Les suspensions colloïdales ont été étudiées par simulations numériques à partir de deux modèles : la dynamique Brownienne (BD) et la SRD-MD (Stochastic Rotation Dynamics - Molecular Dynamics). Ces études ont consisté à reprendre des travaux existants pour les porter sur GPU, tout en cherchant différentes optimisations possibles adaptées à ces simulations. Une amélioration de la recherche de voisinage de la littérature a pu être utilisée pour toutes ces simulations de type BD. Une simulation de SRD-MD avec couplage de force qui n'avait pas encore été parallélisée sur GPU dans la littérature, a été implémentée en utilisant un nouveau schéma de décomposition adapté à cette simulation, améliorant considérablement les performances. Ces simulations ont pu donner lieu par la suite à des études sur des suspensions colloïdales plus complexes : une hétéroagrégation entre deux suspensions avec des particules de même taille, une hétéroagrégation entre deux populations de colloïdes de tailles très différentes, et en dehors des suspensions colloïdales, une simulation de nanoalliages. Enfin, le modèle de SRD a été adapté afin d'être utilisé dans le cadre d'animation physique de fluide réaliste dans le contexte de l'informatique graphique. Des adaptations du modèle pour y incorporer des notions comme la gestion de la compressibilité, de la tension de surface ont dues être apportées. Des premiers résultats ont pu permettre de réaliser quelques simulations, dont une chute d'eau dans une verre. / Colloïdal suspensions have been studied by means of numerical simulation, using two physical models : Brownian dynamics and Stochastic Rotation Dynamics - Molecular Dynamics. These studies consist in parallizing colloïdal simulations from previous studies on GPU, and find some new optimisations for these specific simulations. An improvement of the neigborhood search has been implemented in all our BD type simulations. A SRD-MD with force coupling have been implemented for the first time in the literature, using a new decomposition scheme, which improves significantly its performances. Then, theses simulations have been adapted to study more complex colloidal suspensions : an interfacial heteroaggregation of colloidal suspensions, a heteroaggregation between two types of particles with a large size ratio, and outside this context, a nanoalloy simulation. Finally, the SRD model has been adapted to realistic fluid animtion from computer science context. Theses adaptations require to add to SRD model, the notion of compressibility and surface tension. First results have been released, like a pouring water into a glass simulation.

O Brasil é uma referência mundial na produção e exportação de citros, entretanto esse cultivo pode sofrer diversos problemas e perdas de produtividade por motivos diversos, como por exemplo, pragas. Para reduzir os riscos e perdas, torna-se interessante o uso de sistemas automatizados de monitoramento, justificando a necessidade de realizar a coleta de dados para determinar diversos fatores. Determinadas plantações, como a de citros, não podem ser monitoradas somente via solo ou somente via imagens aéreas, tornando necessário mesclar ambas as abordagens de acordo com o parâmetro a ser monitorado. Para a realização desse monitoramento devem ser utilizados robôs com habilidades distintas, robôs aéreos e robôs terrestres. Assim, é preciso designar as tarefas que cada robô realizará e também coordenar todos os robôs durante a execução do sistema como um todo, visando otimizar o processo de coleta de dados. Esse problema pode ser analisado e modelado como um problema de alocação de tarefas para robôs (Multi-Robot Task Allocation (MRTA)). Para resolver esse problema propõe-se um framework baseado em técnicas de cobertura de conjuntos e em mecanismo de mercado baseado em leilão. Teste simulados são realizados e demonstram que a presente proposta cumpre o papel na alocação das tarefas aos robôs. Além disso, visando a aplicação da solução proposta é projetado e desenvolvido uma plataforma robótica aérea (quadrirotor) de baixo custo utilizando peças prototipadas. Para o controle de estabilidade dessa plataforma, propõe-se um modelo matemático de acordo com os parâmetros inerciais do quadrirotor. Esse quadrirotor é utilizado em diversas aplicações reais, mostrando que o projeto desenvolvido pode ser reproduzido e destinado a execução de tarefas reais, como por exemplo a coleta de dados na agricultura de precisão. / Brazil is a world reference in the production and export of citrus, although this crop can suffer several problems and losses of productivity for diverse reasons, as for example, pests. In order to reduce risks and losses, it is interesting to use automated monitoring systems, justifying the need to perform data collection to determine several factors. Certain plantations, such as citrus plantations, can not be monitored only via soil or only via aerial images, making it necessary to merge both approaches according to the parameter to be monitored. To perform this monitoring, robots with different abilities, such asunmanned aerial vehicle (UAV) and unmanned ground vehicle (UCV) should be used. Therefore, it is necessary to assign the tasks that each robot will perform and also to coordinate all the robots during the execution of the system as a whole, in order to optimize the process of data collection. The problem can be studied and modeled as a task allocation problem for robots (MRTA). To solve this problem we propose a framework based set covering techniques and auction-based market mechanism. Simulated tests are performed and demonstrate that the present proposal fulfills the role in assigning tasks to robots. In addition, aiming at the application of the proposed solution is designed and developed a low cost aerial robotic platform (quadrirotor) which use prototyped parts. This quadrirotor is used in several real applications, showing that the developed project can be reproduced and destined to perform real tasks, such as data collection in precision agriculture.

