Este trabalho de doutorado versa sobre a simulação numérica das etapas construtivas de pontes estaiadas através do método dos elementos finitos, utilizando a plataforma ANSYS, versão 14.5. A justificativa da realização deste trabalho vem da crescente utilização deste tipo de estrutura de ponte no Brasil, devido, principalmente, ao seu bom comportamento estrutural e pela possibilidade de uso de uma estrutura mais leve, esbelta e econômica. Inicialmente, foi feita uma pesquisa abrangente sobre os componentes estruturais e concepções de projetos mais utilizados. Em seguida, estudaram-se as relações constitutivas de cada material constituinte, a fim de melhor representá-los na simulação numérica. Para a representação das equações constitutivas do concreto, implementou-se um novo modelo de material visco-elastoplástico com fissuração, com a ajuda da ferramenta de customização UPF (User Programmable Features), onde foram adicionadas novas sub-rotinas ao programa principal em linguagem FORTRAN. Como o objetivo final desta tese era trabalhar com análise estrutural das etapas construtivas de pontes estaiadas, envolvendo um número muito grande de elementos finitos, a implementação deste novo modelo possibilitou a utilização de elementos tridimensionais quadráticos de 20 nós (SOLID186) com armadura incorporada (REINF264), tornando a solução do problema mais rápida e eficaz. A fim de validar as sub-rotinas acrescentadas ao sistema, foram simuladas dezesseis vigas em concreto armado, ensaiadas por Leonhardt e Walther (1962) e por Bresler e Scordelis (1963), e uma viga segmentada com protensão externa, ensaiada por Aparicio et al (2002). A comparação de resultados das análises numéricas e experimentais mostraram boa aproximação. Após a validação das sub-rotinas, foi modelada numericamente a Ponte do Saber, localizada na cidade do Rio de Janeiro, seguindo fielmente as etapas construtivas adotadas na construção da ponte. A partir dos dados de monitoramento desta ponte, durante a fase executiva, foi possível comparar os valores obtidos no modelo numérico com as informações coletadas em campo, obtendo-se bons resultados. / This doctoral thesis presents the numerical simulation of the construction stages of cable-stayed bridges using the finite element method, through the ANSYS software, version 14.5. This work is justified by the increasing use of such structures in Brazil, mainly due to its good structural behavior and the possibility of using a lighter, slender and economic structure. Initially, it presents the results of a research about the structural components and the commonly used design concepts. Then, the constitutive relations of each constituent material were studied, in order to best represent them in the numerical simulation. For the representation of the constitutive equations of concrete, it was implemented a new model of visco-elastoplastic material with cracking. That was made with the help of the customization tool UPF (User Programmable Features), where new subroutines were added to the main program in FORTRAN language. The final goal of this thesis is to work with structural analysis of the construction stages of cable-stayed bridges, involving a very large number of finite elements. Therefore, the implementation of this new model enabled the use of three-dimensional quadratic elements of 20 nodes (SOLID186) with embedded reinforcement (REINF264), making the solution of the problem faster and more effective. In order to validate the subroutines added to the system, sixteen reinforced concrete beams tested by Leonhardt and Walther (1962) and by Bresler and Scordelis (1963), and a segmental beam with external prestressing, tested by Aparicio et al (2002) were simulated. The comparison of results of numerical and experimental analyzes showed good approximation. After validation of the subroutines, it was numerically modeled the Saber Bridge, located in the city of Rio de Janeiro, following exactly the constructive steps taken in the construction of the bridge. Based on the monitoring data of the bridge during the construction phase, it was possible to compare the values obtained from the numerical model with the information collected in the field, obtaining good results.

