The following report chronicles the University of North Texas Wind Turbine Project at Apogee Stadium. The timeline of events will include the feasibility study conducted by and for the university, grant awards from the Texas State Energy Conservation Office to fund the project, and a three-year sample of real time performance data since installation. The purpose of this case study is to compare the energy generation estimates by various stakeholders to the measured energy generation using a new but uniform performance relationship. In order to optimize energy generation in wind turbine generator systems, the most common wind speeds measured at the site should also be the most efficient wind speeds at which the wind turbine can convert the kinetic energy in the wind into mechanical energy and ultimately electrical energy. The tool used to convey this relationship will be a figure plotting the wind speed profile against the efficiency curve of the wind turbine. Applying this relationship tool to the UNT Apogee Stadium wind turbines provided valuable results. The most common wind speeds at Apogee Stadium are not the most efficient wind speed for the turbine. Also, the most common wind speeds were near the lower limit of the wind turbine’s performance parameters. This scenario was evident in both the energy generation predictions as well as the real-time recorded data. This case study will also present the economic analysis of the Apogee Stadium wind turbines using another tool that was not previously used in the feasibility study. The case study concludes with future steps to improve wind turbine performance, and to budget future cost using past, present and future energy savings.

No description available.

No description available.

Ce mémoire présente l'ensemble des étapes de pré-traitement appliquées aux images provenant de l'imagerie par résonance magnétique de diffusion afin de conseiller les meilleurs paramètres dans une étude de tractographie. L'imagerie de diffusion nous donne l'information locale des déplacements moyens des molécules d'eau dans le cerveau. Cette information nous permet d'inférer l'architecture de la matière blanche. La reconstruction du signal de diffusion fait appel à différentes méthodes plus ou moins aptes à restituer la complexité des configurations de fibres. Dans ce mémoire, nous proposons une nouvelle méthode de reconstruction du phénomène de diffusion basée sur la décomposition en ondelettes sphériques. Ensuite, en combinant ces informations à tous les points du cerveau nous reconstruisons le réseau de fibres de la matière blanche par un algorithme de tractographie déterministe. Afin d'initier cet algorithme, nous proposons une nouvelle méthode d'initialisation dans le but de mieux gérer la complexité des configurations de fibres au sein d'un seul voxel. Les fibres reconstruites sont très difficiles à évaluer dans le cerveau car nous ne connaissons pas la configuration réelle des fibres. Pour être en mesure d'évaluer nos méthodes de reconstruction, nous utilisons un fantôme calquant la complexité des configurations de fibres trouvées dans le cerveau. Dans ce mémoire, nous proposons un ensemble de métriques et un système de notations permettant d'évaluer automatiquement la qualité des résultats d'une tractographie. Nous concluons l'étude concernant les données synthétiques par un ensemble de conseils sur les paramètres à utiliser afin d'obtenir des résultats de tractographie optimaux. Finalement, nous évaluons qualitativement les résultats de tractographie issus de données réelles afin de confirmer nos choix sur les données fantômes.

L'objectif de cette thèse est d'étudier le comportement local d'une mesure de probabilité, notamment au travers d'un indice de régularité locale. Dans la première partie, nous établissons la normalité asymptotique de l'estimateur des kn plus proches voisins de la densité et de l'histogramme. Dans la deuxième, nous définissons un estimateur du mode sous des hypothèses affaiblies. Nous montrons que l'indice de régularité intervient dans ces deux problèmes. Enfin, nous construisons dans une troisième partie différents estimateurs pour l'indice de régularité à partir d'estimateurs de la fonction de répartition, dont nous réalisons une revue bibliographique.

This thesis presents a class of graphical models for directly representing the joint cumulative distribution function (CDF) of many random variables, called cumulative distribution networks (CDNs). Unlike graphical models for probability density and mass functions, in a CDN, the marginal probabilities for any subset of variables are obtained by computing limits of functions in the model. We will show that the conditional independence properties in a CDN are distinct from the conditional independence properties of directed, undirected and factor graph models, but include the conditional independence properties of bidirected graphical models. As a result, CDNs are a parameterization for bidirected models that allows us to represent complex statistical dependence relationships between observable variables. We will provide a method for constructing a factor graph model with additional latent variables for which graph separation of variables in the corresponding CDN implies conditional independence of the separated variables in both the CDN and in the factor graph with the latent variables marginalized out. This will then allow us to construct multivariate extreme value distributions for which both a CDN and a corresponding factor graph representation exist. In order to perform inference in such graphs, we describe the `derivative-sum-product' (DSP) message-passing algorithm where messages correspond to derivatives of the joint cumulative distribution function. We will then apply CDNs to the problem of learning to rank, or estimating parametric models for ranking, where CDNs provide a natural means with which to model multivariate probabilities over ordinal variables such as pairwise preferences. We will show that many previous probability models for rank data, such as the Bradley-Terry and Plackett-Luce models, can be viewed as particular types of CDN. Applications of CDNs will be described for the problems of ranking players in multiplayer team-based games, document retrieval and discovering regulatory sequences in computational biology using the above methods for inference and estimation of CDNs.