On Cu(111) surface and in interaction with a single hexa-tert-butylphenylbenzene moleculegear, the rotation of a graphene nanodisk was studied using the large-scale atomic/molecular massively parallel simulator molecular dynamics simulator. To ensure a transmission of rotation to the molecule-gear, the graphene nanodisk is functionalized on its circumference by tertbutylphenyl chemical groups. The rotational motion can be categorized underdriving, driving and overdriving regimes calculating the locking coefficient of this mechanical machinery as a function of external torque applied to the nanodisk. The rotational friction with the surface of both the phononic and electronic contributions is investigated. For small size graphene nanodisks, the phononic friction is the main contribution. Electronic friction dominates for the larger disks putting constrains on the experimental way of achieving the transfer of rotation from a graphene nanodisk to a single molecule-gear.

Forested shoreline is important perching habitat for bald eagles (Haliaeetus leucocephalus). Bald eagles hunt, feed and loaf on shoreline perches. I measured trees suitable for bald eagle perches along the northern Chesapeake Bay during 1990-1991 to determine the influence of shoreline perch tree availability on bald eagle distribution. The shoreline was divided into 250 x 50 m segments. A segment was considered used if at least 1 eagle had perched on it during 1985-1992. I determined the number of suitable shoreline perch trees, percent forest cover, and distance from the water to the nearest suitable perch tree for each segment. Segments along the Chesapeake Bay had an average of 1 suitable perch tree per 10 m of shoreline. Shoreline segments used by eagles had more suitable perch trees (P = 0.0008) and a larger percent of forest cover (P = 0.0008). Suitable trees on segments with eagle use were closer to water than suitable trees on segments without eagle use (P = 0.0087). The differences in segments with and without eagle use appear to be largely due to the lack of trees in marshes which were used only seldomly. Marsh had few suitable perch trees, less forest cover and a greater mean distance from water to the nearest suitable perch tree than the other land types (P < 0.0001). These factors are unfavorable for foraging eagles and most marsh segments (66.7%) were unused, probably for this reason. The number of suitable perch trees and the percent of forest cover were lower on developed areas than undeveloped, forested areas (P < 0.01 for both tests). Also the distance from water to the nearest suitable perch tree was greater on developed land than forested land (P < 0.01). Thus, development appears to decrease the availability of suitable shoreline perch trees when compared to forested areas. Logistic regression models were created to predict the probability of eagle use, given the conditions at the time of this study. Varying values of development density, percent forest cover, number of suitable perch trees and distance from water to the nearest suitable tree were inputs used in these models to create curves to predict eagle use under different conditions. These curves indicated that, for a given development density, the probability of eagle use increased as the number of suitable perch trees or percent forest cover on the segment increased. Also, for a given development density, the probably of eagle use increases as the distance to water decreases. / Master of Science

We studied piping plovers (Charadrius melodus Ord) on Assateague Island National Seashore, Maryland, during the 1988-1990 breeding seasons. The estimated breeding population declined from 25 pairs in 1988 to 14 pairs in 1990. Nest predation by red foxes (Vulpes vulpes) was high. Predator exclosures constructed around individual nests did not increase nest survival. Chick survival was higher in bay beach and island interior brood-rearing habitats than on the ocean beach. Our evidence supports the hypothesis that the availability of adequate food is driving the differences in survival observed among brood-rearing habitats. Chicks raised on the bay beach or island interior weighed more, had higher foraging rates, and spent a greater proportion of their time foraging than chicks reared on the ocean beach. Indices of invertebrate prey abundance indicated that insects were more abundant on the bay beach and island interior than on the ocean beach. Disturbance did not differ among brood-rearing habitats. Human disturbance was higher in 1990 than in previous years. Overall productivity was 0.71 chicks fledged/breeding pair, well below our estimate of the productivity needed to deter a population decline. Management efforts should focus on reducing nest predation and maintaining overwash access paths to high quality brood-rearing habitat. / Master of Science

