Thesis (M.S.)--University of Wisconsin--Madison, 1966. / eContent provider-neutral record in process. Description based on print version record.

Thèse (M.Sc.)--Université Laval, 2008. / Bibliogr.: f. 69-75. Publié aussi en version électronique dans la Collection Mémoires et thèses électroniques.

Thesis (M.S.) -- Worcester Polytechnic Institute. / Keywords: Electric Propulsion; Plasma thruster; MPD. Includes bibliographical references (leaves 6-13).

Les arcs électriques de défaut se produisant entre deux câbles endommagés peuvent donner lieu à un phénomène appelé " arc tracking ", qui se caractérise par le maintien et la propagation de cet arc le long des câbles, entraînant leur ablation progressive. Dans un réseau aéronautique, un tel défaut peut avoir de graves conséquences lorsqu'il apparaît au sein d'un toron de câbles, d'autant plus qu'il est susceptible de se transférer à une structure avoisinante. Un dispositif expérimental a été développé pour l'étude de ce phénomène. Il est constitué d'une alimentation conçue selon des spécifications propres au domaine aéronautique, permettant de générer un signal continu ou alternatif pendant une durée réglable. Afin de se rapprocher des conditions de vol (pression réduite à haute altitude), les essais sont effectués au sein d'une enceinte fermée, soit à 1 bar, soit à 95 mbar. Trois types de câbles sont testés, un en aluminium et deux en cuivre. En régime continu, ils sont reliés au + et au - de l'alimentation, et à deux phases en régime alternatif. Une plaque en aluminium, représentant la structure de l'avion et connectée au neutre du générateur en régime alternatif, est placée en regard de la partie dénudée des câbles, là où se produit l'arc, permettant un éventuel transfert de l'arc depuis les câbles vers cette plaque. Chaque test s'accompagne de l'acquisition systématique des signaux de courant et de tension, et deux capteurs de flux radiatif sont positionnés autour de l'arc. Une caméra rapide permet la visualisation du comportement de l'arc au cours de l'essai. Enfin, les échantillons de câbles sont pesés avant et après chaque test. Un bilan de puissance est réalisé pour les différentes configurations testées (courant, pression, câbles), afin d'estimer la répartition de la puissance totale entre les électrodes (fusion et vaporisation des câbles, conduction, rayonnement) et la colonne de plasma (rayonnement, conduction, convection). La part transférée aux électrodes est estimée au moyen de la valeur de la chute de tension aux électrodes et du courant. Une partie de cette puissance aux électrodes est à l'origine de la fusion et de la vaporisation des câbles (le matériau isolant comme le métal). Celle-ci est estimée grâce à un calcul thermodynamique à partir de la masse de câble ablatée, en considérant que la totalité a été fondue et que 1% a été vaporisé. La puissance transmise à la colonne d'arc est en partie rayonnée, mais le spectre de longueurs d'onde inférieures à 200 nm est absorbé dès les premiers microns d'air, et une estimation est effectuée à partir du calcul coefficient d'émission nette. Une partie importante de l'étude a porté sur le transfert de l'arc à la plaque en aluminium, en fonction de la distance entre les câbles et cette plaque et selon un critère de température en face arrière, mesurée par thermographie infrarouge. Il a ainsi été montré qu'à pression réduite l'arc est beaucoup plus diffus, et a tendance à s'accrocher à la plaque en aluminium à des distances environ deux fois plus grandes qu'à pression atmosphérique. Cependant, la puissance totale disponible étant inférieure d'environ 20%, les dégâts à la plaque en aluminium sont moins importants. / Fault electric arcs occurring between two damaged wires may lead to a phenomenon called "arc tracking", which consists in a maintaining and propagating arc along the cables, causing their progressive ablation. In an aeronautical network, such a fault may have serious consequences if it happens within a wiring harness, especially as it is likely to transfer to a nearby structure. An experimental setup has been developed to study this phenomenon. It consists of a power supply designed in accordance with aeronautical specifications, which enables to generate a direct or alternating signal during an adjustable time. In order to reproduce flight conditions in a better way (low pressure at high altitude), tests are carried out within a closed chamber, whether at 1 bar pressure, or at 95 mbar pressure. Three types of cables are tested, one mainly made of aluminium and two mainly made of copper. In direct regime, they are connected to the + and the - poles of the generator, and to two phases in alternating regime. An aluminium plate, representing the aircraft structure and connected to the neutral in alternating regime, is placed in front of the stripped part of the cables, where the arc occurs, enabling a potential transfer of the arc from the cables toward the plate. Every test involves a systematic acquisition of the current and voltage signals, and two radiative flux sensors are positioned around the arc. A fast camera enables a viewing of the arc behaviour during the test. Finally, the sample is weighed before and after each test. A power balance is realised for the various tested configurations (current, pressure, cables), in order to estimate the distribution of the total power between the electrodes (fusion and vaporisation of the cables, but also conduction and radiation) and the plasma column (radiation, conduction, convection). The amount of power that is transferred to the electrodes is estimated with the value of the electrode voltage drop and the current. Part of this power makes the cables melt and vaporise (the insulating material as well as the metal), it is estimated with a thermodynamic calculation based on the ablated mass of cable, considering that the whole has been molten, and 1% vaporised. The amount of power transmitted to the arc column is partly radiated, but the wavelength spectrum emitted below 200 nm is absorbed through the very first microns in air, and estimation is made involving the calculation of the net emission coefficient. A significant part of the study was related to the transfer of the arc toward the aluminium plate, as a function of the distance between the cables and that plate, and according to a temperature criterion at the back side, by means of infrared thermography. Thus, it was shown that at low pressure, the arc is much less localised, and tends to transfer to the aluminium plate at distances around twice bigger than at atmospheric pressure. However, the total available power is around 20% smaller, and the damage caused to the aluminium plate is less severe.

