Characteristics and classification of high-current fault arcs on distribution systems

When a fault occurs on transmission or distribution systems due to lightning or overvoltage, often an arc discharge occurs at the fault point. The arc discharge, which is caused by a fault current, has a high current, high temperature, strong light emission, etc., thus it sometimes causes heavy damages to electric power equipment. The arc discharge is influenced by the conditions around the arcs, i.e., gas, insulation materials, gap length, weather, etc. Also, the arc voltage along the arc column indicates the characteristics of the arc. If the voltage waveforms of the arcs caused by the fault on transmission or distribution systems are classified, it is possible to find the location and the equipment where the fault occurred. In this paper, the arc voltage data in 6-kV class XLPE cables and 6-kV class overhead lines are analyzed and an artificial neural network method is applied to classify the arc voltage waveforms. The results obtained from the six artificial neural networks developed show that the artificial neural network method is effective for classification of arc voltage waveforms if adequate input parameters are selected.     

Distance metric learning for graph structured data

Machine Learning - Graphs are versatile tools for representing structured data. As a result, a variety of machine learning methods have been studied for graph data analysis. Although many such...     

Hydroxyapatite crystal growth on calcium hydroxyapatite ceramics

Calcium hydroxyapatite (Ca-HAP) ceramics containing tricalcium phosphate (TCP) were soaked in three solutions: phosphate buffer, tris buffer, and simulated body fluid (SBF). Petal-like crystals of Ca-HAP were deposited on the Ca-HAP ceramics when (i) Ca-HAP ceramics contained -TCP, (ii) the soaking solution contained phosphate ion and (iii) the pH of soaking solution was higher than 7.3. These conditions facilitate the presence of HPO ₄ ^2– and Ca²⁺ ions, the latter from dissolution of -TCP. A well-defined X-ray diffraction pattern for the deposited Ca-HAP crystals indicates preferred growth of {002} planes. Slower crystal growth of Ca-HAP was found for SBF (pH=7.5) than in the phosphate buffer, due possibly to the lower phosphate ion content in SBF.     

Dehydration of Gels and Glasses in the Systems B2 O3-SiO2 and ZrO2-SiO2 Prepared by the Sol-Gel Process from Metal Alkoxides

Dehydration of gels prepared by the sol-gel process from metal alkoxides in the systems B₂O₃-SiO₂ and ZrO₂-SiO₂ was determined by measuring the shrinkage of the gel on heating. Dehydration was enhanced with increased ZrO₂ content, whereas it decreased with B₂O₃ content. Diffusion of water was also measured in the nonporous glasses obtained by heating the gels. The diffusion rate was independent of the composition of the glass.     

The electrocatalytic activities of the Pd-doped MnO2 for the chlorine evolution reaction

The anodic evolution of chlorine on the massive β-MnO₂ doped with various amounts of Pd was investigated in acidic chloride solution. A second-order relation was found between the rate of the chlorine evolution reaction and the number of active Pd sites at the electrode surface. And it was confirmed that, in the relatively low doped electrodes, the active sites are more effectively dispersed on the surface with microscopic homogeneity. From the activation energies, it was also suggested that the reaction mechanism on the MnO₂ doped with relatively small amounts of Pd is somewhat different from that on the pure PdO electrode.     

Local disturbances preceding a running crack front in a viscoelastic solid

The local disturbances ahead of a running crack front in a viscoelastic solid were measured through noncontact electro-optical tools. It was observed that the very high local strain rate up to 200/sec exists even in the present quasistatic loading case. It may be concluded that the running crack propagation velocities, not the conventional average strain rates ranging from about 10^?4/sec to 10^?2/sec, govern the local disturbances, as the crack propagation velocities always exceed 200m/sec for both average strain rates, while the local strain rates observed do not show much differences between both average strain rate cases.     