Dans un scénario avec plusieurs réseaux sans fil de différentes technologies, ce travail a comme objectif la conception d'un moteur cognitif capable de reconnaitre l'environnement radio et de sélectionner un réseau avec le but final de maximiser la "qualité d'expérience" (QoE) de l'utilisateur. Un accent particulier est mis sur la simplicité de tous les éléments impliqués, du hardware aux algorithmes, afin de garder la faisabilité pratique de ce dispositif.Deux aspects ont été étudiés. Pour la reconnaissance de l'environnement radio une identification de réseau et une classification automatique sur la base de caractéristiques de la couche MAC a été proposée et testée. En ce qui concerne la sélection du réseau, des "Key Performance Indicators" (KPIs), qui sont des paramètres de la couche application, ont étés pris en compte afin d'obtenir la QoE désirée. Un modèle général pour la sélection du réseau a été proposé et testé avec de différents types de trafic par des simulations et par la réalisation d'un démonstrateur (application pour Android). De plus, comme il y a le problème de quand mesurer pour estimer la performance d'un réseau et quand l'utiliser effectivement pour transmettre et recevoir, le problème du bandit manchot ("Multi-armed bandit", MAB) a été appliqué à ce contexte et un nouveau modèle de MAB a été proposé afin de mieux répondre aux cas réels considérés. L'impact du nouveau modèle, qui introduit la distinction de deux actions différentes, mesurer et utiliser, a été testé par des simulations en utilisant des algorithmes déjà disponibles dans la littérature et deux algorithmes conçus spécifiquement. / In a scenario where multiple wireless networks of different technologies are available, this work addresses the problem of the design of a cognitive engine, core of a cognitive radio device, able to perform the surrounding radio environment recognition and the network selection with the final goal of maximization of final user Quality of Experience (QoE). Particular focus is put on the requirement of simplicity of all the elements involved, from hardware to algorithms, in order to keep in mind the importance of its practical realizability.Two aspects were investigated. For the surrounding radio environment recognition step, a network identification and automatic classification method based on MAC layer features was proposed and tested. As regards the network selection, Key Performance Indicators (KPIs), i.e. application layer parameters, were considered in order to obtain the desired goal of QoE. A general model for network selection was proposed and tested for different traffic types, both with simulations and a practical realization of a demonstrator (implemented as an application for Android OS). Moreover, as a consequence of the originated problem of when measuring to estimate a network performance and when effectively using the network for data transmission and reception purposes, the multi-armed bandit problem (MAB) was applied to this context and a new MAB model was proposed, in order to better fit the considered real cases scenarios. The impact of the new model, that introduces the distinction of two different actions, to measure and to use, was tested through simulations using algorithms already available in literature and two specifically designed algorithms.

I studien tas måttlistor fram för tre barnplagg med funktionsdetaljer. Plaggen graderas för att kontrollera att skillnadsintervallerna är rimliga att skicka vidare till produktion. Studien genomförs mot ett företag som uppgiften utformas tillsammans med och som bidrar med material i form av tyg, provplagg att utgå ifrån, samt handledning.Metoden för studien är huvudsakligen experimentell forskning i form av avmätning, konstruktion, gradering och sömnad. Underlag för den experimentella forskningen består av litteratur. Huvudfokus ligger på gradering då plaggen i studien omfattar ett stort storleksspann innehållande 12 storlekar vilka är 86-152 centilong. Utöver detta kommer funktionsdetaljer samt passform att behandlas i studien. I rapporten undersöks om det är rimligt att arbeta i ett stort storleksspann där rekommenderad indelning av barnstorlekar frångås. Utöver detta undersöks även om samma kragkonstruktion kan användas till två olika jackor samt hur raglanärm bör graderas utan att minska rörelsen i ärmen.Genom att först mäta av och utvärdera de plagg som företaget tillhandahåller uppnås resultat. Därefter konstrueras och sys plaggen med funktionsdetaljer upp för att slutligen graderas. Avslutningsvis skapas måttlistor för samtliga plagg som baserar sig på framtagen gradering och som slutligen överlämnas till företaget.The main purpose of this study is to create measurement charts for three garments for children with functional features. Garments are graded to control measurements before being sent to production. The study is created in collaboration with a company who contributes with fabric, garment samples and guidance.The method of this study is