Experimental determination of the internuclear separation of HD^+ ion micro-cluster

Measurements of zero-degree breakup fragment energy distribution fromthe Coulomb-Explosions of 1.50965 MeV HD+ ion inicro-cluster beam are reported.Mean value of the internuclear separation of HD+ is found to be 0.12510.003 urn. Aset of high-resolution experimental arrangement and improvement of the Van-de-Graaffaccelerator are described briefly.

SCMR:a semantic-based coherence micro-cluster recognition algorithm for hybrid web data stream

Data aggregation from various web sources is very significant for web data analysis domain.In addition,the recognition of coherence micro cluster is one of the most interesting issues in the field of data aggregation.Until now,many algorithms have been proposed to work on this issue.However,the deficiency of these solutions is that they cannot recognize the micro-cluster data stream accurately.A semantic-based coherent micro-cluster recognition algorithm for hybrid web data stream is proposed.Firsdy,an objective function is proposed to recognize the coherence micro-cluster and then the coherence micro-cluster recognition algorithm for hybrid web data stream based on semantic is raised.Finally,the effectiveness and efficiency evaluation of the algorithm with extensive experiments is verified on real music data sets from Baidu inc.and Migu inc.The experimental results show that the proposed algorithm has better recall rate than the non-semantic micro cluster recognition algorithm and single source data flow micro cluster recognition algorithm.

Approach to inter-satellite time synchronization for micro-satellite cluster

Micro-satellite cluster enables a whole new class of missions for communications, remote sensing, and scientific research for both civilian and military purposes. Synchronizing the time of the satellites in a cluster is important for both cluster sensing capabilities and its autonomous operating. However, the existing time synchronization methods are not suitable for microsatellite cluster, because it requires too many human interventions and occupies too much ground control resource. Although, data post-process may realize the equivalent time synchronization, it requires processing time and powerful computing ability on the ground, which cannot be implemented by cluster itself. In order to autonomously establish and maintain the time benchmark in a cluster, we propose a compact time difference compensation system（TDCS）, which is a kind of time control loop that dynamically adjusts the satellite reference frequency according to the time difference. Consequently, the time synchronization in the cluster can be autonomously achieved on-orbit by synchronizing the clock of other satellites to a chosen one＇s. The experimental result shows that the standard deviation of time synchronization is about 102 ps when the carrier to noise ratio（CNR） is 95 d BHz, and the standard deviation of corresponding frequency difference is approximately0.36 Hz.

Macro and micro mechanics behavior of granite after heat treatment by cluster model in particle flow code

In this paper, a cluster model in particle flow code was used to simulate granite specimens after heat treatment under uniaxial compression. The results demonstrated that micro-cracks are randomly distributed in the specimen when the temperature is below 300?C, and have partial coalescence when the temperature is up to 450?C, then form macro-cracks when the temperature is above 600?C. There is more inter-granular cracking than intra-granular cracking, and their ratio increases with increasing temperature.The micro-cracks are almost constant when the temperature decreases from 900?C to room temperature, except for quartz α–β phase transition temperature(573?C). The fracture evolution process is obviously affected by these cracks, especially at 600–900?C. Elevated temperature leads to easily developed displacement between the grains, and the capacity to store strain energy becomes weaker, corresponding to the plasticity of granite after heat treatment.

