Indiana University-Purdue University Indianapolis (IUPUI) / A crucial task in computer vision and biomedical image applications is to represent images in a numerically compact form for understanding, evaluating and/or mining their content. The fundamental step of this task is the segmentation of images into regions, given some homogeneity criteria, prior appearance and/or shape information criteria. Specifically, segmentation of cells in microscopic images is the first step in analyzing many biomedical applications. This thesis is a part of the project entitled "Construction and profiling of biodegradable cardiac patches for the co-delivery of bFGF and G-CSF growth factors" funded by National Institutes of Health (NIH). We present a method that simultaneously segments the population of cells while partitioning the cell regions into cytoplasm and nucleus in order to evaluate the spatial coordination on the image plane, density and orientation of cells. Having static microscopic images, with no edge information of a cytoplasm boundary and no time sequence constraints, traditional cell segmentation methods would not perform well. The proposed method combines deformable models with a probabilistic framework in a simple graphical model such that it would capture the shape, structure and appearance of a cell. The process aims at the simultaneous cell partitioning into nucleus and cytoplasm. We considered the relative topology of the two distinct cell compartments to derive a better segmentation and compensate for the lack of edge information. The framework is applied to static fluorescent microscopy, where the cultured cells are stained with calcein AM.

Ways of rolling tubes on a continuous rolling mill considered in the work. Overview of three-roll continuous rolling mills are given. Technology for the production of pipes for pipe unit with 3 roller stand continuous mill developed. The geometric model of the process by scanning the tube sheet is constructed. Determination of the amount of deformation is made. Kinematically possible velocity field is constructed. Examples of calculations of a table rolling and modes of compression on a continuous three-roll mill is shown. / В работе рассмотрены способы прокатки труб на непрерывном стане, проведен обзор трехвалковых оправочных станов. Приведена технология производства труб на ТПА с 3-х валковой клетью непрерывного стана. Построена геометрическая модель обжатия путем развертки трубы в лист и проведено определение размера очага деформации. Построено кинематически-возможное поле скоростей. Приведены примеры расчетов таблицы прокатки и режимов обжатия на непрерывном трехвалковом стане.

In India, the natural energy resources (thermal and hydro) are unevenly distributed and are mostly present in the remote areas and the load centers are distributed across various regions of the country. Therefore high voltage lines have become necessary for the devel-opment of large interconnected power networks and for the reliable and economic transfer of power. The increase in electric power demand due to the electric load growth has lead to the expansion of the transmission systems to ultra high voltage levels. Presently, Ultra High Voltage (UHV) power transmission lines are being built to transfer large electric power to distant load centers from the generating stations. Increasing the line voltage increases the surge impedance loading, stability and the thermal capacity of the line. Lightning is one of the major causes for the line outages and interruptions of UHV power lines. A lightning strike generates a very large voltage leading to insulator puncture, melting, burning and pitting of conductors and the supporting hardware. Lightning can lead to transient over-voltages thus leading to ash-over in the power transmission lines which are dangerous for the power equipments as well as for the human beings working in the vicinity. Ground wires are used for the protection of overhead power transmission lines against a lightning stroke. The overhead ground wires are installed such that the lightning attaches to it and shunts the lightning current to the ground through the tower, thus protecting the phase conductors. Shielding failure happens when the lightning strikes the phase conductor instead of the ground wires. Lightning shielding failure is a major con-cern in UHV lines due to their large height, very high operating voltage and wide exposure area of the phase conductors. The lightning over-voltages injected on the phase conductor (shielding failure) nally reaches the substation causing serious threat to the substation components and can lead to temporary or permanent outage of the power transmission system. There have been cases of very high shielding failure ash-overs of UHV lines and thus lightning attachment to power transmission lines need to be studied in detail to reduce the power system line outages. Several models such as electro-geometric model (EGM) and leader progression model (LPM) have been developed to study the shielding failure of power transmission lines. EGM has been extensively used to obtain lightning attachment to power transmission lines but in recent years it is seen that EGM is unable to accurately predict the lightning attach-ment to UHVAC lines. The shielding failure rates obtained by EGM does not match with the observed shielding failure rate for UHV lines. For this reason LPM is considered to obtain lightning attachment to UHV lines but LPM in its initial stage do not deal with the detailed physics of the upward leader inception, i.e., corona inception and unstable as well as stable upward leader inception from the object on the ground. In this thesis a model for the lightning attachment has been developed based on the present knowledge of the lightning physics. The thesis mainly focuses on the modelling of upward leader inception and propagation for lightning attachment to UHV power trans-mission lines. Upward leader inception is modeled based on the corona charge present near