This thesis investigates several problems in social network analysis on misinformation prevention with an emphasis on finding solutions that can scale to massive online networks. In particular, it considers two problem formulations related to the spread of misinformation in a network that cover the elimination of existing misinformation and the prevention of future dissemination of misinformation. Additionally, a comprehensive comparison of several algorithms for the feedback arc set (FAS) problem is presented in order to identify an approach that is both scalable and computes a lightweight solution. The feedback arc set problem is of particular interest since several notable problems in social network analysis, including the elimination of existing misinformation, crucially rely on computing a small FAS as a preliminary. The elimination of existing misinformation is modelled as a graph searching game. The problem can be summarized as constructing a search strategy that will leave the graph clear of any misinformation at the end of the searching process in as few steps as possible. Despite the problem being NP-hard, even on directed acyclic graphs, this thesis presents an efficient approximation algorithm and provides new experimental results that compares the performance of the approximation algorithm to the lower bound on several large online networks. In particular, new scalability goals are achieved through careful algorithmic engineering and a highly optimized pre-processing step. The minimum feedback arc set problem is an NP-hard problem on graphs that seeks a minimum set of arcs which, when removed from the graph, leave it acyclic. A comprehensive comparison of several approximation algorithms for computing a minimum feedback arc set is presented with the goal of comparing the quality of the solutions and the running times. Additionally, careful algorithmic engineering is applied for multiple algorithms in order to improve their scalability. In particular, two approaches that are optimized (one greedy and one randomized) result in simultaneously strong performance for both feedback arc set size and running time. The experiments compare the performance of a wide range of algorithms on a broad selection of large online networks and reveal that the optimized greedy and randomized implementations outperform the other approaches by simultaneously computing a feedback arc set of competitive size and scaling to web-scale graphs with billions of vertices and tens of billions of arcs. Finally, the algorithms considered are extended to the probabilistic case in which arcs are realized with some fixed probability and a detailed experimental comparison is provided. \sloppy Finally, the problem of preventing the spread of misinformation propagating through a social network is considered. In this problem, a ``bad'' campaign starts propagating from a set of seed nodes in the network and the notion of a limiting (or ``good'') campaign is used to counteract the effect of misinformation. The goal is to identify a set of $k$ users that need to be convinced to adopt the limiting campaign so as to minimize the number of people that adopt the ``bad'' campaign at the end of both propagation processes. \emph{RPS} (Reverse Prevention Sampling), an algorithm that provides a scalable solution to the misinformation prevention problem, is presented. The theoretical analysis shows that \emph{RPS} runs in $O((k + l)(n + m)(\frac{1}{1 - \gamma}) \log n / \epsilon^2 )$ expected time and returns a $(1 - 1/e - \epsilon)$-approximate solution with at least $1 - n^{-l}$ probability (where $\gamma$ is a typically small network parameter). The time complexity of \emph{RPS} substantially improves upon the previously best-known algorithms that run in time $\Omega(m n k \cdot POLY(\epsilon^{-1}))$. Additionally, an experimental evaluation of \emph{RPS} on large datasets is presented where it is shown that \emph{RPS} outperforms the state-of-the-art solution by several orders of magnitude in terms of running time. This demonstrates that misinformation prevention can be made practical while still offering strong theoretical guarantees. / Graduate