mainly experimental in form of construction, grading and sewing. Literature is used as reference for the experimental parts of the study. The main focus is grading the garment which involves 12 sizes, 86-152 centilong. Functional features and fit are also a part of the study.The report examines the equity of grading in a large size range that doesn’t correspond with the chosen literature. Furthermore the study examines if the same collar construction can be used for two different jackets and how a raglan sleeve should be graded without reducing the movement in the sleeve.Result is created by measuring and evaluating samples from the company. New garments are constructed, graded and sewn as final samples with functional features. Finally measurement charts are created based on the grading and are presented to the company. / Program: Designteknikerutbildningen

Perante a situação atual de degradação dos recursos hídricos, destacando os reservatórios destinados a usos múltiplos, e a necessidade de se buscar soluções para tal problema, foi proposto o presente trabalho. O reservatório de Barra Bonita, localizado no interior do Estado de São Paulo, foi escolhido para estudo, por estar inserido em uma área densamente povoada, industrializada e ocupada por agricultura intensiva. Os dados adquiridos foram de natureza cartográfica (cartas de pedologia, geomorfologia, geologia, drenagem e malha municipal); orbitais (imagens multiespectrais dos satélites Landsat 5 e 7); censitários (produção agrícola municipal e densidade demográfica); limnológicos (característicasfísicas, químicas e biológicas do reservatório estudado, obtidas em campo e na literatura); pluviométricos e fluviométricos. A organização e o tratamento das diversas informações foram realizados com o auxílio de Sistema de Informação Geográfica (Spring 3.6), obtendo também um banco de dados limnológicos multitemporal. As ferramentas de geotecnologias foram aplicadas na caracterização dos fatores fisiográficos, no mapeamento do uso e cobertura da terra e do potencial poluidor das sub-bacias da bacia de drenagem, sendo os resultados analisados em conjunto com os dados referentes à qualidade da água do reservatório de Barra Bonita, de forma espacial e temporal (período de 1990 a 2002). Os resultados mostraram que ocorreu um incremento significativo do nível trófico da água, de forma diferenciada para cada compartimento do reservatório em questão, durante o período avaliado. A análise integrada dos sistemas aquático e terrestre, no espaço e no tempo, permitiu concluir que a degradação da qualidade da água se deve às alterações no uso e cobertura da terra e ao aumento populacional, traduzidos em fontes difusas e pontuais de poluição. Desta forma, fica evidenciada a importância da gestão de recursos hídricos na bacia estudada e espera-se que as informações obtidas e metodologias aplicadas neste trabalho possam subsidiar ações direcionadas a este propósito. / Nowadays, water quality deterioration is a common problem related to pollution inputs from point and nonpoint sources. The situation becomes worse in the reservoirs due to multiple water uses. Barra Bonita reservoir was selected for this study because it is located in an overpopulated area of the São Paulo State, southern Brazil. Moreover, it is a region of many industries and intensive agriculture. The data used to carry out the study included physical features (pedology, geomorphology, geology, drainage, precipitation and flow) provided by several public agencies in different scales and formats; land use and land cover maps derived from interpretation of Landsat TM ETM satellite images; limnologycal variable measurements derived from field work and literature spanning the period comprised between 1979 and 2002 and census data (population density and agriculture production). The first step in the study was to build a limnologycal database to compile and organize all the information available regarding water quality in the Barra Bonita reservoir. A second major step was to implement and feed the Médio Tietê Basin Geographic Information System with the selected pollution forcing variables. After that, the geotechnogies were applied to diagnose physical features, to map land use and land cover and to model potential pollution instances for each sub-basin. This data was analyzed in the space and time considering the water quality degradation from 1990 to 2002. The results showed an increase in the eutrophication level of Barra Bonita reservoir in the period considered in response to changes in land use and land cover patterns. The water-land integrated analysis allowed to verify that the water degradation is related to the land use and land cover changes and population increase (diffuse and point contribution). Therefore, the water resources management is paramount in order to improve the water quality in this basin. It is expected that information and methodologies applied in this research can be used to support this sustainable process.

