Nowadays, the Norwegian Public Roads Administration plans to replace the ferry connections along the western coast of Norway with fixed road connections through the coastal highway E39 project. Due to the necessity of bridging long distances with considerable depths, non-conventional engineering solutions are being developed. To date, the first fjord planned to be cross is Bjørnafjord, located just in the middle of Trondheim and Kristiansand. In order to bridge the fjord distance, three innovative solutions have been developed: a submerged floating tube bridge, a floating pontoon bridge, and finally, a multi-span suspension bridge. It is known that these structures are affected by periodic environmental forces which may cause significant fatigue problems in some structural components, disturbing their service behaviour. The aim of this master thesis is to deeply analyse the cumulative fatigue damage produced in the mooring lines of a proposed solution for crossing Bjørnafjorden as well as to investigate the influence of the implementation of different mooring line scenarios on the damage. A 3D finite element model of the chosen alternative was conducted and verified. The model provided the stress variations in each mooring line which were processed through Excel software using macros performed as well as Matlab scripts in order to obtain the results. Then, a deep fatigue damage assessment was carried out discussing how different tension-history time period calculations affect to the damage, as well as which fatigue evaluation methodology is adequate for the chosen solution. Moreover, the fatigue damage produced on the initial mooring line system configuration was analysed. It was investigated the repercussions of each direction and the contribution of each environmental load on the total damage. Last but not least, it was researched the influence of different water depths and line configurations on the fatigue damage. The results show that the fatigue damage obtained for the adopted solution remains far from the failure threshold with the worst damage produced in the mooring line chain components. The previous results have been computed employing a 1-hour analysis duration through the rainflow counting method. Also, the worst fatigue damage is produced by environmental states propagating in a westerly direction, corresponding to the maximum fetch. Regarding the load contribution in the fatigue damage, it is proved in this case that the current loads stabilise the mooring lines, reducing the fatigue damage. Finally, investigating the effect of various mooring line scenarios on the damage it has been proved that the fatigue damage is increased as the water depth is decreased and as the lateral mooring lines are separated from the initial configuration.

[pt] A modelagem numérica tem-se mostrado uma ferramenta valiosa na avaliação e previsão do comportamento estrutural a partir de informações experimentais. Neste trabalho faz-se um estudo de certos problemas de identificação de sistemas estruturais. O principal objetivo reside na avaliação e interpretação computacional de parâmetros estruturais com base em resultados de ensaios mecânicos não- destrutivos. Foram desenvolvidos e codificados algoritmos, em diferentes formatos computacionais, para o tratamento deste problema. O desenvolvimento é aplicável a qualquer tipo de estrutura modelada por elementos finitos. No trabalho são exemplificadas estruturas em forma de conjuntos de massas e molas, vigas em flexão no plano, vigas em torção e flexão, arcos, pórticos e treliças. Analisa-se a determinação de termos de rigidez e massa a partir de informações globais e locais sobre deslocamentos, deformações e freqüências de vibração. São apresentados estudos que indicam a possibilidade de determinar, com as técnicas numéricas aqui desenvolvidas, a localização e magnitude dos danos ocorridos nas estruturas, a partir da resposta estática ou dinâmica da estrutura, mesmo com reduzidas informações sobre a resposta da estrutura não danificada. Isto foi aqui estendido à determinação de tensões e carregamentos,através da consideração de efeitos de não-linearidade geométrica. Apresentam-se também critérios para projeto de experimentos, de forma a maximizar a sensibilidade aos parâmetros buscados e minimizar sua correlação. Ilustra-se através de vários exemplos o desenvolvimento teórico e numérico aqui apresentado. / [en] Numerical modeling is a valuable tool in the evaluation and prediction of structural behavior from given experimental data. This work presents a study of some problems of identification of structural systems. The main objective is the computer evaluation and interpretation of structural parameters from the results of non-destructive mechanical tests.Algorithms were developed and coded, in different computational formats, to treat this problem.The development is applicable to any type of structure modeled by finite elements. Examples presented in this work are sets of masses and springs, beams in pane bending, beams in bending and torsion, arches, frames and trusses. The determination of stiffness and mass terms, from local and global information on displacements, strains and frequencies, is discussed. The studies indicate that, using the numerical techniques developed herein, it is possible to determine a location and magnitude of damages occurred in structures, from the static or dynamic response data, resorting to limited information on the response of the undamaged structure. Such techniques were extended to the evaluation of stresses and loadings, by taking into account effects of geometric nonlinearity. Criteria for design of experiments are presented, which allow for maximum sensitivity and minimum correlation of the structural parameters under concern.Several examples are presented to illustrate the theoretical and numerical techniques developed herein.

