Master of Science / Department of Electrical Engineering / William B. Kuhn / During the development of an educational renewable energy production platform, it was found that there were no low-cost, efficient grid-tie interfaces for a 160 W DC wind turbine. Typically, a small DC wind turbine is used in conjunction with a rechargeable battery bank or, if the wind turbine is directly interfaced with a grid-tie inverter, a regulator with a diversion-load. The use of batteries is undesirable due to their high-cost and high-maintenance characteristics. Diversion loads by nature waste power, as any excess energy that cannot be accepted by a battery or inverter is usually converted into heat through a resistive element. Initially, a 24 V DC, 160 W Air Breeze small wind turbine was directly connected to an Enphase Energy M190 grid-tie micro-inverter. The 24 V DC Air Breeze wind turbine is designed to charge a battery or bank of batteries while the M190 micro-inverter is designed to convert the DC output of a 200 W solar panel to grid-tied AC power. As expected, the power-production response time associated with the small wind turbine and the power-accepting, load-matching response time of the micro-inverter were not compatible. The rapidly changing power output of the small wind turbine conflicted with the slow response time of the micro-inverter resulting in little power production. Ultimately, the response time mismatch also produced sufficiently large voltage spikes to damage the turbine electronics. In this thesis, a solution for a low-cost, efficient grid-tie interface using no batteries and no diversion load is presented. A capacitance of eight Farads is placed in parallel with the small wind turbine and the micro inverter. The large capacitance sufficiently smoothes the potential abrupt voltage changes produced by the wind turbine, allowing the micro-inverter adequate time to adjust its load for optimal power conversion. Laboratory experiments and data from an implementation of such a parallel super capacitor wind turbine to grid-tie micro-inverter configuration are provided along with DC and AC power production monitoring circuits interfaced with a micro controller.

While knowledge is one of an organization’s greatest assets, it remains a challenge to facilitate knowledge transfer between people within an organization. Social influence has been studied in its role of facilitating information diffusion, which is necessary for knowledge transfer to occur. Among this research, tie strength, a quantifiable characteristic of a social network that determines the link between two nodes, has been measured to determine the impact of social influence on knowledge transfer and information dissemination within a social network. Current research that explores the impact of social influence on information diffusion has been conducted within public social networks due to the availability of data that can be gathered from public social online network systems, such as Facebook. With the emergence of collaboration technologies that exist in online social network tools being utilized within organizations, there is an opportunity to digitally collect information regarding information dissemination within a collaborative network. This study captured data from an online social network, specifically a unified communication tool, being used within a collaborative social network at a mid-sized South Central corporation. A content analysis of Lync messages for 1,749 connections was performed to quantitatively measure the influence of tie strength on information dissemination within a collaborative social network. The results demonstrated that tie strength had a significant impact on information dissemination using a collaborative system. Multivariate analysis of variance showed that tie strength had the largest impact on information dissemination using the instant messaging modality of a collaboration system.