Feature extraction is the main driving force behind the advancement of the image processing techniques infields suchas image quality assessment, objectdetection, and object recognition. In this work, we perform a comprehensive and in-depth study on feature extraction for the following applications: image macro-uniformity assessment, 2.5D printing quality assessment, streak defect detection, and pedestrian detection. Firstly, a set of multi-scale wavelet-based features is proposed, and a quality predictor is trained to predict the perceived macro-uniformity. Secondly, the 2.5D printing quality is characterized by a set of merits that focus on the surface structure.Thirdly, a set of features is proposed to describe the streaks, based on which two detectors are developed: the first one uses Support Vector Machine (SVM) to train a binary classifier to detect the streak; the second one adopts Hidden Markov Model (HMM) to incorporates the row dependency information within a single streak. Finally, a novel set of pixel-difference features is proposed to develop a computationally efficient feature extraction method for pedestrian detection.

L’art des Turkmènes qarâ qoyunlu (782-873 h./1380-1468) et âq qoyunlu (780-914 h./1378-1508) alongtemps été dédaigné dans l’histoire de l’art du monde iranien. Pourquoi assimiler systématiquement cesoeuvres aux productions timourides ? Ne pourrait-on donc parler d’un « art turkmène » ? À partir d’un corpusde cinquante ensembles décoratifs en céramique, l’objectif de cette étude est de retracer une histoire desproductions turkmènes, de leurs formes, de leurs techniques, et d’en comprendre les articulations artistiques.L’ensemble des monuments qarâ qoyunlu et âq qoyunlu connus est dans un premier temps exploré,regroupés selon une présentation régionale : un parti pris permettant de souligner la force de traditionsartistiques locales autour de centres tels que Tabriz, E½fahân, Yazd ou encore Kâshân. Cette approche liminaireest suivie d’une analyse technique et ornementale des céramiques architecturales. Le dernier axe de cetterecherche s’intéresse à l’organisation de la production. L’enjeu est d’abord de comprendre les confluencesartistiques complexes s’exerçant alors en Iran, avant de s’arrêter sur les acteurs de cette production : lesmécènes et les artisans.Cette étude cherche à apporter un regard nouveau sur certains pans de la céramique architecturaleislamique. La « charnière turcomane » constitue en effet un jalon important de l’histoire de l’art : un point decontact avec les territoires timouride, mamlouk et ottoman frontaliers, et une passerelle vers les Empiresmodernes du siècle suivant. / Qarâ Qoyunlu (782-873 h./1380-1468) and Âq Qoyunlu (782-873 h./1380-1468) Turkman’s art haslong been disregarded in the art history of the Iranian world. This research brings the nature of Turkman art intoquestion, through the example of tiling art. Based on a group of fifty ornamental ceramic groups, this studyattempts to trace the history of Turkman decorative features, their designs, their techniques, and to understandtheir artistic connections.This research first proposes to explore every known qarâ qoyunlu and âq qoyunlu monument, sharedout among regional groups so that the strength of the regional traditions could be underlined around suchartistic centers as Tabriz, E½fahân, Yazd, or Kâshân. This part is followed by a technical and ornamental studyof Turkman tiles. The last component of that research is the organization of the production. It aims to givestatements about artistic confluences in Iran, and then to present actors of tiling production : the patrons andcraftsmen.This study intends to bring a new eye on some groups of Islamic tiles. The Turkman transition makesan important step in history of art: a meeting point between Timurid, Mamluk and Ottoman’s borderlineterritories, and a bridge to Modern Empires of the next century.