Made available in DSpace on 2014-12-17T14:57:53Z (GMT). No. of bitstreams: 1 JoaoMS.pdf: 1515806 bytes, checksum: fd1b665e0003e03823a56f9ce0f2eaaf (MD5) Previous issue date: 2009-09-11 / Petr?leo Brasileiro SA - PETROBRAS / The modern technology of materials and structural integrity of pipelines requests the use of inspection tools named inspection pigs to detect, localize and measure the length, width and depth dimensions of the thickness losses of walls of buried and underwater pipelines in service. These tools run them internally, performing and recording measurements, with performance that varies according to the pig s technology. It has been developed recently an instrumented pig technology, called feller pig. This work aims to indicate factors that influence the feller pig technology performance in the detection and in the accuracy of measurement of the length, width and depth dimensions of the thickness losses on the internal surface of an oil pipeline wall under normal conditions of oil pipe inspection with pig. In this work, is made a collection of factors and an analyses of the technology based on the available literature, as well as an experiment to observe the technology and the factors operating. In the experiment, a feeler pig is used in a pipeline built in carbon steel and in operation that flows petroleum, in witch are observed areas with internal thickness losses occurred naturally. Some of these areas and their dimensions taken by automated ultra-sound scanner are compared with the ones indicated by the feller pig. Based on the data collection, on the analysis and on the experiment, the influence of factors object of this research is discussed. It is concluded that, among these, there are factors related to pipe fabrication tolerances, to wear of pig components, to internal adhesive wear of pipeline, to other pipeline damages and to technology characteristics. Finally, actions are suggested to know better, improve and define the applicability of this technology / A moderna tecnologia de materiais e de integridade estrutural de dutos pressup?e o uso de ferramentas de inspe??o denominadas pigs de inspe??o para detectar, localizar e medir comprimento, largura e profundidade das perdas de espessura de parede de dutos em servi?o enterrados ou submersos. Essas ferramentas os percorrem internamente, realizando e registrando medi??es, com desempenho que varia de acordo com a tecnologia do pig. Recentemente foi desenvolvida uma tecnologia de pig de inspe??o, chamada de pig de perfilagem. O objetivo desta pesquisa ? indicar fatores que influenciam o desempenho da tecnologia de pig de perfilagem na detec??o e na exatid?o de medi??o de comprimento, largura e profundidade de perdas de espessura na superf?cie interna da parede de um oleoduto, em condi??es normais de inspe??o de oleodutos com pig. Neste trabalho, faz-se um levantamento de tais fatores e uma an?lise desta tecnologia, baseados na literatura dispon?vel, bem como um experimento para observar a tecnologia e fatores operando. No experimento, utiliza-se um pig de perfilagem em um duto constru?do em a?o carbono e em opera??o escoando petr?leo, no qual se constata ?reas com perdas de espessura internas ocorridas naturalmente. Algumas destas ?reas e suas dimens?es, medidas por varredura de ultra-som automatizado, s?o comparadas com as indicadas pelo pig de perfilagem. Com base no levantamento, na an?lise e no experimento, discute-se a influ?ncia de fatores objetos desta pesquisa. Conclui-se que, entre estes, ocorrem fatores relacionados ? toler?ncia de fabrica??o dos tubos, ao desgaste de componentes do pig, ao desgaste adesivo interno ao duto, a outros danos no duto e a caracter?sticas da tecnologia. Por fim, s?o sugeridas a??es para conhecer melhor, aprimorar e delimitar a aplicabilidade desta tecnologia

Ce travail de thèse est focalisé sur l’établissement d’un modèle de coupe orthogonale simulant l’usinage de l’alliage de Titane (Ti–6Al–4V). L’objectif principal est de contribuer à des approches et des méthodologies numériques permettant la modélisation de la formation du copeau de ce type d’alliage et l’amélioration de la compréhension des phénomènes multi-physiques induits par l’enlèvement de matière et qui découle de l’interaction outil-matière. Pour se faire, ce travail de recherche commence par la présentation d’une méthodologie macroscopique et robuste permettant d’établir un modèle de coupe en adoptant la loi de comportement phénoménologique de Johnson et Cook. L’une des améliorations apportés dans ce modèle consiste à adopter le critère de l’énergie de rupture pour contrôler l’évolution de l’endommagement de la matière et pourvoir par conséquent simuler pertinemment la genèse du copeau. Cela a permis de capturer la localisation de la déformation maximale par cisaillement et par la suite de dimensionner les formes caractéristiques liées au phénomène de segmentation du copeau selon les conditions de coupe adoptées. Le modèle de frottement à l’interface outil-matière découpée a également été affiné en intégrant une contrainte de cisaillement limite à l’interface copeau-outil. L’effet de la température sur cette contrainte limite a également été considéré et implémenté dans une routine utilisateur. Cela a permis de modéliser la formation du copeau même sous des conditions de frottement agressives et avec des rayons d’acuité d’arête de coupe élevés. L’étude paramétrique ainsi effectuée montre une très bonne concordance avec les résultats expérimentaux. Dans un second temps un modèle qui peut-être qualifié de multi-echelles a été élaboré. Ce modèle de coupe tient compte de la microstructure du matériau à découper. Il est basé sur la théorie de la plasticité cristalline pour simuler la coupe orthogonale de l’alliage de Titane Ti–6Al–4V. Dans cette approche, les grains de la matière traitée sont présentés explicitement dans le maillage avec des formes hexagonales caractérisées par différentes orientations et systèmes de glissement. Dans ce cas d’étude le processus de dégradation de la matière sous différentes sollicitations thermomécaniques est le résultat de la cœxistence de deux types