真空精炼系统故障诊断的模糊聚类分析

大型真空冶金系统 (DVMS)的故障点多、故障现象与故障原因间对应关系复杂 ,很难从抽气机理上来精确地进行故障诊断。DVMS工艺要求能对该设备进行快速、准确的故障诊断。为此 ,介绍了一种基于模糊聚类算法的智能诊断模型。该模型具有很强的自学习、自组织能力 ,适用于大型复杂真空系统的故障诊断。以RH KTB真空冶金系统的智能故障诊断为例 ,给出了真空冶金系统模糊故障诊断的实际过程。通过分析证实了该算法对大型复杂真空冶金系统智能故障诊断的有效性     

CABOSFV algorithm for high dimensional sparse data clustering

An algorithm, Clustering Algorithm Based On Sparse Feature Vector (CABOSFV), was proposed for the high dimensional clustering of binary sparse data, This algorithm compresses the data effectively by using a tool ‘Sparse Feature Vector‘, thus reduces the data scale enormously, and can get the clustering result with only one data scan, Both theoretical analysis and empirical tests showed that CABOSFV is of low computational complexity. The algorithm finds clusters in high dimensional large datasets efficiently and handles noise effectively.     

基于RBF-FCM-ROLS的砼安全性评估方法

提出了一种基于径向基函数神经网络RBFNN(Radial Basis Function Neural Networks)、模糊C-均值聚类(FCM)算法和递归正交最小二乘法(ROLS)的混凝土安全性专家系统评估预测方法。此方法先用FCM算法初选多个RBFNN的函数中心，再采用ROLS训练网络，最后结合后向选择法，减少初选的中心数目，以得到最终的有效中心值。该方法加快了RBFNN的训练速度，提高了网络的运算效率。将其运用到混凝土安全性评估专家系统中，获得了满意的结果。将这种新算法得出的评估数据与传统的BP网络计算出的数据进行了比较，进一步证明了RBFNN及其学习算法的优越性和实用性。     

被动单脉冲导引头干扰源角度分辨

讨论了被动单脉冲导引头对窄带高斯噪声干扰源的角度分辨．在导引头主波束内只有一个干扰源情况下，导引头给出的平均到达角指向干扰源真实到达角方向．在导引头主波束内同时存在两个相互独立干扰源情况下，通过平均到达角理论分析和计算机仿真，导引头给出的平均到达角指向两干扰源功率重心方向．利用复角技术，可判断导引头主波束内是单目标还是多目标．并提出一种利用复角技术和聚类分析相结合的两干扰源角度分辨方法，计算机仿真证实了方法的有效性和正确性．     

应用聚类有效性函数指导数据分类算法的研究

研究了应用聚类有效性函数指导数据分类的算法,指出文[6]中算法的不足并对其进行了修改。实验结果表明,修改后的算法是可行的。     

基于二进制数计算相似度的高属性维稀疏数据聚类方法

针对一种特定类型高属性维数据———区间变量型高属性维稀疏数据聚类问题,提出高属性维稀疏信息系统,稀疏特征编码,基于二进制数计算相似度概念,给出一种新的基于二进制数计算相似度的高属性维稀疏数据聚类算法,由于计算属性稀疏特征相似度所采用的是二进制数布尔AND运算,因此,相比目前人们所使用的聚类算法,它是一种计算简单、精度高、聚类质量较高的聚类算法。该算法在高属性维稀疏数据挖掘及聚类分析中有着重要的应用。通过数值算例分析表明该聚类方法有效。     

基于K均值的迭代局部搜索聚类算法

K均值聚类算法(KM)是解决聚类问题的一个常用的方法,该方法的主要缺点是其找到的局部极小值与全局最优值的偏差往往较大。论文构造一种基于KM算法的迭代局部搜索算法(称之为IKM)。该算法以KM算法所得到的解作为初始解,从该初始解开始作局部搜索,在搜索过程中接受部分劣解。当解无法改进时,算法对所得到的局部极小解做适当强度的扰动后进行下一次的迭代,以跳出局部极小,从而拓展了搜索的范围。试验结果表明IKM算法得到的聚类结果比KM算法得到的聚类结果有明显的改进,平均改进达100%以上。当数据集越大,簇的个数越多时,改进的效果越是显著,可以达到300%以上。因而,IKM算法是一个确实可行的有效的方法。     

一种新型的自适应人工免疫算法

论文简略叙述了免疫系统的基本原理,并对两种典型的免疫算法aiNet和RLAIS进行了详细分析,进而提出了一种新型的自适应人工免疫网络算法用于离散数据的聚类分析。此算法能得到稳定和良好的聚类结果,并能达到用户要求的数据压缩比,另外需用户确定的控制参数很少。     

Reliable Distributed Network Management by Replication

This paper presents a new clustering architecture for SNMP agents that supports semi-active replication of managed objects. A cluster of agents provides fault-tolerant object functionality: replicated managed objects of a crashed agent of a given cluster may be accessed through a peer cluster. The proposed architecture is structured in three layers. The lower layer corresponds to the managed objects at the network elements. The middle layer contains management entities called clusters that monitor and replicate managed objects. The upper layer allows the definition of management clusters as well as the relationship between clusters. A practical tool was implemented and is presented. The impact of replication on network performance is evaluated as well as a probabilistic analysis of replicated object consistency.     

Automatic localization of craniofacial landmarks for assisted cephalometry

In this paper we propose a system for localization of cephalometric landmarks. The process of localization is carried out in two steps: deriving a smaller expectation window for each landmark using a trained neuro-fuzzy system (NFS) then applying a template-matching algorithm to pin point the exact location of the landmark. Four points are located on each image using edge detection. The four points are used to extract more features such as distances, shifts and rotation angles of the skull. Limited numbers of representative groups that will be used for training are selected based on k-means clustering. The most effective features are selected based on a Fisher discriminant for each feature set. Using fuzzy linguistics if-then rules, membership degree is assigned to each of the selected features and fed to the FNS. The FNS is trained, utilizing gradient descent, to learn the relation between the sizes, rotations and translations of landmarks and their locations. The data for training is obtained manually from one image from each cluster. Images whose features are located closer to the center of their cluster are used for extracting data for the training set. The expected locations on target images can then be predicted using the trained FNS. For each landmark a parametric template space is constructed from a set of templates extracted from several images based on the clarity of the landmark in that image. The template is matched to the search windows to find the exact location of the landmark. Decomposition of landmark shapes is used to desensitize the algorithm to size differences. The system is trained to locate 20 landmarks on a database of 565 images. Preliminary results show a recognition rate of more than 90%.