No description available.

The fully softened shear strength was defined by Skempton (1970) as the peak drained shear strength of a clay in a normally consolidated state. All the experience available on the applicability of the fully softened shear strength for slopes is based on back-analyses. Back-analyses of first-time failures in cuts in stiff-fissured clays and embankments constructed of fat clays have shown that, over a long period of time, the shear strength gets reduced from what is measured in the laboratory using undisturbed samples to the fully softened shear strength. These back-analyses require knowledge or assumption of pore pressures in the slope, which will have a significant influence on the shear strength obtained. Karl Terzaghi, in 1936, was the first person that qualitatively explained the behavior of cut slopes in stiff-fissured clays. According to Terzaghi (1936), a softening process is initiated by the water percolating into the fissures causing swelling and decreasing the overall shear strength of the clay mass. Investigations presented later by Skempton and his colleagues showed that the controlling shear strength for cuts in stiff-fissured clays was equal to the fully softened shear strength and recommended this shear strength to be used for design (Skempton 1970; Chandler and Skempton 1974; Chandler 1974; Skempton 1977). Skempton (1977) concluded that displacements caused by progressive failure decrease the shear strength of stiff clays toward the fully softened shear strength. At first, it was believed that only stiff-fissured clays were subjected to softening and that intact clays should be designed using the peak shear strength measured using undisturbed samples (Skempton and Brown 1961; Skempton 1964, 1970). Recent publications have showed that the likelihood of a clay experiencing softening is more dependent on the plasticity of the clay rather than the fissures (Bjerrum 1967; Chandler 1984a; Mesri and Abdel-Ghaffar 1993). Fat clays, when compared to lean clays, tend to be more brittle. This means that fat clays have a more pronounced decrease in shear strength after the peak shear strength is achieved and for this reason are more susceptible to progressive failure. First-time failures in stiff clays usually occur a long period of time after construction. For this reason, steady state seepage was used in the back-analyses of the case histories presented by Skempton and his colleagues. They found that a pore pressure ratio of 0.3 was applicable to first-time failures in cuts in stiff-fissured clays (James 1970; Vaughan and Walbancke 1973; Chandler 1974; Skempton 1977). Investigations presented by Professor Steve Wright and his colleagues of the University of Texas at Austin showed, based on back-analyses, that the fully softened shear strength is also the controlling shear strength of compacted embankments constructed of highly plastic clays (Green and Wright 1986; Kayyal and Wright 1991; Wright 2005; Wright et al. 2007). Steve Wright and his colleagues concluded that weathering, expressed in cycles of wetting and drying, was the main mechanism decreasing the shear strength of compacted clay embankments toward the fully softened shear strength. Failures in this type of projects were found to be shallow (less than 10 ft deep) and to occur numerous years after construction (USACE 1983; Stauffer and Wright 1984; Kayyal and Wright 1991; Wright et al. 2007). A pore pressure ratio ranging from 0.4 to 0.6 was found to be applicable for the case histories analyzed by Wright and his colleagues. Day and Axten (1989) recommended the use of the infinite slope method with seepage parallel to the slope face for slope stability analyses. This same recommendation was presented by Lade (2010). A seepage parallel to the slope face corresponds to a pore pressure ratio ranging from 0.4 to 0.5 for slopes with ratios of 2H:1V to 5H:1V. Failures on compacted clay embankments related to softening have been reported in Texas (Stauffer and Wright 1984; Kayyal and Wright 1991; Wright 2005; Wright et al. 2007), and Mississippi (USACE 1983). According to McCook (2012), softening of this type of structures also occur in Louisiana To perform slope stability analyses using fully softened shear strength parameter, the type of soils, type of projects, and depths where this shear strength is applicable, and the pore pressures and factor of safety to be used in design should be determined. As stated above, the fully softened shear strength has been found to be the controlling shear strength of cuts in stiff clays and compacted embankments constructed of highly plastic clays. Steady state seepage conditions should be used to design cuts in stiff clays, and a pore pressure ratio ranging from 0.4 to 0.6 or a phreatic surface at the surface of the slope should be used to design compacted embankments made of fat clays. In cuts in stiff clays, both shallow and deep failures related to fully softened shear strength have been observed. For this type of project, the recommended methodology for design is to assign a curved fully softened failure envelope to the whole slope, search for the critical failure surface, and obtain the factor of safety. This approach will provide the correct factor of safety but the critical surface obtained might not be what is expected to occur in situ. Pore pressures corresponding to steady state seepage should be used for design. It should be emphasized that the recommendation to use fully softened shear strength in first-time failures in stiff clays is based on the back-analyses of case histories. Research is required to better understand progressive failure and its influence on the shear strength mobilized in situ. In compacted embankments constructed of fat clays, only shallow failures related