Approach to MAI cancellation for micro-satellite clusters

With the development of micro-satellite technology,traditional monolithic satellites can be replaced by micro-satellite clusters to achieve high flexibility and dynamic reconfiguration capability.For satellite clusters based on the frequency division-code division multiple access(FD-CDMA)communication system,the inter-satellite ranging precision is usually constrained due to the influence ofmulti-address interference(MAI).Themulti-user detection(MUD)is a solution to MAI,which can be divided into two categories:the linear detector(LD)and the non-linear detector(NLD).The general idea of the LD is aiming to make a better decision during the symbol decision process by using the information of all channels.However,it is not beneficial for the signal phase tracking precision.Instead,the principle of the NLD is to rebuild the interference signal and cancel it from the original one,which can improve the ranging performance at the expense of considerable delays.In order to enable simultaneous ranging and communication and reduce multi-node ranging performance degradation,this paper proposes an NLD scheme based on a delay locked loop(DLL),which simplifies the receiver structure and introduces no delay in the decision process.This scheme utilizes the information obtained from the interference channel to reconstruct the interference signal and then cancels it from the original delayed signal.Therefore,the DLL input signal-to-interference ratio(SIR)of the desired channel can be significantly improved.The experimental results show that with the proposed scheme,the standard deviation of the tracking steady error is decreased from 5.59 cm to 3.97 cm for SIR=5 dB,and 13.53 cm to 5.77 cm for SIR=-5 dB,respectively.

2种栽培模式对酿酒葡萄马瑟兰果际微域环境和果实发育的影响

【目的】研究2种栽培模式对新疆吐哈盆地产区酿酒葡萄马瑟兰果际微域环境和果实发育的影响,为产区酿酒葡萄栽培模式选择提供参考。【方法】以新疆吐哈盆地产区酿酒葡萄马瑟兰为试材,采用低‘厂’字形(结果高度40 cm,处理组)和高‘厂’字形(结果高度80 cm,对照组)2种模式栽培,比较分析2种栽培模式下果实生育期果际微域环境指标和果实发育指标变化及采收期果实品质、花色苷组分。【结果】与高‘厂’字形栽培相比,低‘厂’字形栽培果实生育期果际日最高温、日均温度分别降低1.66和2.21℃,月均超过35℃温差总和与月均≥35℃高温时长分别减小39.25℃和40.12 h;日最大湿度、日最小湿度和日均湿度分别提高2.89%、1.70%和2.64%,日均低湿时长减小11.56 h;叶幕透射辐射、土壤反射辐射和总辐射分别降低36.94%、50.38%和44.91%。低‘厂’字形栽培相比高‘厂’字形栽培更有利于果实发育,转色期浆果具有更高的酒石酸、苹果酸和抗坏血酸质量浓度,采收期果穗质量和果粒质量分别增大38.74%和25.61%,萎蔫率降低14.20%,可溶性固形物减小2.67°Brix,总酸质量浓度提高0.83 g/L,皮总黄烷醇含量和总花色苷含量分别提高23.83%和5.80%,但低‘厂’字形栽培降低了3′,5′-羟基取代类花色苷比例、花色苷甲基化修饰比例和花色苷总修饰比例。【结论】新疆吐哈盆地产区酿酒葡萄马瑟兰采用低‘厂’字形栽培能够改善果际微域环境,提高果实品质。

基于谈判博弈的微电网群多主体共享储能容量优化配置策略

大力发展新能源是构建新型电力系统、推动能源转型的必然选择。文章提出了基于谈判博弈的微电网群多主体共享储能容量优化配置方法。考虑微电网群运营商(MGCO)与共享储能运营商(SESO)之间的协同交互机制,建立基于共享储能的微电网群容量优化框架;考虑微电网之间的能量共享以及共享储能容量租赁,构建考虑能量共享的微电网群容量优化配置模型;计及共享储能动态租赁,构建SESO储能容量优化配置模型;在此基础上,构建基于谈判博弈的MGCO和SESO储能容量配置模型,并提出基于ADMM的分布式求解方法。基于MATLAB进行算例分析,结果表明,所提储能优化配置方法能够在降低MGCO成本的同时提高SESO的收益,达到了双赢的效果。

基于混合密度和微簇聚合的密度峰值聚类算法

密度峰值聚类算法是一种简单高效聚类新算法,但该算法在处理密度分布不均匀数据集时,很难找到正确的类簇中心,并且在样本分配过程中容易出现错误连带现象,导致聚类效果不佳。针对上述问题,提出一种基于混合密度和微簇聚合的密度峰值聚类算法(HMDPC)。HMDPC算法首先根据反向K近邻和样本间的归属关系定义样本的混合密度;其次,将数据划分为多个微簇,定义微簇之间的相似度,基于此相似度对多个微簇进行聚合,从而获得最终的聚类结果。在人工数据集和UCI数据集上进行实验,并将HMDPC算法与其它6种聚类算法比较,实验结果表明HMDPC算法聚类效果较好。

