一种基于语义相似度的文本聚类算法

文本聚类在很多文本挖掘和信息检索系统中发挥着重要的作用。现有的聚类算法大多数都是基于向量空间模型,文档集合中出现的单词词频作为特征项。这些算法都存在数据维数过高、聚簇难以描述的问题,而且忽略了单词间的语义联系。本文提出了一种基于语义相似度的文本聚类算法——TCU SS(Text clustering usingsem an ticsim ilarity)算法。TCU SS算法将文档表示成概念列表,有效地解决了数据维数高和聚簇描述难的问题,并给出如何利用概念列表进行聚簇描述的方法。TCU SS算法利用两个概念列表中单词间的语义相似度作为文档间相近程度的度量,并以图为基础进行聚类分析,避免有些聚类算法对聚簇形状的限制。实验证明,TCU SS算法提高了聚类质量。     

Topology optimization of support structure of telescope skin based on bit-matrix representation NSGA-II

Non-dominated sorting genetic algorithm II（NSGA-II）with multiple constraints handling is employed for multi-objective optimization of the topological structure of telescope skin,in which a bit-matrix is used as the representation of a chromosome,and genetic algorithm（GA）operators are introduced based on the matrix.Objectives including mass,in-plane performance,and out-of-plane load-bearing ability of the individuals are obtained by fnite element analysis（FEA）using ANSYS,and the matrix-based optimization algorithm is realized in MATLAB by handling multiple constraints such as structural connectivity and in-plane strain requirements.Feasible confgurations of the support structure are achieved.The results confrm that the matrix-based NSGA-II with multiple constraints handling provides an effective method for two-dimensional multi-objective topology optimization.     

无线传感网的分布式非测距三维定位算法

为解决三维空间无线传感器网络节点定位问题,基于区域立体网格化表示的思想,提出一种分布式非测距三维定位算法3D-DRL(three-Dimensional Distributed Range-free Localization).通过对立体网格投票,选取得票值最高的所有网格的质心作为未知节点的估计位置.3D-DRL无需未知节点间相互通信,具有较小的通信开销,不依赖于锚节点比例,且对网络拓扑结构具有鲁棒性.仿真结果表明,在无线传播环境理想、未知节点通信半径R=50 m、所有节点均随机部署在100 m×100 m×100 m三维区域的情况下,定位误差小于未知节点通信半径的8%,且通过调整VANR,能够实现所有节点的定位.     

大型复杂装备研发成本控制专家系统

根据大型复杂装备研发周期长、耗资大、风险大的特点,引入前馈控制机制,建立了大型复杂装备研发项目成本控制专家系统的体系结构,同时以该体系结构为主体,给出专家系统的知识表示及推理机的一种实现方法,最终使得知识库和推理机得以独立实现.实现后的系统以知识库和专家系统相结合的方式对装备研发成本进行实时监控和预测,对于输入的问题,系统经过推理后给出相应的参考建议,为成本管理智能化提供可行依据.由于知识库和推理机是相对独立实现,因此该系统相对传统专家系统来说具有更好的可维护性和可靠性.     

作战命令的分词技术研究

作战命令的分词技术研究是作战命令自动理解的重要基础,就作战命令的分词问题提出了2种数学模型。利用数据融合方法建立了2个和3个成熟汉语分词系统构成的词法分析器,并通过仿真分析对2种方法进行了比较,最终得出采用多个成熟的汉语分词系统进行作战命令的分词是可行和有效的。     

基于稀疏信号表示的空间碎片成像

地基雷达一直是空间碎片监测的有力工具,研究如何利用它们对尺寸小于距离分辨率的目标进行成像具有重要意义。基于空间碎片通常会围绕它的主轴进行简单的自旋转运动这一事实,通过对目标回波模型的分析,本文提出了一种基于稀疏信号表示的空间碎片成像方法。该方法将成像问题转化为稀疏信号分解问题,首先以各个散射点的散射特性作为元素构建字典,然后通过采用现有的方法对单个距离单元横向回波数据的稀疏信号表示进行求解,最后对求解的结果重新排列以得到目标的二维高分辨图像。仿真实验结果表明了本方法的正确性。     

O群E(×)T1C-G系数详解

本文用传统方法计算了O群的克莱布施-戈登系数,结合坐标变换给出了克莱布施-戈登系数的数学意义。     

An improved approach to model ionospheric delays for single-frequency Precise Point Positioning

PPP with low-cost, single-frequency receivers has been receiving increasing interest in recent years because of its large amount of possible users. One crucial issue in single-frequency PPP is the mitigation of ionospheric delays which cannot be removed by combining observations on different frequencies. For this purpose, several approaches have been developed, such as, the approach using ionospheric model corrections with proper weight, the GRAPHIC (Group and Phase Ionosphere Calibration) approach, and the method to model ionospheric delays over a station with a low polynomial or stochastic process. From our investigation on the stochastic characteristics of the ionospheric delay over a station, it cannot be precisely represented by either a deterministic model in the form of a low-order polynomial or a stochastic process for each satellite, because of its strong irregular spatial and temporal variations. Therefore, a novel approach is developed accordingly in which the deterministic representation is further refined by a stochastic process for each satellite with an empirical model for its power density. Furthermore, ionospheric delay corrections from a constructed model using GNSS data are also included as pseudo-observations for a better solution. A large data set collected from about 200 IGS stations over one month in 2010 is processed with the new approach and several commonly adopted approaches for validation. The results show its significant improvements in terms of positioning accuracy and convergence time with a negligible extra processing time, which is also demonstrated by data collected with a low-cost, handheld, single-frequency receiver.     

基于ELMD与改进SMSVM的机械故障诊断方法

机械振动信号携带大量重要的机械状态信息,然而机械故障振动信号在复杂工作状态下通常呈现非平稳、非线性特性。因此,从振动信号抽取和选择有效的机械故障特征、提高故障识别性能,成为机械故障诊断研究的热点。针对上述问题,本文提出了基于集成局部均值分解(Ensemble local means decomposition,ELMD)与改进的稀疏多尺度支持向量机(Sparse multiscale support vector machine,SMSVM)的机械故障诊断方法。该方法首先使用自适应非线性、非平稳信号处理方法 ELMD把多模态调制故障信号分解成为多个单模态解调信号,有效地增强了故障特征。把压缩感知和多尺度分析技术融合于故障模式分类中,提出改进SMSVM旋转机械故障识别方法,提高多类机械微弱故障数据模式识别性能。该方法融合稀疏表示、多尺度分析和SVM的优点,无需求解复杂的优化问题,易于推广至更多尺度SVM,具有计算量少、泛化性与鲁棒性好、物理意义明显等优点。人工数据和实验设备数据验证了本文算法的优越性。     

基于稀疏和冗余表示的图像融合方法

为了产生能提供更多信息的融合图像,提出了一种基于SRR(Sparse and Redundant Represent,稀疏和冗余表示)理论的图像融合方法。该方法首先采用训练所得的冗余字典对原始图像进行稀疏表示,然后对系数矩阵采用1-范数取大融合规则进行融合,最后对融合系数矩阵和冗余字典进行重建以得到融合图像。将该方法与拉普拉斯变换、PCA(Principal Component Analysis,主成分分析)、DWT(Discrete Wavelet Transform,离散小波变换)、CVT(Cur Velet Transform,曲波变换)和NSCT(Non-Subsampled Contourlet Transform,非下采样轮廓波变换)等方法在几组图像上进行比较,实验结果表明,该方法在一定程度上提高了融合图像的质量,在主观和客观上都具有较好的性能。