to fully softened shear strength have been observed. For this type of projects, the recommended methodology for design is to assign a curved fully softened failure envelope to the whole embankment, search for the critical failure surface, and obtain the factor of safety. If for any reason deep failures are to be considered in designing compacted embankments constructed of fat clays, based on the fact that failures in this type of projects are usually shallow, the first 10 ft below the surface of the slope should be assumed to have a shear strength equal to the fully softened shear strength. Pore pressures should be calculated based on a water table coincident with the slope face. The fully softened shear strength should not be used in the foundation soil. If any softening occurred in the foundation soil, this should be reflected in the shear strength measured using undisturbed samples. Softening of the foundation soil is not expected to occur after the embankment is constructed. The consequences of shallow and a deep failures are usually not the same. For this reason, is reasonable that the same factor of safety should not be required for both cases. A shallow failure may be considered by some agencies solely as a maintenance issue. The factor of safety should be based on the uncertainties in the parameters being used for design and the consequences of failure of the structure (Duncan and Wright 2005). The parameters that have more impact on the factor of safety obtained for slope stability are shear strength and pore pressures. The fully softened shear strength is the lowest shear strength expected to be mobilized in first-time slides. This shear strength, coupled with a conservative assumption of pore pressure gives a low uncertainty in the parameters that have the most influence in the factor of safety. For shallow failures, the consequences of failure are very low. For this reason, if the fully softened shear strength is used, coupled with a water table corresponding to the worst case scenario possible, a factor of safety as low as 1.25 can be used. For deep failures, the consequences of failure will vary depending on the structure. The pore pressure for this type of analyses should be based on the worst seepage condition expected throughout the life of the project. In this case, for structures with low to mid consequences of failure, a factor of safety of 1.35 can be used. For structures with a high consequence of failure, a factor of safety of 1.50 can be used. These factors of safety are based on the recommendations presented by Duncan and Wright (2005) for factors of safety based on uncertainties in the parameters and consequences of failures. The fully softened shear strength should be measured using normally consolidated remolded specimens as recommended by Skempton (1977). Soil samples should be hydrated for two days using distilled or site-specific water. The soil sample should then be washed or pushed through a No. 40 (425 µm) sieve. To achieve the desired water content, the soil sample cab be air-dried or more water should be added. Water contents equal to or higher than the liquid limit should be used to prepare test specimens for fully softened shear strength measurements. The direct shear device is recommended for fully softened shear strength measurements. The Bromhead ring shear device does not provide accurate values of fully softened shear strength. The triaxial device requires more time and effort to measure the fully softened shear strength and provides about the same fully softened shear strength as the direct shear device. The fully softened shear strength failure envelope can be estimated using the correlation presented in Figure 6.59 for the parameters required for Equation 4.1. This correlation is only intended to be used in preliminary design or if better information is not available. Laboratory determination of fully softened shear strength is always recommended for final designs. If this is not possible, the confidence limits presented in Figure 6.59 should be used to determine the fully softened shear strength parameters. / Ph. D.

The problem of determining the probability of default or credit risk for companies is crucial when providing financial services. This problem is often modeled based on snapshot data that does not take the time dimension into account. Instead, we approach the problem with enterprise resource planning data in time series. With the added complexity the time series introduce, we pose that deep learning models could be suitable for the task. A comparison of a fully convolutional network and a transformer encoder was made to the current state-of-the-art model for the probability of default problem, XGBoost. The comparison showed that XGBoost generalized very well to the time series domain, even well enough to beat the deep learning models across all evaluation metrics. Furthermore, time series data with monthly, quarterly and yearly timestamps over three years was tested. Also, public features that could be extracted from quarterly and annual financial reports were compared with internal enterprise resource planning data. We found that the introduction of time series to the problem improves the performance and that models based on internal data outperform the ones based on public data. To be more precise, we argue that the dataset being based on small to medium-sized companies lessens the impact of highly granular data, and makes the selection of what features to include more prominent. This is something XGBoost takes advantage of in a very efficient way, especially when extracting features that capture the behavior of the time series, causing it to beat the deep learning competitors even though it does not pick up on the sequential aspect of the data.