基于方向权值标签传播的微学习单元聚类算法

为帮助学习者从大量在线学资源中找到适合自身个性化的学习资源及顺序集合,提出了一种基于有向边方向权值的标签传播算法(label propagation algorithm on directed edge weights,LPADEW)用于发现适合特定学习者并属于同一学习周期的微学习单元序列簇群。该算法对标签传播算法进行两个改进:根据单元节点的利用度确定标签的更新顺序,降低在节点更新顺序上的随机性;利用当前单元节点的前置邻居和后置邻居的有向边权累加值进行标签更新,并将标签权重引入标签更新策略,既可降低标签更新的随机性,也可避免形成巨型簇群。实验结果表明,LPADEW算法在微学习真实数据集和人工数据集中均取得了较好的结果。

微电网群系统的分布式控制方法

为应对日益增长的微网群系统稳定性、经济性运行控制需求,提出了一种考虑并网和孤岛运行模式的微网群系统分布式控制方法。方法基于一致性算法实现微电网分布式分层控制,并考虑了微网群内各微网的协调控制。Simulink仿真结果表明,在负荷投切等情况发生时,工作在并网或孤岛模式的微电网群系统频率、电压波动在合理范围内,分布式电源始终按照等微增率原则分配出力。所提微电网群系统分布式控制方法能够维持微电网群的频率稳定和电压稳定,并通过分布式电源出力分配实现经济性目标。

LeaDen-Stream: A Leader Density-Based Clustering Algorithm over Evolving Data Stream

Clustering evolving data streams is important to be performed in a limited time with a reasonable quality. The existing micro clustering based methods do not consider the distribution of data points inside the micro cluster. We propose LeaDen-Stream (Leader Density-based clustering algorithm over evolving data Stream), a density-based clustering algorithm using leader clustering. The algorithm is based on a two-phase clustering. The online phase selects the proper mini-micro or micro-cluster leaders based on the distribution of data points in the micro clusters. Then, the leader centers are sent to the offline phase to form final clusters. In LeaDen-Stream, by carefully choosing between two kinds of micro leaders, we decrease time complexity of the clustering while maintaining the cluster quality. A pruning strategy is also used to filter out real data from noise by introducing dense and sparse mini-micro and micro-cluster leaders. Our performance study over a number of real and synthetic data sets demonstrates the effectiveness and efficiency of our method.

结合微聚类和主动学习的流分类方法

数据流分类是数据挖掘中重要的研究内容,但是数据流中的概念漂移和标记成本昂贵的问题给分类带来了巨大的挑战。现有的研究工作大多采用基于主动学习的在线分类技术,一定程度上缓解了概念漂移和有限标签的问题,但是这些方法的分类效率较低,并且忽略了内存开销的问题。针对这些问题提出了一种结合微聚类和主动学习的流分类方法(a data stream classification method combining micro-clustering and active learning,CALC)。提出一种新的主动学习混合查询策略,将其与基于错误的表示学习相结合,从而在维护过程中衡量每个微聚类的重要性,通过动态维护一组微聚类以适应数据流中产生的概念漂移。采用基于微聚类的惰性学习方法,实现对数据流的分类,并完成对缓存微聚类的在线更新。使用三个真实数据集和三个人工合成数据集进行实验,结果显示CALC在分类准确率和内存开销方面优于现有的数据流分类算法。与基准模型(online reliable semi-supervised learning on evolving data streams,ORSL)相比,CALC的分类准确率有一定的提升,在六个数据集上的平均准确率分别提高了5.07、2.41、1.04、1.03、3.47、0.64个百分点。