Statistical parametric speech synthesis (SPSS) has seen improvements over recent years, especially in terms of intelligibility. Synthetic speech is often clear and understandable, but it can also be bland and monotonous. Proper generation of natural speech prosody is still a largely unsolved problem. This is relevant especially in the context of expressive audiobook speech synthesis, where speech is expected to be fluid and captivating. In general, prosody can be seen as a layer that is superimposed on the segmental (phone) sequence. Listeners can perceive the same melody or rhythm in different utterances, and the same segmental sequence can be uttered with a different prosodic layer to convey a different message. For this reason, prosody is commonly accepted to be inherently suprasegmental. It is governed by longer units within the utterance (e.g. syllables, words, phrases) and beyond the utterance (e.g. discourse). However, common techniques for the modeling of speech prosody - and speech in general - operate mainly on very short intervals, either at the state or frame level, in both hidden Markov model (HMM) and deep neural network (DNN) based speech synthesis. This thesis presents contributions supporting the claim that stronger representations of suprasegmental variation are essential for the natural generation of fundamental frequency for statistical parametric speech synthesis. We conceptualize the problem by dividing it into three sub-problems: (1) representations of acoustic signals, (2) representations of linguistic contexts, and (3) the mapping of one representation to another. The contributions of this thesis provide novel methods and insights relating to these three sub-problems. In terms of sub-problem 1, we propose a multi-level representation of f0 using the continuous wavelet transform and the discrete cosine transform, as well as a wavelet-based decomposition strategy that is linguistically and perceptually motivated. In terms of sub-problem 2, we investigate additional linguistic features such as text-derived word embeddings and syllable bag-of-phones and we propose a novel method for learning word vector representations based on acoustic counts. Finally, considering sub-problem 3, insights are given regarding hierarchical models such as parallel and cascaded deep neural networks.

Nas últimas décadas, as texturas dinâmicas ou texturas temporais, que são texturas com movimento, tornaram-se objetos de intenso interesse por parte de pesquisadores das áreas de processamento digital de imagens e visão computacional. Várias técnicas vêm sendo desenvolvidas, ou aperfeiçoadas, para a extração de características baseada em texturas dinâmicas. Essas técnicas, em vários casos, são a combinação de duas ou mais metodologias pré-existentes que visam apenas a extração de características e não a melhora da qualidade das características extraídas. Além disso, para os casos em que as características são \"pobres\" em qualidade, o resultado final do processamento poderá apresentar queda de desempenho. Assim, este trabalho propõe descritores que extraiam características dinâmicas de sequências de vídeos e realize a fusão de informações buscando aumentar o desempenho geral na segmentação e/ou reconhecimento de texturas ou cenas em movimento. Os resultados obtidos utilizando-se duas bases de vídeos demonstram que os descritores propostos chamados de D-LMP e D-SLMP foram superiores ao descritor da literatura comparado e denominado de LBP-TOP. Além de apresentarem taxas globais de acurácia, precisão e sensibilidade superiores, os descritores propostos extraem características em um tempo inferior ao descritor LBP-TOP, o que os tornam mais práticos para a maioria das aplicações. A fusão de dados oriundos de regiões com diferentes características dinâmicas aumentou o desempenho dos descritores, demonstrando assim, que a técnica pode ser aplicada não somente para a classificação de texturas dinâmicas em sí, mas também para a classificação de cenas gerais em vídeos. / In the last decades, the dynamic textures or temporal textures, which are textures with movement, have become objects of intense interest on the part of researchers of the areas of digital image processing and computer vision. Several techniques have been developed, or perfected, for feature extraction based on dynamic textures. These techniques, in several cases, are the combination of two or more pre-existing methodologies that aim only the feature extraction and not the improvement of the quality of the extracted features. Moreover, in cases that the features are \"poor\" in quality, the final result of processing may present low performance. Thus, this work proposes descriptors that extract dynamic features of video sequences and perform the fusion of information seeking to increase the overall performance in the segmentation and/or recognition of textures or moving scenes. The results obtained using two video bases show that the proposed descriptors called D-LMP and D-SLMP were superior to the descriptor of the literature compared and denominated of LBP-TOP. In addition to presenting higher overall accuracy, precision and sensitivity rates, the proposed descriptors extract features at a shorter time than the LBP-TOP descriptor, which makes them more practical for most applications. The fusion of data from regions with different dynamic characteristics increased the performance of the descriptors, thus demonstrating that the technique can be applied not only to the classification of dynamic textures, but also to the classification of general scenes in videos.