Prostate segmentation is a required step in determining the volume of a prostate, which is very important in the diagnosis and the treatment of prostate cancer. In the past, radiologists manually segment the two-dimensional cross-sectional ultrasound images. Typically, it is necessary for them to outline at least a hundred of cross-sectional images in order to get an accurate estimate of the prostate's volume. This approach is very time-consuming. To be more efficient in accomplishing this task, an automated procedure has to be developed. However, because of the quality of the ultrasound image, it is very difficult to develop a computerized method for defining boundary of an object in an ultrasound image. The goal of this thesis is to find an automated segmentation algorithm for detecting the boundary of the prostate in ultrasound images. As the first step in this endeavour, a semi-automatic segmentation method is designed. This method is only semi-automatic because it requires the user to enter four initialization points, which are the data required in defining the initial contour. The discrete dynamic contour (DDC) algorithm is then used to automatically update the contour. The DDC model is made up of a set of connected vertices. When provided with an energy field that describes the features of the ultrasound image, the model automatically adjusts the vertices of the contour to attain a maximum energy. In the proposed algorithm, Mallat's dyadic wavelet transform is used to determine the energy field. Using the dyadic wavelet transform, approximate coefficients and detailed coefficients at different scales can be generated. In particular, the two sets of detailed coefficients represent the gradient of the smoothed ultrasound image. Since the gradient modulus is high at the locations where edge features appear, it is assigned to be the energy field used to drive the DDC model. The ultimate goal of this work is to develop a fully-automatic segmentation algorithm. Since only the initialization stage requires human supervision in the proposed semi-automatic initialization algorithm, the task of developing a fully-automatic segmentation algorithm is reduced to designing a fully-automatic initialization process. Such a process is introduced in this thesis. In this work, the contours defined by the semi-automatic and the fully-automatic segmentation algorithm are compared with the boundary outlined by an expert observer. Tested using 8 sample images, the mean absolute difference between the semi-automatically defined and the manually outlined boundary is less than 2. 5 pixels, and that between the fully-automatically defined and the manually outlined boundary is less than 4 pixels. Automated segmentation tools that achieve this level of accuracy would be very useful in assisting radiologists to accomplish the task of segmenting prostate boundary much more efficiently.

<div> Recent advancements of health monitoring sensing technologies are enabling plethora of new applications in a variety of biomedical areas. In this work, we present a new sensing technology that enables a fully autonomous monitoring of urinary tract infection (UTI). UTI is the second most common infection in the human body caused by bacterial pathogens, and costs millions of dollars each year to the patients and the health care industry. UTI is easily treatable using antibiotics if identified in early stages. However, when early stage identification is missed, UTI can be a major source of serious complications such as ascending infections, loss of kidney function, bacteremia, and sepsis. Unfortunately, the limitations of existing UTI monitoring technologies such as high cost, time-intensive sample preparation, and relatively high false alarm rate prohibit reliable detection of UTI in early stages. The problem becomes more serious in certain patient groups such as infants and geriatric patients suffering from neurodegenerative diseases, who have difficulties in realizing the symptoms and communicating the symptoms with their caregivers. In addition to the aforementioned difficulties, the fact that UTI is often asymptomatic makes early stage identifications quite challenging, and the reliable monitoring and detection of UTI in early stages remain as a serious problem.</div><div> To address these issues, we propose a diaper-embedded, self-powered, and fully autonomous UTI monitoring sensor module that enables autonomous monitoring and detection of UTI in early stages with minimal effort. The sensor module consists of a paper-based colorimetric nitrite sensor, urine-activated batteries, a boost dc-dc converter, a low-power sensor interface utilizing pulse width modulation, a Bluetooth low energy module for wireless transmission, and a software performing calibration at run-time. </div><div> To further optimize the sensor module, a new fully integrated DC-DC converter with low-profile and low ripple is developed. The proposed DC-DC converter maintains an extremely low level of output voltage ripples in the face of different battery output voltages, which is crucial for realizing low-noise sensor interfaces. Since the DC-DC converter is a part of a module embedded into a diaper, it is highly desirable for the DC-DC converter to have a small physical form factor in both area and height. To address this issue, the proposed DC-DC converter adopts a new charge recycling technique that enables a fully integrated design without utilizing any off-chip components. In addition, the DC-DC converter utilizes sub-module sharing techniques – multiple modules share a voltage buffer and a recycle capacitor to reduce power consumption and save chip area. The DC-DC converter provides a regulated voltage of 1.2V and achieves a maximum efficiency of 80% with a 300ohm load resistance. The output voltage ripple is in the range of 19.6mV to 26.6mV for an input voltage ranging from 0.66 to 0.86V.<br></div>