the conductor region and the upward leader propagation model is based on the correlation between the lightning induced voltage on the conductor and the voltage drop along the upward leader channel. The present model considers corona inception and modelling of unstable and stable upward leader inception from the ground object for the analysis of the lightning attachment process. The upward leader inception model developed is compared with the previous inception models and the results obtained using the present and previous models are found to be comparable. Lightning striking distances ( nal jump) for various lightning return stroke current were computed for di erent conductor heights using present lightning attachment model. It is seen that the striking distance increases with the increase in lightning re-turn stroke current and increases with increase in conductor heights. The striking distance computed using the present model matches with the value calculated using the equation proposed by the IEEE working group for the applicable conductor heights of up to 8 m. The in uence of the conductor operating voltage, cloud electric eld, lightning down-ward leader lateral distance, conductor length, transmission line tower and conductor sag on the upward lightning leader inception are analysed and reported in the thesis. It is found that the lightning attraction to power transmission line increases with increase in conductor positive operating voltage and decreases with increase in conductor negative op-erating voltage. The presence of transmission line tower reduces the lightning attachment to the conductor lines and the probability of lightning strike decreases with the increase in downward leader lateral distance from the conductor lines. The present lightning attachment model is applied to study the shielding failure of UHV power transmission lines rated for 1200 kV ac (delta and horizontal con guration) and for 800 kV dc (with and without a dedicated metallic return conductor) and thereby the lightning shielding failure ash-over rate is computed for the UHV power transmission lines. It is seen that the lightning shielding rate for UHV power transmission lines depend on the lateral distance of the downward leader channel, instantaneous 50 Hz voltage on the transmission line conductor, height of the transmission line conductor, induced voltages on the conductor and the lightning return stroke current.

Les environnements CAO fournissent des outils puissants pour la programmation graphique et la manipulation d’entités géométriques complexes. Dans cette thèse, nous proposons d’exploiter ce potentiel dans le domaine de la conception de robots parallèles. Ces robots sont considérés comme une alternative intéressante vis-à-vis de leurs homologues sériels dans différentes applications comme le « pick and place » et l’usinage. Cependant, leur utilisation industrielle est encore restreinte en raison d’un espace de travail limité, de modèles géométriques difficiles à résoudre et l’existence de configurations singulières délimitant leur domaine d’exploitation. L’analyse et la caractérisation de l’espace de travail jouent alors un rôle fondamental dans la phase de conception de robots manipulateurs parallèles. Dans ce travail de thèse, nous proposons des approches géométriques originales donnant lieu à un ensemble de méthodes et techniques basées CAO pour l’analyse et la caractérisation de l’espace de travail de robots parallèles plans et spatiaux. L’espace de travail est généré comme un solide dans l’environnement CAO à partir d’un paramétrage géométrique, d’esquisses et d’opérations élémentaires telles que le balayage hélicoïdal et l’intersection. Nous avons montré que ces méthodes constituent des outils pertinents et efficaces d’aide à la conception des mécanismes parallèles. Ils permettent également la résolution du problème géométrique direct et la génération de trajectoires libres de singularités. Plusieurs types de manipulateurs ont été considérés dans ce travail pour mettre en avant et illustrer les techniques CAO / Géométriques proposées : robots parallèles plans à 3 degrés de mobilité de type 3-RPR, 3-RRR, 3-PPR et 3-PRR, robots parallèles spatiaux à 6 degrés de mobilité de type ou 3-CRS ou 3-PRRS. / CAD environments provide very powerful tools for graphical programming and manipulation of complex geometric entities. In this thesis, we propose to exploit such potential in the design of parallel robots. These robots are considered an attractive and important alternative towards their serials counterparts in various applications, like “pick and place” and machining. However, their industrial applications are restricted due to limited workspace, complexity related to resolution of the direct geometric model, and in addition the existence of the singular configurations which bound their application field. The analysis and the characterization of the workspace therefore play an essential role in the design phase of parallel robotic manipulators. In this thesis, we suggest original geometric approaches giving rise to a set of methodologies and techniques based on the use of CAD in order to analyze and characterize the workspace of planar and spatial parallel robotic manipulators. Workspace is generated as a solid in CAD environment by using a parametric geometric model, sketches, and elementary operations such as helical scanning and performing then Boolean intersection operation. We have shown in this thesis, that the proposed methodologies represent relevant and efficient tools which assist designers of parallel mechanisms. Moreover, they allow us to solve the direct geometric problem and to plan singularity-free trajectories. Several types of robotic manipulators have been considered in this work to highlight and illustrate the proposed CAD / Geometric techniques : planar parallel manipulators having three degree of freedom such as 3-RPR, 3-RRR, 3-PPR, and 3-PRR, and spatial parallel robotic manipulators having six degree of freedom 3-CRS-type.