4款单一麦芽威士忌特征风味成分解析

以苏格兰麦卡伦和日本山崎麦芽威士忌为研究对象,采用顶空固相微萃取结合气相色谱-质谱联用技术解析4款威士忌中的风味化合物。结果表明:酯类、酸类和醇类化合物在威士忌挥发性化合物中的相对含量最高,超90%。在243种挥发性化合物中有58种具有气味活度值(odor activity value,OAV)。苏格兰麦卡伦威士忌中乙酸苯乙酯、金合欢醇、反式-橙花叔醇等物质OAV更高,赋予麦卡伦威士忌丰富的花香味。日本山崎威士忌中异戊酸乙酯、2-甲基丁酸乙酯、1-癸醇等物质OAV更高,赋予山崎威士忌更丰富的水果香气。二维层次聚类分析结果显示,山崎威士忌中典型风味物质是2-甲基丁酸乙酯、丁酸乙酯、异戊酸乙酯等,麦卡伦威士忌中典型风味物质是金合欢醇、反式-橙花叔醇、大马士酮等。主成分分析结果显示每款威士忌位于得分图的不同象限,表明每款威士忌有独特的风味特点。本研究结果表明,风味是评价不同威士忌的重要指标。

基于改进K-均值的微博热点话题发现方法

微博文本数据高维度、同义、多义特征明显,传统基于向量空间模型(VSM)联合K-均值的热点话题发现方法存在准确率低,计算复杂,聚类中心难以确定等问题。提出一种相关向量机(RVM)优化VSM的微博文本向量化方法,首先利用RVM的自适应特征选择能力对VSM特征向量进行降维,然后利用主成分分析(PCA)方法确定K-均值算法的初始聚类中心,进而采用K-均值算法得到聚类结果,最后根据微博转发、评论和高影响力用户数量定义热度指数,热度指数最大的话题即为当前热点话题。采用实际微博文本数据集开展实验,结果表明所提方法相对于2种传统方法的准确率分别提升7.3%和1.1%,实时性分别提升45%和53%。

基于自组织映射网络的微表情运动规律分析方法

表情是人类情感交互的重要方式。神经生理学研究表明:微表情很难由主观意识所控制,是人类真实情感的流露。与宏表情不同,微表情发生常常伴随着面部左右和上下的不对称移动,运动模式复杂。但由于微表情运动幅度小,人眼难以直接观察,微表情运动规律尚未被深入解析。在公共安保等领域,对微表情识别算法的可靠性与可解释性有很高要求。因此,本文旨在研究微表情运动规律分析方法,实现微表情运动规律的系统性解析。本文工作包括:研究基于自组织映射网络(Self-Organizing Maps, SOM)的微表情特征无监督聚类方法,得到微表情运动模式及其概率。定义微表情运动相似度指标,它从面部14个感兴趣区域运动趋势的角度来衡量两个微表情运动特征的差异程度,并作为调节SOM网络权值的依据。本文对CASMEII、SAMM、SMIC和MMEW数据集的微表情样本进行分析,并根据SOM网络的聚类结果总结微表情运动规律,该规律可以指导微表情识别算法特征提取,提升可靠性。

密度峰值聚类算法研究综述

密度峰值聚类(DPC)是一种新提出的基于密度和距离的聚类算法,由于其原理简单,无需迭代和能处理形状数据集等优点,正在数据挖掘领域得到广泛应用。但DPC算法也有着一定的缺陷,如:对截断距离参数敏感,初始聚类中心的选择非自动化,后续标签分配存在链式问题,时间复杂度较高等。文章对DPC算法的研究现状进行了总结与整理,首先介绍了DPC的算法原理和流程;其次,针对DPC算法的不足对DPC算法的优化进行概括和分析,指出了优化算法的核心技术以及优缺点;最后,对DPC算法未来可能面对的挑战和发展趋势进行展望。