L’installation et la maintenance des ancrages passifs scellés au rocher représentent un coût important dans le budget de fonctionnement des infrastructures de transport sujettes à l’aléa rocheux. Dans ce contexte, clarifier et optimiser les méthodes de dimensionnement actuellement employées représente un enjeu fort. Or le comportement d’un ancrage passif scellé au rocher est très complexe : il fait intervenir non seulement la rupture des matériaux constituants mais également la rupture des interfaces. Par ailleurs, la résistance de l’interface barre-scellement lors de l’arrachement d’un ancrage dépend du caractère dilatant de cette interface, liée à la géométrie de la barre (verrous) et à la plastification du matériau de scellement. Le travail de cette thèse porte sur une meilleure compréhension du comportement en traction d’un ancrage passif scellé au rocher, l’objectif étant d’améliorer la conception et le dimensionnement des ancrages sollicités axialement. Premièrement, le travail a consisté à définir le comportement de l’interface barre-scellement, Pour cela, un modèle numérique a été développé afin de reproduire des essais d’arrachement, sous différentes contraintes de confinement radial, de barres d’acier de haute adhérence (HA) scellées sur une longueur de 15 cm dans un cylindre de coulis de ciment (Moosavi et al. 2005). La géométrie réelle de l’interface acier–scellement a été considérée dans le modèle. Le comportement du coulis de scellement a été représenté par un modèle de type «concrete damage plasticity» (CDP), modèle de Lubliner (1986), implémenté dans Abaqus (2012). La représentation dans une approche continue de la plastification locale et de la fissuration, a nécessité un calage pour tenir compte des effets d’échelle. Le comportement de l’interface barre-scellement a ainsi été identifié comme cohésif dommageable avec frottement, dans une bande de cisaillement de largeur entre 2 et 3 fois la hauteur des verrous. Dans l’objectif de simuler un ancrage sur le terrain, des éléments d’interface représentant le contact acierscellement (sans les verrous) ont été développés.. La performance des éléments a été testée par la modélisation d’essais réalisés par Benmokrane et al. (1995) pour des barres de HA scellées en forage dans un bloc de béton, c’est-à-dire avec des conditions-limites radiales à rigidité constante. En parallèle, afin de tester l’influence de différents paramètres (géométrie, propriétés de matériau, accessoires de mise en oeuvre) sur le comportement de l’ancrage, une campagne insitu de 36 essais d’arrachement de barres d’acier HA scellées dans un massif calcaire résistant a été réalisée. Certaines barres équipées de fibre optique ont permis d’étudier finement la mobilisation de l’interface entre la barre d’acier et le matériau de scellement lors de la sollicitation axiale. Ces observations expérimentales, comparées aux résultats de la simulation numérique des essais, ont permis de valider le modèle numérique développé et en particulier les éléments d’interface. Par ailleurs, dans nos conditions d’essais, il n’est pas noté d’effet de la longueur de scellement ou de la résistance du coulis de scellement sur la résistance de l’ancrage : la résistance est limitée par la résistance en traction de la barre d’acier. Quant à la rigidité de l’ancrage, c’est le rapport diamètre de la barre/diamètre du forage ou un coulis plus résistant qui tend à rigidifier le comportement de l’ancrage. Pour un rocher résistant et homogène, la longueur de scellement efficace est au maximum de 65cm pour les diamètres usuels de barre. Elle tend à augmenter quand le système est plus déformable : coulis de scellement moins résistant ou, pour un diamètre de barre donné, diamètre de forage plus grand. La présence de la canule d’injection ne semble pas perturber le comportement de l’ancrage. Par contre, une attention particulière doit être portée à la mise en oeuvre de la chaussette géotextile. / Fully grouted rockbolts have been used for decades for transport infrastructure exposed to rockfall hazards. However their installation and maintenance are usually costly. Therefore, understanding and optimizing of the current design method for this type of anchorage is a major challenge. This study is not simple because of the complexity of fully grouted rockbolt behavior: its failure involves not only the failure of constituent materials (rockbolt, grout, rock), but also the failure of bolt-grout and grout-rock interfaces. Moreover, the strength of bolt-grout interface depends on the interface dilatancy, which is likely linked to the geometry of ribs and to the plastification of grout material. This thesis which includes two different parts, a numerical modeling and an experimental work, contributes hence to a better understanding of fully grouted rockbolt behavior under an axial tensile load, in order