Approved for public release, distribution is unlimited / ine discrimination of similar soil classes was produced by the BRDF variations in the high-spatial resolution panchromatic image. Texture analysis results depended on the directionality of the gray level co-occurrence matrix (GLCM) calculation. Combining the different modalities of analysis did not improve the overall classification, perhaps illustrating the consequences of the Hughes paradox (Hughes, 1968) / Flight Lieutenant, Royal Thai Air Force

Cette thèse porte sur l'étude de la production d'ions négatifs (INs) en surface pour des applications à la fusion contrôlée. Les INs formés en surface d'un échantillon dans un plasma d'hydrogène à partir de bombardement par les ions positifs sont collectés et analysés en énergie par un spectromètre de masse (SM). Les fonctions de distribution en énergie des INs (FDIs) que mesure le SM sont différentes de celles émises par la surface f(E, Θ) du fait des modifications de trajectoires et d'énergie induites par la traversée du plasma et du SM. Afin de déterminer la FDI émis par la surface f(E, Θ) en se servant des fonctions de distribution mesurée par le SM f''(E) nous avons développé un modèle qui calcule les trajectoires des ions entre la surface et le détecteur du SM. A partir d'une fonction test f(E,Θ) il est alors possible de calculer f''(E) et de comparer à l'expérience. L'élément critique de cette méthode est le choix de f (E, Θ). Le choix effectué dans cette thèse est la fonction de distribution des neutres calculée par le logiciel SRIM pour les particules pulvérisées et rétrodiffusées lors de bombardement d'une surface dans des conditions similaires aux conditions expérimentales. Les résultats du modèle montrent un bon accord entre les FDIs expérimentales et les fonctions calculées, et valident nos calculs et le choix de SRIM.Dans le but de comparer les mécanismes de production et les rendements des INs, une étude comparative a été réalisée sur différents matériaux carbonés. Les FDIs mesurées présentent la même forme à la température ambiante. / This thesis deals with negative ions (INs) surface production for applications in controlled fusion. Negative ions (NIs) formed at the sample surface from positive ions bombardment in hydrogen plasma are collected and analyzed with energy mass spectrometer (MS). The NI energy distribution functions (NIDF) measured by the MS are different from those emitted from surface f(E, Θ) due to modifications trajectories and energies which result when NI cross plasma and MS. In order to determine the NIDF emitted by the surface f(E,Θ) using the NIDF measured by MS f''(E), we developed a model that calculates the ion trajectories between the surface and MS detector. Then from a test function f(E,Θ) it is possible to calculate f''(E) and compare it to the experimental one. The critical issue is this method is the choice of f(E, Θ). The approach used in this thesis is the neutral backscattered and sputtered distribution function calculated by SRIM software during a surface bombardment similar to the experimental conditions. The model resulting show a good agreement between experimental and calculated NIDF, and validate our calculations and the choice of SRIM.In order to compare production mechanisms and NIs yields, a comparative study on different carbons materials was performed. Measured NIDFs show the same shape at room temperature. This indicates that the mechanisms involved in the NI production and the contribution of these mechanisms in the NIDF are the same for all materials. The best NI yield at low temperature is observed on DLC surface. The highest NI yield for all temperatures is observed on Boron doped diamond (BDD) surface at 400°C.

A resistência à ruptura de materiais frágeis como o concreto de granulometria fina é fortemente dependente de sua heterogeneidade, do tamanho da amostra e da heterogeneidade das tensões. A partir de uma abordagem probabilista, fundamentada no modelo de Weibull, este trabalho trata da influência de tais efeitos sobre a resistência à tração de painéis compósitos formados por concreto de granulometria fina e telas soldadas. Desde que as armaduras se mantenham em regime elástico, mostra-se que o modelo de Weibull de dois parâmetros pode ser empregado com sucesso na previsão da ruptura dos painéis compósitos em estudo. Os parâmetros foram então identificados com medidas de ensaios de flexão em três e em quatro pontos de barras de concreto de granulometria fina de dois tamanhos diferentes. A utilização do modelo de Weibull confirmou, por outro lado, a validade da teoria de misturas como ferramenta para estimar as resistências de rupturas locais correspondentes a diferentes níveis de formação de fissuras / The rupture strength of brittle materials such as the microconcrete is strongly dependent on the microstructural heterogeneity, the sample size and the stress heterogeneity. Using a probabilistic approach, based on the Weibull distribution, this work deals with the influence of such effects on the tensile strength of composite panels formed by microconcrete and welded wire grids. Provide the reinforcements behaves in elastic regime, it is shown that the model of Weibull of two parameters can be used with success to predict the failure of the composite panels studied. The parameters were then identified from measurements of flexural tests in three and in four points of two different sizes of microconcrete samples. The feasibility of the Weibull model validates, as well, the rule of mixture as a satisfactory tool to estimate local failure strengths of related to different levels of cracking

Com a crescente demanda de modelagem de riscos dependentes, enfatizamos neste trabalho a teoria de cópulas e algumas medidas de dependência tais como coeficiente de correlação linear, coeficiente de correlação de Spearman. Mostramos algumas interpretações errôneas sobre o coeficiente de correlação linear e como podemos realizar simulações de variáveis aleatórias com determinadas marginais e dependência. Realizamos uma aplicação na área de seguros para determinar o capital alocado da seguradora. / With the growing demand for modeling dependent risk, in this study we emphasize the theory of copulas and some measures of dependence such as linear correlation coefficient and Spearman correlation coefficient. We show some misleading interpretations on the linear correlation coefficient, and how we can perform simulations of random variables with some marginals and dependence. We conduct an application in the insurance area to determine the allocated capital of the insurer.