Doctor of Philosophy / Department of Chemistry / Paul E. Smith / The force fields used in computer simulations play an important role in describing a particular system. In order to estimate the accuracy of a force field, physical or thermodynamic properties are usually compared with simulation results. Recently, we have been developing a force field which is called the Kirkwood-Buff Force Field (KBFF). This force field is established by transforming experimental data into Kirkwood-Buff (KB) integrals and then attempting to reproduce those KBIs with molecular dynamic (MD) simulations. Here we investigate a variety of intermolecular interactions in aqueous solutions through KB theory and molecular simulations. First, we describe a force field for the simulation of alkali halide aqueous solutions. These models are developed specifically to reproduce the experimentally determined Kirkwood-Buff integrals and solution activities as a function of molality. Additionally, other experimentally known properties including ion diffusion constants, relative permittivities, the densities and heats of mixing are reproduced by these models. Second, In an effort to understand the interactions which occur between amino acids in solution we have developed new force fields for simple amino acids and their analogs including glycine, betaine, β-alanine, dl-alanine, NH4Cl, NH4Br, N(CH3)4Cl, N(CH3)4Br, CH3NH3Cl, and CH3COONa. The new force fields reproduce the experimental Kirkwood-Buff integrals which describe the relative distribution of all the species in a solution mixture. Furthermore, it is shown that these simple amino acids can be understood in terms of the interactions of their functional groups and that, to a very good approximation, the transferability and additivity usually assumed in the development of biomolecular force fields appear to hold true. Third, an analysis of the effect of a cosolvent on the association of a solute in solution is presented by using the Kirkwood-Buff theory of solutions. The derived expressions provide a foundation for the investigation of cosolvent effects on molecular and biomolecular equilibria, including protein association, aggregation, and cellular crowding. Finally, in an effort to understand peptide aggregation at the atomic level we have performed simulations of polyglycine ((gly)n) using our recently developed force fields. Experimentally, the association of glycine polypeptides increases with n. Our force fields reproduce this behavior, and we investigated the reasons behind this trend. In addition to studying closed ensembles, we also simulate these systems in a semi-open ensemble that was designed to mimic cellular environments typically open to water, using a simple direct approach. The differences between the two ensembles are investigated and compared with our recent theoretical descriptions of aggregating systems using Kirkwood-Buff theory.

Muscle Dysmorphia (MD) is characterized by preoccupation with muscularity. Although there is a growing body of research concerning MD, there is a lack of research concerning the potential role exercise training environment has on the clinical features of MD. The purpose of this study was to compare MD symptomology in traditional strength-trained (TRAD) individuals to individuals training in communal high intensity functional training environments (HIFT). Participants were recruited from both types of facilities. Participants (N=376) completed online (Qualtrics) demographics survey and Muscle Dysmorphia Inventory (MDI). One-way ANOVA compared the effect of training environment on MDI scores among HIFT, TRAD, both HIFT and TRAD (BOTH), home gym (HOME), and “OTHER.” Training environment significantly affected MDI for the 5 environments [F (4, 345) = 3.765, p = .005, d = 0.737]. Mean score for TRAD (M = 111.73, SD = 20. 39, [107.78, 115.68]) was significantly higher than HIFT (M = 102.20, SD = 19.59, [99.17, 105.23]). MDI for BOTH (M= 107.06, SD = 18.01, [100.77, 113.34]), HOME (M = 108.89, SD = 22.80, [99.86, 117.90]), and OTHER condition (M = 108.19, SD = 22.43, [97.97, 118.40]) did not significantly differ from HIFT or TRAD. Results suggest training environment is correlated with levels of MD symptomology. Specifically, males and females with higher levels of MD symptoms prefer to train in a traditional training environment, which is potentially more conducive to facilitating and perpetuating MD symptomology. The results of this study provide insight into the social physique anxiety associated with MD, as participants with higher levels of MD symptoms do not prefer to train in a HIFT environment where training occurs communally and other gym members provide extrinsic motivation. Additionally, the results of the present study further our understanding into the psychopathology of MD in that the motivating factors related to aesthetics (high level of body focus) associated with a TRAD environment take precedence over the motivating factors relating to selfimprovement and the desire to increase functional fitness that is associated with a HIFT environment. Results may provide knowledge for creating optimal treatment programs for individuals with clinical MD.