to improve the current design method. Firstly the bolt-grout interface behavior in particular the force transmission between the bolt and the grout, was studied. For this purpose, a numerical model was proposed to simulate the pull-out test of a short length of 15cm of bolt grouted in a cylinder of cement hardened paste, under constant confinement pressure (Moosavi et al., 2005). In this model, the bolt-grout interface was modeled with its real geometry, including ribs. To consider the different types of behavior and failure (tension, compression and shear) of the grout, this one was modeled by the concrete damage plasticity model (CDP model), which was developed by Lubliner et al. (1986) and implemented in Abaqus (2012). Moreover, the localization of plastic strain and the fracture of grout imply different size effects, which were also taken into consideration within the calibration of the parameters of the CDP model. The bolt-grout interface behavior was also identified to be a cohesive-damage friction interface within a 2 or 3 times the height of the ribs wide shear band. From these numerical modeling results, a constitutive model for the bolt-grout interface was developed to replace the real geometry of bolt-grout contact, in order to model in-situ anchors, whose length is much longer than that of anchors in laboratory tests. The interface model was then validated by performing the numerical simulation of the pullout tests of grouted bolt in borehole drilled in concrete block, carried out by Benmokranne et al. (1995), under constant rigidity of confinement instead of constant pressure of confinement. For the second part of this study, a series of 36 pull-out field tests of fully grouted rockbolt in a rigid limestone rock wall was conducted in order to study the influence of different parameters on the behavior of rock anchors (geometry, material characteristics, accessories). Certain of the bolts equipped with optical fiber allowed bolt strains to be measured along their length, and hence, the stress mobilization along the bolt-grout interface to be studied. The comparison between experimental and numerical results permitted us to validate the numerical simulation, in particular the proposed interface model. With our field test conditions, no influence of grouting length or grout resistance on the strength of anchor was noticed and the bolt strength was limited by the tensile strength of the steel rebar. It is the ratio of bolt diameter/borehole diameter or a stronger grout that tends to stiffen anchor. In a resistant and homogeneous rock and for usual rebar diameters, the efficient grouting length is lower than 65cm. This length increases when the system is more deformable, by using a less resistant grout or a higher grout thickness. The instalation of the cannula seem not to affect the anchor’s strength. However, a particular attention must be paid to the use of the geotextile sock.

La cryptographie à base de réseaux euclidiens a généré un vif intérêt durant les deux dernièresdécennies grâce à des propriétés intéressantes, incluant une conjecture de résistance àl’ordinateur quantique, de fortes garanties de sécurité provenant d’hypothèses de difficulté sur lepire cas et la construction de schémas de chiffrement pleinement homomorphes. Cela dit, bienqu’elle soit cruciale à bon nombre de schémas à base de réseaux euclidiens, la génération debruit gaussien reste peu étudiée et continue de limiter l’efficacité de cette cryptographie nouvelle.Cette thèse s’attelle dans un premier temps à améliorer l’efficacité des générateurs de bruitgaussien pour les signatures hache-puis-signe à base de réseaux euclidiens. Nous proposons unnouvel algorithme non-centré, avec un compromis temps-mémoire flexible, aussi rapide que savariante centrée pour des tables pré-calculées de tailles acceptables en pratique. Nousemployons également la divergence de Rényi afin de réduire la précision nécessaire à la doubleprécision standard. Notre second propos tient à construire Falcon, un nouveau schéma designature hache-puis-signe, basé sur la méthode théorique de Gentry, Peikert et Vaikuntanathanpour les signatures à base de réseaux euclidiens. Nous instancions cette méthode sur les réseauxNTRU avec un nouvel algorithme de génération de trappes. / Lattice-based cryptography has generated considerable interest in the last two decades due toattractive features, including conjectured security against quantum attacks, strong securityguarantees from worst-case hardness assumptions and constructions of fully homomorphicencryption schemes. On the other hand, even though it is a crucial part of many lattice-basedschemes, Gaussian sampling is still lagging and continues to limit the effectiveness of this newcryptography. The first goal of this thesis is to improve the efficiency of Gaussian sampling forlattice-based hash-and-sign signature schemes. We propose a non-centered algorithm, with aflexible time-memory tradeoff, as fast as its centered variant for practicable size of precomputedtables. We also use the Rényi divergence to bound the precision requirement to the standarddouble precision. Our second objective is to construct Falcon, a new hash-and-sign signaturescheme, based on the theoretical framework of Gentry, Peikert and Vaikuntanathan for latticebasedsignatures. We instantiate that framework over NTRU lattices with a new trapdoor sampler.