O presente estudo investiga a aquisição da expressão reflexiva \'ele mesmo\' por crianças adquirindo Português Brasileiro como língua materna, considerando a influência que tal expressão sofre da alternância entre os traços semântico-pragmáticos de reflexividade [+ provável] e reflexividade [- provável] dos predicados em que se encontra. Dois experimentos foram conduzidos, um de produção e um de compreensão, para descobrirmos que, nem mesmo dentre as crianças mais velhas testadas por nós (5;7 - 6;6), o desempenho adulto com a proforma complexa foi verificado. A variação semântica dos predicados, tendo se mostrado apreendida já pelas crianças entre 3;6 - 4;6, foi descartada como agente causador dos erros cometidos pelos sujeitos, razão pela qual recorremos à explicação que responsabiliza, pelo domínio tardio da expressão, a sobrecarga computacional que as operações mentais relevantes à interpretação do \'ele mesmo\' impõem sobre a limitada memória de trabalho infantil. / The present study investigates the acquisition of the reflexive expression \'ele mesmo\' by children acquiring Brazilian Portuguese as their mother language, considering the influence that such expression suffers from the variation between the semantic-pragmatic features of [+ probable] reflexivity and [- probable] reflexivity of the predicates in which it is included. Two experiments were conducted, one of production and another of comprehension, and we discovered that, not even among the oldest children tested (between 5;7 - 6;6 years of age), adult performance with the complex proform was achieved. The younger children tested (between 3;6 - 4;6 years of age) showed mastery of the semantic variation of the predicates. Therefore, the mistakes committed by the subjects cannot be explained by a lack of knowledge of these pragmatic features. In order to account for such non-adult behavior, we resorted to an explanation that blames the computational overload that the relevant mental operations to the interpretation of \'ele mesmo\' impose over the limited working memory of the children.

Quatro linhagens capazes de crescer com acefato foram isoladas a partir de solos com históricos de aplicação de deste composto. Stenotrophomonas maltophilia, Rhodococcus sp., Staphylococcus sp. e Pandoreae sp. foram identificadas através do rDNA 16S e perfil de ácidos graxos de membrana. Rhodococcus sp. foi o isolado mais eficiente na degradação de acefato, removendo 99.24% deste composto em meio de cultura com acefato como única fonte de carbono. Quando avaliado com acefato como fonte combinada de carbono e nitrogênio, este organismo degradou 19% de acefato com formação de metamidofós (17%). Staphylococcus sp. apresentou 21% de degradação de acefato utilizando-o como fonte de carbono e nitrogênio, mas não manteve o crescimento com este composto como fonte de carbono. Stenotrophomonas maltophilia e Pandoreae sp. não mantiveram crescimento com acefato como única fonte de carbono isoladamente. Estas linhagens apresentaram crescimento em acefato como fonte de nitrogênio e enxofre, porém, as análises de GC/MS demonstraram que não houve degradação nestas condições. / Four strains of microorganisms capable of growth on acephate were isolated from soil samples with a history of acephate application. Stenotrophomonas maltophilia, Rhodococcus sp., Staphylococcus sp. and Pandoreae sp. were identified based on 16S rRNA gene sequencing and fatty acid profiling. Rhodococcus sp. was the most efficient acephate degrader of the isolates, it removed 99.24% of acephate from defined growth media when the compound was provided as sole carbon source. When provided as a combined carbon and nitrogen source, the organisms degraded 19% of acephate with formation of methamidophos (17%). Staphylococcus sp. degraded 21% of acephate when provided as sole nitrogen and carbon source but did not grow on the compound as a sole source of carbon. Pandoreae sp. and Stenotrophomonas maltophilia failed to grow on acephate as sole source of carbon in defined medium. These strains grew in media where the pesticide was provided as a combined nitrogen and carbon source, but no acephate biodegradation could be demonstrated in these instances.