d’endommagement. En effet, il est supposé une rupture de la matière au niveau des grains qualifiée de dommage intra-granulaire et un dommage éventuel, aux interfaces des grains, qualifié d’inter-granulaire. Ce dernier est simulé par des éléments cohésifs d’épaisseurs nulles traduisant la possibilité de la décohésion des gains de la matière. Ce modèle est validé par des résultats de compression provenant de la littérature. / The objective of this study is to model material removal with cutting tool in the case of the machining of Titanium alloy (Ti–6Al–4V) and to bring a multi-physic comprehension of chip formation and the tool/workpiece interaction by adopting finite element approaches and methodologies. For that, the present contribution begins by a macroscopic modeling of the orthogonal cutting process. The cut material behavior considered is supposed based on JC law. Moreover, in order to simulate properly the chip genesis, the material fracture energy concept is adopted for controlling the material damage evolution. This allows capturing the shear strain localization and consequently the chip segmentation for a given set of cutting parameters. The frictional contact model considers the influence of temperature on the limiting shear stress at the tool/chip interface. As a result, this reliable model has the capability to simulate the cutting process even with high coefficient of friction and with large cutting edge radius. The parametric study carried out by referring to this model shows a very interesting corroboration with experimental results. In a second step, the present research work presents a material microstructure-level cutting model (MML cutting model) for cutting simulation. The crystal plasticity theory is adopted for modeling the cutting of the Titanium alloy Ti–6Al–4V in orthogonal case. In this model, the grains of the studied material are explicitly considered, and their orientation angles and slip system strength anisotropy are considered as the main source of the microstructure heterogeneity in the cutting material. To simulate the material degradation process, the continuum intra-granular damage and discrete cohesive zone inter-granular damage models are developed, wherein the zero thickness cohesive elements are implemented to simulate the bond between grain interfaces. The material model is validated by a comparison of compression tests from literature. Finally, simulation results demonstrate the possibility to capture the influence of the microstructure on the material removal in terms of chip formation. It is demonstrated that the grain orientation angle plays an important role for the chip segmentation and its periodicity during the cutting process. This certainly can affect the evolution of the cutting force. Additionally, the surface integrity is discussed based on the MML cutting model for different cutting speeds and feed rates. Indeed, a parametric study shows that the surface integrity is seriously affected by the machining parameters, and the affected region is limited within three layer grains for the present MML cutting model.

<p dir="ltr">Bridge safety has been a longstanding priority for civil engineers. Engineers and researchers devote significant effort toward establishing the ability to detect and monitor damage and to manage the performance of our bridges over their lifecycle. However, current practices, heavily reliant on manual involvement, still present challenges in efficiency and effectiveness. For instance, rapid bridge vulnerability assessment methods are being developed, but these methods normally require information about the bridge, e.g., substructure type, which is not readily available in most current bridge databases (e.g., National Bridge Inventory). Manually collecting the necessary information for each bridge is time-consuming and would influence the use of those rapid bridge vulnerability assessment methods. Similarly, routine bridge inspection, mandated every two years in the United States, requires organizing inspection photos, evaluating bridge condition, etc., which are also time-consuming tasks. A two-year inspection cycle may be overly cautious, especially in the early stages of its life when there is typically little degradation. Furthermore, when conducting an inspection there is no reference for when to use advanced inspection techniques, so visual inspection, the simplest method, is adopted for most bridges. Given these challenges in bridge asset management, the integration of machine learning to use the data from historical records of bridge performance can aid in and serve to expedite the above tasks.</p><p dir="ltr">The objective of this research is to develop machine learning-based methods that can assist humans in completing certain tasks associated with bridge asset management. Towards this objective, the following research tasks are carried out. In Task 1, a CNN-based bridge substructure identifier is developed to automatically recognize the bridge substructure type from an inspection image. A method is developed to set a rational budget for this work based on risk tolerance. In Task 2, the automated bridge inspection image organization tool (ABIRT) is developed to automate the process of organizing inspection images and generating an inspection report. In Task 3, a technique for machine-assisted bridge damage analysis using visual data is developed and validated. Additionally, a decision-making method is established to assist bridge inspectors in adopting this technique, with a focus on managing costs and minimizing risks. And in Task 4, reinforcement learning-based approach is developed to manage the bridge inspection process. Through this research, several key machine learning techniques are explored to assist bridge managers with the more tedious steps involved in the seismic vulnerability analysis and condition management tasks, allowing the engineer to dedicate more time to making decisions. This highly cross-disciplinary research is scalable and expandable to many other applications and will serve to improve the future safety and reliability of our infrastructure.</p>