Les structures en béton armé sont amenées à répondre à différentes exigences pouvant dépasser la simple résistance mécanique. Pendant le processus de fissuration, les contraintes dans le béton armé sont progressivement redistribuées entre l’acier et le béton via l’interface entre ces deux matériaux. Cette redistribution de contraintes a un impact direct sur l’état de fissuration final et doit donc être prise en compte dans la modélisation. Il existe différents modèles numériques capables de représenter les effets de la liaison acier-béton. Cependant, leur usage est, pour l’instant, incompatible avec les applications concernant les structures de grandes dimensions (difficultés de maillage, coût de calcul…). Dans ce cadre d’application, l’hypothèse de liaison parfaite entre l’acier et le béton (déplacement identique) est donc toujours utilisée. On se propose ici de développer un nouveau modèle éléments finis de liaison acier-béton qui soit à la fois représentatif des phénomènes physiques se produisant à l’interface entre les deux matériaux et compatible avec les contraintes de modélisation des structures de grandes dimensions. Ce travail de thèse se découpe en trois grandes parties : - le développement d’un modèle élément fini de liaison acier-béton adapté aux contraintes de modélisation des structures de grandes dimensions. Ce modèle numérique permet ainsi de tenir compte des interactions mécaniques entre le béton et les armatures d’acier représentées à l’aide d’éléments barres. - la caractérisation du comportement de la liaison acier-béton. Un modèle de loi d’adhérence (évolution de la contrainte d’adhérence en fonction du glissement) basé sur des observations expérimentales (campagne expérimentale de pull-out menée au cours de la thèse et données bibliographiques) est proposé. Il permet en particulier de différencier le cas d’une rupture par arrachement, d’une rupture par éclatement en tenant compte des caractéristiques matériaux et géométriques de la structure. - l’application du modèle proposé à un élément structurel (poutre). Un essai de poutre en flexion quatre points visant à caractériser l’évolution de la fissuration (évolution de l’ouverture de fissure mesurée à l’aide de la technique de corrélation d’images notamment) a ainsi été proposé. Ces résultats ont ensuite été comparés à ceux de simulations numériques tenant compte de la liaison acier-béton d’une part ou de l’hypothèse de liaison parfaite d’autre part. Les deux modélisations donnent une bonne approximation du comportement extérieur de la structure (comportement global et ouvertures de fissure des surfaces extérieures de la poutre). Le modèle de liaison acier-béton apporte cependant une meilleure caractérisation de la phase de fissuration active (apparition des fissures) et modifie plus particulièrement le comportement local de la structure à proximité directe des armatures (limitant le développement de l’endommagement du béton le long des renforts). / Reinforced concrete structures may have to fulfill functions that go beyond their simple mechanical resistance. During the cracking process, stresses are progressively transferred from steel to concrete through the steel-concrete interface. This stress transfer has a direct impact on the crack properties. Taking into account these effects seems thus essential to predict correctly the cracking of reinforced concrete structures. Different models exist to represent the steel-concrete bond behavior. However, these models are rarely compatible with large scale simulations (meshing difficulties, heavy computational cost…). To overcome these difficulties, a perfect relation between steel and concrete (same displacements) is generally considered for structural applications. In this contribution, a new finite element approach is proposed to represent the steel-concrete bond effects in a context adapted for large scale simulations. This thesis is divided in three parts: - the development of a finite element steel-concrete bond model adapted for large scale structural applications . This model takes into account mechanical interactions between concrete and steel reinforcement represented by truss elements. - the characterization of the steel-concrete bond behavior. A model for the bond stress-slip law based on experimental observations (experimental campaign on pull-out test carried out during the thesis and data of literature) is proposed. This model differentiates the case of a pull-out failure and of splitting failure and takes into account the material properties and the geometric characteristics of the structure. - an application of the proposed model on a structural element (beam). A four point bending beam is experimentally tested. This test aims to characterize the crack evolution (in particular the crack opening using the image correlation technique). Experimental results are then compared with numerical simulations taking into account the bond–slip effect between steel and concrete or considering the perfect relation hypothesis. The two simulations give a good approximation of the external behavior of the structure (global behavior and crack opening on the external face of the beam). Nevertheless, the bond model improves the cracking description during the active cracking phase (beginning of crack apparition) and influences the local behavior of the structure especially near the steel bars (avoiding the propagation of the damage of concrete along the steel reinforcement).