Aggregation arrestant pheromone of the German cockroach,Blattella germanica (L.) (Dictyoptera: Blattellidae): Isolation and structure elucidation of blattellastanoside-A and -B

The aggregation pheromone of the German cockroach,Blattella germanica, consists of attractant and arrestant, which can be detected by olfactometer and choice-chamber assay, respectively. Both were extracted from the frass-contaminated filter paper being used as a shelter. They were separated by solvent partition withn-butanol and water. The arrestant from then-butanol phase was purified by open column chromatography and then successive HPLC isolated two major arrestant components. Spectral evidence from SI-MS, HR-EI-MS, and NMR experiments with pulse techniques provided possible structures as 1-(6-chloro-4,5-epoxy-5-stigmast-3-yl)--d-glucopyranoside and 1-(6-chloro-5-hydroxy-5-stigmast-3-yl)--d-glu-copyranoside, denoted as blattellastanoside-A and blattellastanoside-B, respectively. They represented arrestant activity as median effective doses (ED₅₀) at 0.044 (A) and 3.2 (B) nmol on 1.0 cm² of Whatman No. 1 filter paper.     

Generalized measuring-worm algorithm: high-accuracy mapping and movement via cooperating swarm robots

Recently, many extensive studies have been conducted on robot control via self-positioning estimation techniques. In the simultaneous localization and mapping (SLAM) method, which is one approach to self-positioning estimation, robots generally use both autonomous position information from internal sensors and observed information on external landmarks. SLAM can yield higher accuracy positioning estimations depending on the number of landmarks; however, this technique involves a degree of uncertainty and has a high computational cost, because it utilizes image processing to detect and recognize landmarks. To overcome this problem, we propose a state-of-the-art method called a generalized measuring-worm (GMW) algorithm for map creation and position estimation, which uses multiple cooperating robots that serve as moving landmarks for each other. This approach allows problems of uncertainty and computational cost to be overcome, because a robot must find only a simple two-dimensional marker rather than feature-point landmarks. In the GMW method, the robots are given a two-dimensional marker of known shape and size and use a front-positioned camera to determine the marker distance and direction. The robots use this information to estimate each other’s positions and to calibrate their movement. To evaluate the proposed method experimentally, we fabricated two real robots and observed their behavior in an indoor environment. The experimental results revealed that the distance measurement and control error could be reduced to less than 3 %.     

Microscopic image restoration based on tensor factorization of rotated patches

In microscopic image processing for analyzing biological objects, structural characters of objects such as symmetry and orientation can be used as a prior knowledge to improve the results. In this study, we incorporated filamentous local structures of neurons into a statistical model of image patches and then devised an image processing method based on tensor factorization with image patch rotation. Tensor factorization enabled us to incorporate correlation structure between neighboring pixels, and patch rotation helped us obtain image bases that well reproduce filamentous structures of neurons. We applied the proposed model to a microscopic image and found significant improvement in image restoration performance over existing methods, even with smaller number of bases.     

Efficient Leave-<Emphasis Type="Italic">m</Emphasis>-out Cross-Validation of Support Vector Regression by Generalizing Decremental Algorithm

We propose a computationally efficient method for cross-validation of the Support Vector Regression (SVR) by generalizing the decremental algorithm of SVR. Incremental and decremental algorithm of Support Vector Machines (SVM) ^{2, 8, 9)} efficiently update the trained SVM model when a single data point is added to or removed from the training set. The computational cost of leave-one-out cross-validation can be reduced using the decremental algorithm. However, when we perform leave-m-out cross-validation (m > 1), we have to repeatedly apply the decremental algorithm for each data point. In this paper, we extend the decremental algorithm of SVR^{8, 9)} in such a way that several data points can be removed more efficiently. Experimental results indicate that the proposed approach can reduce the computational cost. In particular, we observed that the number of breakpoints, which is the main computational cost of the involved path-following, were reduced from \({\mathcal O}(m)\) to \({\mathcal O}(\sqrt{m})\).