Des modèles géométriques ont été proposés pour reconstruire la géométrie de plis associés aux rampes (par exemple pli sur flexure de faille), en identifiant en particulier la profondeur de niveau de décollement et le déplacement total sur la rampe. Ces méthodes de reconstruction géométrique sont appliquées pour des plis partiellement érodés. Au cours de l'érosion, le cut-off de la rampe peut être érodé et, par conséquent, le déplacement sur la rampe est difficile à quantifier. Dans cette thèse, nous développons onze modèles thermo-géométriques. Les modèles combinent les données géométriques et les données d’enfouissement pour proposer une évolution cinématique d’un pli avec cut-off érodé. Nous supposons que la mise en place d'une unité tectonique produit une anomalie thermique dans le mur de la faille, et que cette anomalie thermique pourrait indiquer une épaisseur de bloc chevauchant. Les modèles fournissent une estimation de la profondeur de décollement et le déplacement total sur une rampe érodée, qui ne dépend pas de taux d’érosion. Dans le cas de chevauchements actifs, les modèles proposent un taux de déplacement et un âge de l'initiation de la faille en fonction de taux d'érosion. Ces données sont utilisées pour proposer un développement cinématique de coupes érodées. Nous appliquons les modèles sur les plis érodés et actif à Taiwan dans les zones de Choshui et Miaoli. On propose des coupes régionales équilibrées en utilisant la technique de modélisation directe. Dans la section Choshui, nous proposons un niveau de détachement de ~5 km à ~14 km, marquée par deux sauts successifs de rampes de ~5 km and ~4 km. En supposant un taux d'érosion à 4 mm/an, l'âge de l’initiation de chevauchement active est entre 3,3 Ma dans la partie intérieure de prisme (Chevauchement de Tili) à 0,9 Ma dans la partie extérieur (Chevauchement de Chelungpu). Le raccourcissement totale sur la coupe de Choshui est ~100 km et le taux de déplacement calculé est ~1 cm/an. Pour tester nos modèles thermo-géométriques dans une chaîne plissée inactive, on applique nos modèles sur les plis érodés associés aux failles de Pine Mountain et Jones Valley dans la chaîne plissée des Appalaches. L'application des modèles thermo-géométriques nous permet d’estimer une quantité de déplacement sur les deux failles et expliquer de manière satisfaisante l'anomalie thermique dans le mur des failles de Pine Mountain et Jones Valley. Afin d'améliorer la description de l’anomalie thermique qui se développe dans le soubassement des failles, on a étudié l'évolution des minéraux magnétiques des roches argileuses le long de quatre sections dans la chaîne plissée à Taiwan. On a remarqué que la greigite (Fe3S4) domine l'assemblage magnétique dans les roches enfouies à moins à moins de de 70°C. La magnétite (Fe3O4) se développe pour des températures d’enfouissement de ~50°C et domine l’assemblage magnétique jusqu'à ~350° C. A partir ~300°C, la pyrrhotite monoclinique (Fe7S8) se développe aux dépens de la magnétite, et à ~350°C, la magnétite n'est plus détecté. Ces résultats peuvent être utilisés en complément d'autres géothermomètres pour identifier les anomalies thermiques dans une gamme de de 50-70°C et de 300-350°C où les caractéristiques des minéraux magnétiques sont identifiées / Geometric models have been proposed to account satisfactorily for ramp-related folds (e.g. fault-bend fold), identifying in particular detachment depth and total shortening. These methods of geometric reconstruction are applied on partially eroded folds. During erosion, the fault cut-off may be removed and as a result, the displacement is difficult to quantify. In this thesis, we develop 11 thermo-geometric models combining geometric description of folds and burial data to propose kinematic evolution of folds with eroded cut-offs. We assume that the emplacement of a tectonic unit will result in a thermal anomaly in the footwall, and that this thermal anomaly might indicate a thickness of the overriding unit. The models provide an estimation of the detachment depth and the total shortening on an eroded ramp, independent of the erosion rate. In the case of active thrusts, the models provide an estimation of the slip rate and the age of the initiation of the thrust as a function of the erosion rate. These data are used to unravel the kinematic development of eroded cross-sections. We apply the models on eroded folds from Taiwan underlined by active thrusts in the Choshui and Miaoli sections. We propose regional balanced cross-sections using forward modeling technique. In the Choshui section, we propose a detachment profile with a depth between ~ 5 km and ~ 14 