O modelo de bielas e tirantes apresenta como uma de suas vantagens a generalidade, ou seja, é capaz de representar, de modo aproximado, porém realista e sistemático, grande parte dos elementos de concreto estrutural da atualidade. Além disso, permite ao engenheiro fácil visualização físico-intuitiva do comportamento do concreto estrutural. Por outro lado, o modelo ainda tem um enorme potencial não aproveitado. Ainda não se tem um critério exato para determinação dos nós e das seções transversais das bielas de alguns elementos. Apesar de haver certas incertezas no caso de vigas-parede e principalmente de blocos de fundação, nas vigas esses elementos do modelo podem ser determinados sem grandes dificuldades. Possivelmente, a solução seria variar as dimensões das bielas e as posições dos nós, ambas hipotéticas, a fim de confrontar várias situações com resultados experimentais. Daí a grande importância da difusão do modelo de bielas e tirantes: um maior número de análises de modelagens e de resultados de ensaios levará a um maior domínio sobre o modelo. Este trabalho consiste na análise de vigas de concreto armado utilizando modelos de bielas e tirantes. São apresentados os conceitos que levaram à concepção do modelo, desde a Analogia Clássica da Treliça, chegando aos critérios para verificação dos elementos da treliça e às recomendações atuais de normas e pesquisadores. Com o auxílio do programa computacional CAST (2000), foram modeladas quatro vigas, sendo as três primeiras biapoiadas sem descontinuidades e a quarta com balanço e descontinuidade geométrica (abertura na alma). A primeira viga biapoiada teve algumas de suas características iniciais alteradas a fim de gerar o segundo e o terceiro modelos, procurando-se estabelecer limites de carregamentos e análises comparativas. Os resultados das modelagens permitiram comparações com os resultados das verificações realizadas segundo os critérios da NBR 6118:2003 e do MC CEB-FIP (1990) e, permitiram também, identificar os aspectos de maior dificuldade na concepção de um modelo de bielas e tirantes e os pontos críticos dos mesmos, nos quais há maior possibilidade de falha nas verificações / One of the main advantages presented by the strut-and-tie model is generality. This model is able to approximately represent, in a realistic and systematic way, the majority of today’s reinforced concrete elements. Furthermore, it allows the physics-intuitive visualization of the behavior presented by structural concrete. On the other hand, its potentials are not fully explored yet. Up to date, there is no accurate criterion for the determination of nodes and transverse sections of some strut elements. Despite presenting some uncertainties related to the determination of wall-beams (and mainly of foundation blocks), the model can determine beams without major difficulties. Probably, the key is to vary the dimensions of the struts and the positions of the nodes, both hypothetical, in order to confront various situations with experimental data. Hence, it is very important to diffuse the strut-and-tie model: the bigger the number of modeling analysis and experimental data, the better the comprehension of the model. This work aims to analyze reinforced concrete beams employing the strut-and-tie model. Here are presented the concepts that led to the development of the model, ranging from the Truss Classic Analogy to the criteria employed to verify truss elements and the recommendations from researchers and established standards (rules or norms). Four examples of beams were modeled. Three of them were double-based and the fourth presented balance and no geometric discontinuity (with an opening). Some of the characteristics of the first double-based beam were altered in order to generate the second and the third model, thus establishing loading limits and comparative analysis. The modeling allowed comparisons between the verifications performed in accordance with the NBR 6118:2003 and MC CEB-FIP criteria. It also allowed the identification of major difficulties and critical aspects related to the development of struts and ties, the ones that are most prone to failure in the verification process

