Surface coating on carbon nanofibers with alumina precursor by different synthesis routes

Alumina-reinforced carbon nanofiber nanocomposites were prepared using different routes; powders mixture, colloidal route and sol-gel process followed by spark plasma sintering (SPS). CNFs/xAl₂O₃ (x = 10-50 vol.%) were prepared through nanopowders mixing in a high-energy attrition milling. The main limitations in the preparation of this kind of nanocomposites are related to the difficulty in obtaining materials with a homogeneous distribution of both phases and the different chemical nature of CNFs and Al₂O₃, which causes poor interaction between them. A surface coating of CNFs by wet chemical routes with an alumina precursor is proposed as a very effective way to improve the interaction between CNFs and Al₂O₃. An improvement of 50% in fracture strength was found for similar nanocomposite compositions when the surface coating was used. The improved mechanical properties of these nanocomposites are caused by stronger interaction between the CNFs and Al₂O₃.     

Discharge studies in micromegas detectors in a 150 GeV/c pion beam

A detailed study of several Micromegas detectors prototyped for the COMPASS and CLAS12 experiments is presented. Using a 150 GeV/c pion beam, the discharge probability was measured for several detector variants including bulk and non-bulk Micromegas. A detector equipped with an additional GEM foil as pre-amplification stage was also tested. A resistive coating of the readout strips was found to reduce the amplitude of the discharge by at least two orders of magnitude which was below the detection limit of the experimental setup. The effects of the micro-mesh type and material were investigated as well as the influence of the drift gap. Response in the presence of a 1.5 T transverse magnetic field was also studied. The measurements presented were performed during a RD51 beam test period.     

一种海量数据快速聚类算法

为满足海量数据处理要求,提出了一种基于网格的K-means快速聚类算法（SPGK）.设计基于网格质心的聚类簇个数选取算法,对数据进行网格划分得到每个网格的质心,将质心作为K-means聚类的样本点,从而减少K-means的欧氏距离计算次数.该算法基于Spark平台实现并行计算,进一步地提高了算法的运行效率.SPGK不但能够获得良好的聚类效果,而且缩减了欧氏距离计算次数,适用于海量数据的快速聚类.在千万级数据集上的实验结果表明,SPGK的性能明显优于现有的K-means++和基于K均值聚类的递归划分方法.     

一种Spark作业配置参数智能优化方法

Spark的配置参数对作业运行性能有较大影响,针对配置参数种类多、参数搜索空间大、参数间相互影响导致人工配置参数调优效率低下的问题,提出了一种Spark作业配置参数智能优化方法。本文首先在Spark众多配置参数中选择对作业运行性能影响较大的关键配置参数,建立典型Spark作业的运行数据集,利用支持向量回归算法,构建作业性能预测模型,通过改变数据集的规模,对比分析了模型预测值和作业的真实运行时间,模型评估指标证明了作业性能预测模型的有效性和准确性。其次基于作业性能预测模型,设计并实现了基于爬山算法、模拟退火算法、递归随机搜索算法以及粒子群算法的配置参数优化算法,并对几种算法的求解质量进行对比分析,实验表明递归随机搜索算法在几种不同类型的作业上收敛结果较优且标准差较小,证明该算法对不同类型作业的适应性较强、稳定性较好。将本文的智能优化配置与传统经验优化配置相比,实验结果表明,智能优化配置为典型Spark作业分别带来了4%、15%、22%的平均性能提升,证明智能优化配置能够高效地获取到具备较好作业适应性的配置,提升作业运行性能。     

基于Hive的空气质量大数据查询优化方法

为了提高对环境空气质量监测系统中省级环境监测中心站里已汇集的海量监测数据的统计和分析效率,提出了一种基于Spark 集群在Hive上进行多维数据分区的查询优化方法。以湖北省环境监测中心站中的空气质量监测数据为研究对象,将数据转移到Spark集群利用Spark SQL连接Hive并进行分区存储。设计了12种查询,查询4个数据集,通过与采用传统查询方法的实验对比得出结论。实验结果表明:基于Hive的分区优化方法对空气质量大数据的查询时间有47%到96%的优化,而随着查询的复杂程度和数据量的增加,该方法的优化效果越明显。     

Morphology and magnetic properties of Sm2Co7/α-Fe nanocomposite magnets produced by high energy ball milling and spark plasma sintering