Conventional physical vapour deposition (PVD) techniques usually result in films of lower quality than the corresponding bulk material. A major problem with PVD films is the presence of columns and voids throughout the thickness of the film. The films may have a low packing density, low micro-hardness and in many cases poor adhesion to the substrate. Many of these problems are a direct consequence of the low energy of the depositing atoms arriving at the substrate during film growth. The resulting film porosity gives rise to a reduction in mechanical strength, and in the case of dielectric optical films, a reduction in the refractive index. The properties of deposited films are greatly improved when the substrate or the growing film is bombarded with more energetic particles. An ideal deposition process requires a high flux of film atoms with an energy of approximately 5-50 eV in order to achieve sufficient surface mobility at the substrate to overcome columnar growth.

The super-giant Oyu Tolgoi porphyry copper-gold deposits in the South Gobi desert, Mongolia, consist of multiple discrete porphyry centers aligned within a north-northeast trending, >6.5 km long, arc-transverse mineralized corridor. The porphyries are linked to a tectono-magmatic event at ~372 Ma within a Devonian to Carboniferous volcanic arc, and U-Pb (zircon) geochronology records magmatic activity from ~390 Ma to ~320 Ma. The Oyu Tolgoi district underwent at least three discrete periods of syn- to post-mineral shortening and there is evidence for at least three unconformities within the Paleozoic sequence. Although the deposits were formed in an active orogenic environment characterized by rapid uplift, their preservation is a reflection of climactic effects as well insulation from erosion by rapid burial under mass-wasted and pyroclastic material in the volcaniclastic apron of late-mineral dacitic volcanoes. The porphyry copper-gold deposits are spatially and temporally related to medium- to high-K calc-alkaline quartz monzodiorite (~372 Ma) and granodiorite (~366 Ma) intrusive phases that comprise the Late Devonian Oyu Tolgoi Igneous Complex (OTIC). Adakite-like wholerock compositions as well as zircon grains with high CeN/CeN*, EuN/EuN* and Yb/Gd in the sample populations from syn- and late-mineral porphyry intrusions are different from younger intrusions that are not related to porphyry Cu-Au deposit formation. Moreover, mixed zircon populations within OTIC intrusions indicate that efficient assimilation of material from different host rocks by a convecting magma chamber occurred. Mafic to intermediate volcanic units evolved from tholeiitic to calc-alkaline compositions, which is interpreted to be a reflection of marine arc maturation and thickening. Felsic rock suites are dominantly high-K calc-alkaline, regardless of age. Nd-isotopic geochemistry from all suites is consistent with magma derivation from depleted mantle in an intra-oceanic volcanic arc and lead isotopic compositions indicate that the sulfides in the porphyry Cu-Au deposits are genetically linked to the Late Devonian magmas. Magma mixing, adakite-like magmatism and rapid uplift and erosion in a juvenile marine arc setting differentiate the ore-stage geologic environment at Oyu Tolgoi from other settings in active and fossil volcanic arcs. / Science, Faculty of / Earth, Ocean and Atmospheric Sciences, Department of / Graduate

Automation of welding systems is often restricted by the requirements of spatial information of the seams to be welded. When this cannot be obtained from the design of the welded parts and maintained using accurate xturing, the use of a seam teaching or tracking system becomes necessary. Optical seam teaching and tracking systems have many advantages compared to systems implemented with other sensor families. Direct vision promises to be a viable strategy for implementing optical seam tracking, which has been mainly done with laser vision. The current work investigated direct vision as a strategy for optical seam teaching and tracking. A robotic vision system has been implemented, consisting of an articulated robot, a hand mounted camera and a control computer. A description of the calibration methods and the seam and feature detection and three-dimensional scene reconstruction is given. The results showed that direct vision is a suitable strategy for seam detection and learning. A discussion of generalizing the method used as an architecture for simultanious system calibration and measurement estimation is provided.

The purpose of this thesis is to present a definition and some properties of a curved arc in a plane and to present a definition and some properties of the Jordan curve.

This thesis deals with arc detection using AFDD. Here is explained the general theory of electric arc. It also describes the principle of operation of the AFDD device, a specific type of instrument from OEZ s.r.o. and describe the test procedure for this instrument according to the technical standard. In the practical part, selected tests are performed and verified the function of the AFDD device in the event of a disturbance electric arc in the circuit.