Este trabalho estuda o comportamento estrutural de blocos sobre estacas em concreto armado situados em divisas de terrenos. Foi feita revisão bibliográfica abordando os assuntos de blocos de coroamento, vigas de equilíbrio, modelo de bielas e tirantes e um breve comentário sobre o programa computacional Diana (2010). O trabalho apresenta um roteiro para análise estrutural e dimensionamento de vigas de equilíbrio sobre estacas com pilar excêntrico ao apoio. É apresentado um exemplo feito por Andrade (1989), comparado com o método proposto por Burke (1979) e com análise numérica. Além deste, foram analisados casos de projetos reais utilizando o método Burke (1979) adaptado neste trabalho. Alguns desses exemplos foram modelados em regime elástico e linear no programa computacional Diana (2010). São apresentadas comparações de resultados obtidos com o método Burke (1979) e análise numérica, comprovando a eficácia desse método. Em todos os exemplos estudados o método proposto por Burke (1979) indicou bons resultados quando comparados com os encontrados em análise computacional. A dissertação apresenta soluções para projetos de bloco sobre estacas, situados em divisas de terrenos e associados a vigas de equilíbrio. / This research investigates the structural behavior of reinforced concrete blocks piles located on the limit of land. Literature review was done about blocks piles, balance beams, strut and tie model and a short comment about the Diana computer program (2010). The dissertation presents a roadmap for structural analysis and sizing of balance beams on stakes with eccentric support pillar. An example, which was studied by Andrade (1989), is compared with the same example done using the method proposed by Burke (1979) and with numerical analysis. In addition to this, are analyzed real cases of projects using the method of Burke (1979). Some of these examples were modeled in linear and elastic characteristics using Diana program computer (2010). Here, it is presented comparisons between the method proposed by Burke (1979) and numerical analysis. It is presented comparisons between the method Burke (1979) and numerical analysis, proving the effectiveness of this method. In all examples studied the method proposed by Burke (1979) provided with good accuracy the results for computer analysis. The conclusion presents suggestions for projects of block piles located in limit of land associated with balance beams.

Este trabalho estuda o comportamento de blocos rígidos de concreto armado sobre uma, duas, três, quatro e cinco estacas, submetidos à ação de força centrada. Com o objetivo de contribuir para critérios de projeto, utilizaram-se resultados obtidos por meio de modelos analíticos e realizou-se análise numérica utilizando-se programa baseado no método dos elementos finitos. Foi desenvolvida, ainda, uma análise comparativa entre os processos de dimensionamento adotados em projeto, na qual se verificou grande variabilidade dos resultados. Para análise numérica adotou-se comportamento do material como elástico linear e os resultados de interesse foram os fluxos de tensões em suas direções principais. Nos modelos adotados variaram-se os diâmetros de estacas e dimensões de pilar, a fim de se verificar as diferenças na formação dos campos e trajetórias de tensões. Concluiu-se que o modelo de treliça utilizado em projetos é simplificado e foram feitas algumas sugestões para a utilização de um modelo de bielas e tirantes mais refinado. Foi possível a verificação da influência da variação da geometria de estacas e de pilares no projeto de blocos sobre e a revisão dos critérios para os arranjos das armaduras principais. Para os modelos de blocos sobre cinco estacas adotados concluiu-se que o comportamento não é exatamente como considerado na prática / This work describes the behavior of rigid reinforced concrete pile-caps with one, two, three, four and five piles subjected to patches of loading on the center top surfaces of the column. Programs based on finite elements method were used to obtain numerical results and results obtained by means of analytic models were used. A comparative analysis was developed among the processes adopted in design, which large variability of the results was verified. An elastic linear behavior of the material was adopted for numerical analysis and interest results were stress paths along main directions. Piles diameters and column dimensions of the models were been varied, in order to verify the differences in the formation of the regions and trajectories of stress. It is ended that used the truss model is simplified. Some suggestions were made for the use of a more refined strut-and-tie model than the one used in design and it was possible the verification of the variation geometry of piles and columns influence. It was still made a revision of the reinforcement layouts. The conclusion for adopted five-pile-caps was that the behavior is not exactly as considered in the practice