基于模糊C均值聚类的混合动力汽车典型运行工况构建

为了优化某款混合动力汽车在实际道路工况下的燃油经济性,采用模糊C均值聚类算法构建了其在目标城市下的典型运行工况。与运行工况实测数据相比,其特征参数平均误差为4.22%;速度-加速度联合概率分布吻合较好;试验车型的Cruise仿真模型在典型运行工况下的百公里油耗相对误差为1.89%。此外,将典型运行工况与NEDC、WLTC进行对比。可知,典型运行工况具有平均车速低的特点,体现了目标城市道路工况与现行法规工况之间的差异。

磁控溅射法制备Cu/ITO薄膜及其耐蚀性能研究

采用磁控溅射技术在ITO基底上制备出了Cu薄膜,考察了溅射压强和溅射功率对Cu薄膜微观结构的影响,采用电子扫描显微镜、X射线衍射仪对薄膜的形貌和结构进行了表征,并采用电化学腐蚀实验研究了薄膜的耐蚀性能。结果表明:随着压强从3 Pa增加到8 Pa,薄膜的晶粒尺寸先增加后减小,结晶度先减弱后增强,压强的增大有利于表面粒子的扩散,使薄膜更加平整,但对表面粗糙度影响不大,压强为5 Pa时,晶粒尺寸较大。溅射功率对薄膜的结晶度和晶粒尺寸影响较大,随着功率从200 W增加到400 W,薄膜的结晶度逐渐增加,功率增加到400 W时,晶粒尺寸明显增加,此时薄膜表面粗糙度较大。晶粒尺寸的增加有利于增强薄膜的耐蚀性能,表面粗糙度的增加使薄膜耐蚀性减弱。在压强为5 Pa,功率为200 W和400 W时,所得薄膜耐蚀性较好,铜薄膜的耐蚀性由薄膜的晶粒尺寸、表面粗糙度等因素共同决定,二者对其耐蚀性影响呈现出一种竞争关系。

二阶自然最近邻和多簇合并的密度峰值聚类算法

密度峰值聚类(density peaks clustering, DPC)算法基于局部密度和相对距离识别簇中心,忽视了样本所处环境对样本点密度的影响,因此不容易发现低密度区域的簇中心;DPC算法采用的单步分配策略的容错性差,一旦一个样本点分配错误,将导致后续一系列样本点分配错误。针对上述问题,提出二阶自然最近邻和多簇合并的密度峰值聚类算法(TNMM-DPC)。首先,引入二阶自然邻居的概念,同时考虑样本点的密度与样本点所处的环境,重新定义了样本点的局部密度,以降低类簇的疏密对类簇中心选择的影响;其次,定义了核心点集来选取初始微簇,依据样本点与微簇间的关联度对样本点进行分配;最后引入了邻居边界点集的概念对相邻的子簇进行合并,得到最终的聚类结果,避免了分配错误连带效应。在人工数据集和UCI数据集上,将TNMM-DPC算法与DPC及其改进算法进行了对比,实验结果表明,TNMM-DPC算法能够解决DPC算法所存在的问题,可以有效聚类人工数据集和UCI数据集。

基于机器学习的超薄板焊接接头微尺度信息检测和识别

超薄板焊接对接头处的任一细微变化,均要求实时调节焊接热输入和焊枪对中,以保证焊接质量。因此,对超薄板对接接头微尺度信息的检测和识别尤为重要。本研究基于超薄板焊接对接接头的视觉传感检测,以焊接接头图像的像素点RGB值为试验对象,通过K-means聚类模型进行深度学习,训练后的二值化图像采用Canny算子提取轮廓,实现了100μm超薄板焊接对接接头不同微尺度信息的检测和识别,并计算出了焊接接头间隙宽度的细微变化和接头中心位置的偏移程度。试验结果表明,采用对像素点RGB值进行K-means聚类的方法,能更有效地分割焊接接头图像。进行一定次数的机器学习后,可以精确识别100μm厚对接接头的细微变化,通过Canny算子可以可靠提取由机器学习训练出的焊接接头二值化图像的细小毛刺、间隙直线度和间隙尺寸的细微变化,计算出的接头间隙宽度和中心位置偏移量,可为精确实时调节焊接热输入和焊枪对中提供有力支撑。