Pagrindinis šio magistro baigiamojo darbo tikslas buvo apžvelgti kai kurių algoritmų ir jų kombinacijų pritaikymą daugiamačiams sekų šablonams analizuoti ir įgyvendinti algoritmą, gebantį tai atlikti. Buvo aprašyta FP-Tree medžio struktūra, kuri yra skirta kompaktiškai saugoti kritiniams (pvz., dažnai pasikartojantiems) duomenims, pateiktas FP-Growth algoritmas, galintis analizuoti tokią duomenų struktūrą ir rezultate pateikiantis visų dažnų elementų šablonų aibę. Pristatyta modifikuotų FP-Growth ir PrefixSpan algoritmų kombinacija – MD-PS-FPG algoritmas, pateikti kai kurių atliktų testavimų rezultatai, tolimesnių darbų pagrindiniai tikslai ir pan. / The main goal of this master final work was to present some of the algorithms and their combinations for the multidimensional sequence pattern mining and implement an algorithm, that is capable of doing that. FP-Tree, that is used to store critical (for example, often repeated) data, was described. FP-Growth algorithm, that can analyze FP-Tree structure and give frequent pattern set as a result, was presented. MD-PS-FPG algorithm – a combination of modified FP-Growth and PrefixSpan algorithms – was introduced. The results of some tests, further work objectives and other things were also presented.

In this thesis, first-principles calculations based on density functional theory have been employed to investigate metal hydrogen interaction in transition, p-block and rare earth metals. Furthermore, the accuracy of the stochastic quenching method was tested in describing the structure of amorphous Fe(1-x)Zrx. The investigated systems of transition metal hydrides are V-H and ScZr(CoNi)2-H. For V-H, the main focus of the studies is the effect that strain has on the potential energy landscape which governs the metal hydrogen interactions. The investigation has focused on how the properties of hydrogen occupancy in the interstitial sites changes with strain and also how the hydrogen atoms themselves exert strain on the vanadium structure to lower the energy. Results on diffusion, induced strain and zero-point energy are presented which all reveal the considerable difference between tetrahedral and octahedral site occupancy. Diffusion was studied by employing ab initio molecular dynamics simulations to obtain diffusion coefficients and to map the movement of the hydrogen atom. A description of hydrogen in vanadium is provided from a fundamental basis that is expected to be applicable to any lattice gas system. For ScZr(CoNi)2-H, the difference of hydrogen occupancy in various interstitial sites and the hydrogen-induced strain was also investigated through calculations of the change in total volume as a function of hydrogen concentration. The fundamental properties of metal hydrogen bonding were investigated by studying the Zintl phase hydrides that are constituted of the electropositive metal of Nd or Gd and the electronegative metal Ga. Mixing metals of very different electronegativity gives rise to an intricate potential energy landscape in which the incorporation of hydrogen will have a big effect on both the electronic and atomic structure. From the theoretical side of the investigation, structural parameters are presented along with the density of states and Bader charge analysis to describe the hydrogen induced changes to the atomic and electronic structures. Finally, the accuracy of the stochastic quenching method in describing amorphous Fe(1-x)Zrx was evaluated by comparing simulated and measured EXAFS spectra. Once the structural agreement had been established the simulated structures were characterized through radial distribution functions and an analysis of the short-range order from Voronoi tessellation. The structural changes with respect to the composition parameter x were also evaluated.

Sequential experimentation is often employed in process optimization wherein a series of small experiments are run successively in order to determine which experimental factor levels are likely to yield a desirable response. Although there currently exists a framework for identifying optimal follow-up designs after an initial experiment has been run, the accepted methods frequently point to multiple designs leaving the practitioner to choose one arbitrarily. In this thesis, we apply preposterior analysis and Bayesian model-averaging to develop a methodology for further discriminating between optimal follow-up designs while controlling for both parameter and model uncertainty.