O projeto estrutural deve considerar as incertezas provenientes das variáveis envolvidas no dimensionamento dos elementos estruturais, tais como propriedades dos materiais, dimensões dos elementos estruturais e intensidade dos carregamentos atuantes. Existe também a variabilidade inerente aos métodos de análise estrutural, originária de imprecisões na idealização da distribuição das tensões e deformações dos elementos estruturais. Sendo assim, esta dissertação visa analisar a variabilidade dos desvios dos modelos de cálculo propostos por Blevot e Frémy (1967), Fusco (1994) e Santos (2013), todos baseados no método das bielas e tirantes, para blocos sobre duas, três e quatro estacas submetidos a carregamento centrado. Estes elementos estruturais possuem significativa importância, visto que atuam na transmissão dos esforços provenientes dos pilares às fundações profundas. Os modelos de cálculo foram confrontados entre si e com resultados experimentais de Blevot e Frémy (1967), Mautoni (1971), Clarke (1973), Miguel (2000) e Suzuki et al. (1998 / 1999 / 2000 / 2001) para determinação dos parâmetros estatísticos dos desvios desses modelos em relação aos ensaios. Os resultados apontaram desempenho satisfatório para os três modelos de cálculo. Em geral, todos os métodos se mostraram favoráveis à segurança nos casos avaliados para os três tipos de blocos de fundação. / Structural design must consider the uncertainties coming from the variables involved in dimensioning of the structural elements, such as material properties, dimensions of structural elements, and intensity of active loads. There is also inherent variability in structural analysis methods, originated from imprecision in idealization of the distribution of tensions and deformations of structural elements. Thus, this dissertation aims to analyze the variability of calculation models deviations proposed by Blevot and Frémy (1967), Fusco (1994) and Santos (2013), all based on strut and tie method, for two, three and four pile caps submitted to center load. These structural elements are of significant importance, since they act in transmission of loads from columns to deep foundations. Calculation models were compared with each other and with experimental results from Blevot and Frémy (1967), Mautoni (1971), Clarke (1973), Miguel (2000) and Suzuki et al. (1998 / 1999 / 2000 / 2001) to determine the statistical parameters of these model deviations in relation to the tests. Results indicated satisfactory performance for the three calculation models. In general, all methods were favorable to safety in cases evaluated for the three pile caps types.

Förster Resonance Energy Transfer (FRET)-based techniques are gaining an increasing importance in cell biology and cell-matrix adhesion studies because they allow both the detection of conformational changes of target proteins and their localisation in cells. Frequency Domain-Fluorescence Lifetime Microscopy (FD-FLIM) is currently considered one of the most reliable methods to measure FRET in live cells. However, due to its dependence on many technical prerequisites, its use is not yet widespread. The purpose of this work was to first establish FD-FLIM measurements of FRET on a new FD-FLIM microscope module. Then we aimed to apply FD-FLIM-FRET measurements to the study of conformational changes of the cell matrix-adhesion proteins vinculin and integrin and of the growth factor receptor Tie-2. In the first part of the work, published FRET probes including distance-sensors and two sets of vinculin-based probes were extensively tested with FD-FLIM, sensitised emission and ratiometric FRET. FD-FLIM was shown to be the most accurate method in approximating molecular distances between fluorophores. Moreover this study unveiled specific caveats associated with both existing vinculin FRET probes. FD-FLIM was then used to study conformational changes of the extracellular matrix receptor alphavβ3 integrin and of the angiopoietin receptor Tie-2 using specific FRET probes designed by us. While data showed that the alphav-integrin-FRET probe localised to adhesion sites, more experiments will be required to evaluate its full functionality. The Tie-2-FRET probe was fully functional and, upon ligand binding, allowed the detection of a bending movement of the extracellular domain towards the cell membrane. Finally, a combination of FRET, immunofluorescence and tension release experiments were used to show that intracellular tension is not required to maintain integrins in their activated conformation. However, intracellular tension is required to recruit other key proteins such as vinculin, talin and tensin to adhesions sites. Overall this work demonstrates the importance of FD-FLIM-FRET as a tool to investigate conformational changes of adhesion proteins and transmembrane receptors within the cell environment.