km, marked by two steps of ~ 5 km. Assuming erosion rate at 4 mm/a, the age of initiation of the active thrusts is ranging from 3.3 Ma inward (Tili thrust) to 0.9 Ma outward (Chelungpu thrust). The total shortening from the whole section is ~100 km and the calculated slip rate is about 1 cm/a. To test our models in a non-active fold-and-thrust belt, we study eroded folds associated to the Pine Mountain thrust and Jones Valley thrust from the Appalachian belt. The application of the thermo-geometric models provides a value of the total shortening and explains satisfactorily the thermal anomaly in the footwall of the Jones Valley thrust. In order to improve the description of the thermal anomaly, we have studied the evolution of magnetic minerals of argillaceous rocks in four sections from the Taiwan thrust belt. We found that the iron sulfide greigite (Fe3S4) is dominating the magnetic assemblage in the less buried rocks (<70°C). The magnetite (Fe3O4) develops at burial temperature of ~50°C and is dominating the magnetic assemblage up to ~350°C. By ~300°C, the monoclinic pyrrhotite (Fe7S8) develops at the expense of magnetite, and at ~350°C, the magnetite is no longer detected. These results can be used complementary to other geothermometers to identify thermal anomalies in the range 50-70°C and 300-350°C where characteristic magnetic minerals are identified

Inspired by Gould, who claimed life would be arriving at a different outcome each time it were allowed to run from the same beginning, I have attempted to determine the repeatability of the adaptive course of one Pseudomonas fluorescens lineage. In addition, my study aimed to establish whether the likelihood of parallel evolution of the two synonymous single-nucleotide substitutions was contingent upon a prior motility-impairing deletion or a prior increase in fitness. Further, the study was designed to provide empirical data addressing the long-standing question of the effect of starting fitness on the ensuing rate of adaptation. Although no exact replay of the initial evolutionary trajectory was observed, I have demonstrated that gtsB, but not gtsC gene, is likely to be a mutational hotspot under the low glucose with a recovery of two undescribed mutations in gtsB. My data are consistent with a notion that substitutions in gtsB may be contingent upon Δ35kB(fliJ-PFLU4466) motility-impairing deletion, but not the fitness increase associated with it. Finally, the features of the adaptive landscape of P. fluorescens in the minimal glucose provide languid support for Fisher’s hypothesis of a decrease in adaptation rate with the rise in the starting fitness. Taken together, these original results reinforce the non-negligible role of history in shaping the outcomes of biological evolution and call for caution in attempting a formulation of rigid predictive models of evolutionary change. Inspiré par les travaux de Stephen J. Gould qui affirmait que la vie sur terre arriverait à une forme différente si elle repartait à zéro, je présente ici mes travaux où je teste la reproductibilité du cours adaptatif d’une lignée expérimentale de Pseudomonas fluorescens. L’objectif de cette étude était de déterminer si la probabilité que deux mutations synonymes évoluent en parallèle est affectée par la présence d’une délétion affectant la motilité de la bactérie ou de l’augmentation de la valeur sélective de celle-ci. De plus, le design expérimental de cette étude permet de tester si la valeur sélective initiale d’une population affecte le taux d’adaptation de cette même population. Bien d’une reproductibilité exacte du cours adaptatif initial ne fut pas observée, je démontre que le gène gtsB est probablement un « hotspot »mutationnel permettant l’adaptation à de bas niveau de glucose, ayant trouvé deux mutations dans ce site; alors que le gène gtsC ne l’est pas. Mes données sont également conséquentes avec le fait que les mutation dans le gène gtsB dépendent de l’effet de la délétion Δ35kB(fliJ-PFLU4466) affectant la motilité de la bactérie, mais non de l’augmentation de la valeur sélective qui y est associée. Finalement, la forme du plateau adaptative associé à de bas niveaux de glucose chez P. fluorescens supporte l’hypothèse émise par Fisher qui stipule que le taux d’adaptation d’un organisme diminue avec la valeur sélective initiale qui y est associée. L’ensemble de ces résultats supporte le rôle non-négligeable de l’histoire de vie d’une population en ce qui attrait à l’évolution future de cette même population. Aussi, ces résultats appelle à la prudence quand vient le temps de formuler des modèles prédictifs des changements évolutifs d’une population.