In this article,the Sm₂Co₇/α-Fe nanocomposite magnets were prepared by high energy ball milling and spark plasma sintering method.The effect of soft phase content on the magnetic properties was studied.Up to 30 wt% α-Fe was added into Sm₂Co₇ matrix without the decrease of remanence.Optimal energy product(BH)max of 9.2 MGOe was obtained with 20 wt% α-Fe.TEM observation shows that the grain size of α-Fe is 20-50 nm which ensures a good coupling effect between soft and hard phase.One more thing needs to be mentioned is that there exists inter-diffusion between Sm-Co phase and α-Fe phase.Moreover,our results can also illustrate that the Sm₂Co₇/α-Fe nanocomposite magnets are able to acquire better magnetic properties than the SmCo₅/α-Fe magnets prepared by the same process due to the large domain width of Sm₂Co₇ phase.     

Alumina–Carbon Nanofibers Nanocomposites Obtained by Spark Plasma Sintering for Proton Exchange Membrane Fuel Cell Bipolar Plates

There is an increasing demand of multifunctional materials for a wide variety of technological developments. Bipolar plates for proton exchange membrane fuel cells are an example of complex functionality components that must show among other properties high mechanical strength, electrical, and thermal conductivity. The present research explored the possibility of using alumina–carbon nanofibers (CNFs) nanocomposites for this purpose. In this study, it was studied for the first time the whole range of powder compositions in this system. Homogeneous powders mixtures were prepared and subsequently sintered by spark plasma sintering. The materials obtained were thoroughly characterized and compared in terms of properties required to be used as bipolar plates. The control on material microstructure and composition allows designing materials where mechanical or electrical performances are enhanced. A 50/50 vol.% alumina–CNFs composite appears to be a very promising material for this kind of application.     

Heat of combustion of tantalum-tungsten oxide thermite composites

The heat of combustion of two distinctly synthesized stoichiometric tantalum-tungsten oxide energetic composites was investigated by bomb calorimetry. One composite was synthesized using a sol-gel (SG) derived method in which micrometric-scale tantalum is immobilized in a tungsten oxide three-dimensional nanostructured network structure. The second energetic composite was made from the mixing of micrometric-scale tantalum and commercially available (CA) nanometric tungsten oxide powders. The energetic composites were consolidated using the spark plasma sintering (SPS) technique under a 300 MPa pressure and at temperatures of 25, 400, and 500 °C. For samples consolidated at 25 °C, the density of the CA composite is 61.65 ± 1.07% in comparison to 56.41 ± 1.19% for the SG derived composite. In contrast, the resulting densities of the SG composite are higher than the CA composite for samples consolidated at 400 and 500 °C. The theoretical maximum density for the SG composite consolidated to 400 and 500 °C are 81.30 ± 0.58% and 84.42 ± 0.62%, respectively. The theoretical maximum density of the CA composite consolidated to 400 and 500 °C are 74.54 ± 0.80% and 77.90 ± 0.79%, respectively. X-ray diffraction analyses showed an increase of pre-reaction of the constituents with an increase in the consolidation temperature. The increase in pre-reaction results in lower stored energy content for samples consolidated to 400 and 500 °C in comparison to samples consolidated at 25 °C.     

Statistical analysis of electrostatic spark ignition of lean H2/O2/Ar mixtures

The concept of minimum ignition energy (MIE) has traditionally formed the basis for studying ignition hazards of fuels. However, the viewpoint of ignition as a statistical phenomenon appears to be more consistent with the inherent variability in engineering test data. We have developed a very low-energy capacitive spark ignition system to produce short sparks with fixed lengths of 1-2 mm, and the ignition system is used to perform spark ignition tests using a range of spark energies in lean hydrogen-oxygen-argon test mixtures used in aviation safety testing. The test results are analyzed using statistical tools to obtain probability distributions for ignition versus spark energy. A second low-energy spark ignition system was also developed to generate longer sparks with varying lengths up to 10 mm. A second set of ignition tests is performed in one of the test mixtures using a range of spark energies and spark lengths. The results are analyzed to obtain a probability distribution for ignition versus the spark energy per unit spark length. Preliminary results show that a single threshold MIE value does not exist, but rather that ignition is statistical in nature and highly dependent on mixture composition and spark length.