砂浆的早期强度。激光拉曼光谱测试表明HPMA与MPEG的混合物和反应合成物与水泥共同粉磨后吸附在水泥颗粒表面,具有相似的拉曼位移图谱。