Inspired by Gould, who claimed life would be arriving at a different outcome each time it were allowed to run from the same beginning, I have attempted to determine the repeatability of the adaptive course of one Pseudomonas fluorescens lineage. In addition, my study aimed to establish whether the likelihood of parallel evolution of the two synonymous single-nucleotide substitutions was contingent upon a prior motility-impairing deletion or a prior increase in fitness. Further, the study was designed to provide empirical data addressing the long-standing question of the effect of starting fitness on the ensuing rate of adaptation. Although no exact replay of the initial evolutionary trajectory was observed, I have demonstrated that gtsB, but not gtsC gene, is likely to be a mutational hotspot under the low glucose with a recovery of two undescribed mutations in gtsB. My data are consistent with a notion that substitutions in gtsB may be contingent upon Δ35kB(fliJ-PFLU4466) motility-impairing deletion, but not the fitness increase associated with it. Finally, the features of the adaptive landscape of P. fluorescens in the minimal glucose provide languid support for Fisher’s hypothesis of a decrease in adaptation rate with the rise in the starting fitness. Taken together, these original results reinforce the non-negligible role of history in shaping the outcomes of biological evolution and call for caution in attempting a formulation of rigid predictive models of evolutionary change. Inspiré par les travaux de Stephen J. Gould qui affirmait que la vie sur terre arriverait à une forme différente si elle repartait à zéro, je présente ici mes travaux où je teste la reproductibilité du cours adaptatif d’une lignée expérimentale de Pseudomonas fluorescens. L’objectif de cette étude était de déterminer si la probabilité que deux mutations synonymes évoluent en parallèle est affectée par la présence d’une délétion affectant la motilité de la bactérie ou de l’augmentation de la valeur sélective de celle-ci. De plus, le design expérimental de cette étude permet de tester si la valeur sélective initiale d’une population affecte le taux d’adaptation de cette même population. Bien d’une reproductibilité exacte du cours adaptatif initial ne fut pas observée, je démontre que le gène gtsB est probablement un « hotspot »mutationnel permettant l’adaptation à de bas niveau de glucose, ayant trouvé deux mutations dans ce site; alors que le gène gtsC ne l’est pas. Mes données sont également conséquentes avec le fait que les mutation dans le gène gtsB dépendent de l’effet de la délétion Δ35kB(fliJ-PFLU4466) affectant la motilité de la bactérie, mais non de l’augmentation de la valeur sélective qui y est associée. Finalement, la forme du plateau adaptative associé à de bas niveaux de glucose chez P. fluorescens supporte l’hypothèse émise par Fisher qui stipule que le taux d’adaptation d’un organisme diminue avec la valeur sélective initiale qui y est associée. L’ensemble de ces résultats supporte le rôle non-négligeable de l’histoire de vie d’une population en ce qui attrait à l’évolution future de cette même population. Aussi, ces résultats appelle à la prudence quand vient le temps de formuler des modèles prédictifs des changements évolutifs d’une population.

Diploma thesis is focused on the description of the parameters of nonlinear material models of concrete, which are implemented in a computational system LS-DYNA, interacting with performance of nonlinear test calculations in system LS-DYNA on selected problems, which are formed mainly by simulations of tests of mechanical and physical properties of concrete in uniaxial compressive and tensile on cylinders with applying different boundary conditions and by simulation of bending slab, with subsequent comparison of some results of test calculations with results of the experiment. The thesis includes creation of appropriate geometric models of selected problems, meshing of these geometric models, description of parameters and application of nonlinear material models of concrete on selected problems, application of loads and boundary conditions on selected problems and performance of nonlinear calculations in a computational system LS-DYNA. Evaluation of results is made on the basis of stress-strain diagrams and load-displacement diagrams based on nonlinear calculations taking into account strain rate effects and on the basis of hysteresis curves based on nonlinear calculations in case of application of cyclic loading on selected problems. Verification of nonlinear material models of concrete is made on the basis of comparison of some results of test calculations with results obtained from the experiment.