Recently, facing the fierce competition around the world, the headquarters of many MNCs are under pressure to make a decision quickly and effectively for their global marketing strategy. Under such a situation, how effectively the headquarters can obtain knowledge regarding the market situation of subsidiaries around the world is important. In this research, the concept of reverse knowledge transfer is applied to the space of the headquarters' acquisition of knowledge from subsidiaries, from the perspective of decision-making for global marketing strategies. In the extant literature, however, studies on reverse knowledge transfer analyzing the mechanisms dynamically in a multi-faceted way from the perspective of transferring marketing-related knowledge are rare. In order to capture the reverse knowledge transfer mechanisms in a MNC precisely and dynamically, three case studies of the headquarters of an American MNC and its Japanese subsidiary were conducted. Through analysis of the case studies, two new factors were found; one is 'bypass effect', and the other is 'heeding capacity'. By bringing in those new factors, a new dynamic model for reverse knowledge transfer is proposed.

Submitted by Joana Azevedo (joanad@id.uff.br) on 2017-07-03T17:09:22Z No. of bitstreams: 1 Dissert Katia Lucia da C A Ugulino.pdf: 5832202 bytes, checksum: 80dc87973043e14d4b57ed7af931bfba (MD5) / Approved for entry into archive by Biblioteca da Escola de Engenharia (bee@ndc.uff.br) on 2017-07-06T15:11:02Z (GMT) No. of bitstreams: 1 Dissert Katia Lucia da C A Ugulino.pdf: 5832202 bytes, checksum: 80dc87973043e14d4b57ed7af931bfba (MD5) / Made available in DSpace on 2017-07-06T15:11:02Z (GMT). No. of bitstreams: 1 Dissert Katia Lucia da C A Ugulino.pdf: 5832202 bytes, checksum: 80dc87973043e14d4b57ed7af931bfba (MD5) Previous issue date: 2015-12-03 / Durante o ciclo de vida de uma instalação de produção de petróleo, estão presentes riscos relacionados à segurança operacional. Na fase de operação, devido à presença de substâncias perigosas (inflamáveis, tóxicas e ou asfixiantes) submetidas a severas condições de processo, os cenários de risco são mais críticos do que nas demais fases. Nesta etapa do ciclo de vida, a gestão inadequada dos riscos de processos favorece a ocorrência de acidentes com alto potencial de danos: elevado número de fatalidades, custos com indenizações, investimentos para a recuperação de áreas contaminadas, pagamentos de multas a órgãos reguladores, perdas de continuidade operacional e consequentemente elevação do custo de produção. Estes eventos, de forma individual ou coletiva, impactam fortemente a sustentabilidade do negócio. Sendo assim, a identificação, análise e gestão destes riscos representam um grande desafio para os gestores do setor de petroquímico. Neste contexto uma das maiores dificuldades é gerenciar os riscos de processo de forma dinâmica. A abordagem proposta neste trabalho baseia-se em um sistema dinâmico de suporte a decisão que adota conceitos da Metodologia Bow Tie e do Método de Análise Hierárquica de Processo (AHP). O modelo proposto foi aplicado na forma de estudo de caso para uma única plataforma e os resultados deste piloto serão analisados a fim de ajustar os limites para alcançar o melhor desempenho em segurança de processo das instalações. / There are many process risks during the lifecycle of an oil production facility. In the operating phase due to the presence of hazardous substances (flammable, toxic and / or asphyxiating) subjected to severe process conditions, the risk scenery are more critical than in other phases. On this phase of lifecycle the inadequate process risk management favors the occurrence of accidents with high potential for damage: high number of fatalities, severance costs, investment for the rehabilitation of contaminated sites, payment of fines to regulatory agent, loss of operational continuity and consequently increase the cost of production. These events, individually or collectively, have a strong impact on business´s sustainability. Thus, the identification, analysis and process risk management in oil and gas production represents a big challenge for the managers of the petrochemical industry. In this context, one of the biggest difficulties is to manage process risks dynamically. The approach proposed in this work based on a dynamic decision support system that adopts concepts of the Bow Tie Methodology and the Hierarchical Process Analysis Method (AHP). The proposed model has been applied in a case study in a single platform and the results of this pilot will be analyze to adjust the limits to achieve the best performance of the facility process safety.