Esta tese se concentra em aplicar os modelos de mistura de distribuições na segmentação de imagens SAR polarimétricas multi-look. Dentro deste contexto, utilizou-se o algoritmo SEM em conjunto com os estimadores obtidos pelo método dos momentos para calcular as estimativas dos parâmetros do modelo de mistura das distribuições Wishart, Kp ou G0p. Cada uma destas distribuições possui parâmetros específicos que as diferem no ajuste dos dados com graus de homogeneidade variados. A distribuição Wishart descreve bem regiões com características mais homogêneas, como cultivo. Esta distribuição é muito utilizada na análise de dados SAR polarimétricos multi-look. As distribuições Kp e G0p possuem um parâmetro de rugosidade que as permitem descrever tanto regiões mais heterogêneas, como vegetação e áreas urbanas, quanto regiões homogêneas. Além dos modelos de mistura de uma única família de distribuições, também foi analisado o caso de um dicionário contendo as três famílias. Há comparações do método SEM proposto para os diferentes modelos com os métodos da literatura k-médias e EM utilizando imagens reais da banda L. O método SEM com a mistura de distribuições G0p forneceu os melhores resultados quando os outliers da imagem são desconsiderados. A distribuição G0p foi a mais flexível ao ajuste dos diferentes tipos de alvo. A distribuição Wishart foi robusta às diferentes inicializações. O método k-médias com a distribuição Wishart é robusto à segmentação de imagens contendo outliers, mas não é muito flexível à variabilidade das regiões heterogêneas. O modelo de mistura do dicionário de famílias melhora a log-verossimilhança do método SEM, mas apresenta resultados parecidos com os do modelo de mistura G0p. Para todos os tipos de inicialização e grupos, a distribuição G0p predominou no processo de seleção das distribuições do dicionário de famílias. / The main focus of this thesis consists of the application of mixture models in multi-look polarimetric SAR image segmentation. Within this context, the SEM algorithm, together with the method of moments, were applied in the estimation of the Wishart, Kp and G0p mixture model parameters. Each one of these distributions has specific parameters that allows fitting data with different degrees of homogeneity. The Wishart distribution is suitable for modeling homogeneous regions, like crop fields for example. This distribution is widely used in multi-look polarimetric SAR data analysis. The distributions Kp and G0p have a roughness parameter that allows them to describe both heterogeneous regions, as vegetation and urban areas, and homogeneous regions. Besides adopting mixture models of a single family of distributions, the use of a dictionary with all the three family of distributions was proposed and analyzed. Also, a comparison between the performance of the proposed SEM method, considering the different models in real L-band images and two widely known techniques described in literature (k-means and EM algorithms), are shown and discussed. The proposed SEM method, considering a G0p mixture model combined with a outlier removal stage, provided the best classication results. The G0p distribution was the most flexible for fitting the different kinds of data. The Wishart distribution was robust for different initializations. The k-means algorithm with Wishart distribution is robust for segmentation of SAR images containing outliers, but it is not so flexible to variabilities in heterogeneous regions. The mixture model considering the dictionary of distributions improves the SEM method log-likelihood, but presents similar results to those of G0p mixture model. For all types of initializations and clusters, the G0p prevailed in the distribution selection process of the dictionary of distributions.

Trabecular bone is a porous composite material which consists of a mineral phase (mainly hydroxyapatite), organic phase (mostly type I collagen) and water assembled into a complex, hierarchical structure. In biomechanical modelling, its mechanical response to loads is generally assumed to be instantaneous, i.e. it is treated as a time-independent material. It is, however, recognised that the response of trabecular bone to loads is time-dependent. Study of this time-dependent behaviour is important in several contexts such as: to understand energy dissipation ability of bone; to understand the age-related non-traumatic fractures; to predict implant loosening due to cyclic loading; to understand progressive vertebral deformity; and for pre-clinical evaluation of total joint replacement. To investigate time-dependent behaviour, bovine trabecular bone samples were subjected to compressive loading, creep, unloading and recovery at multiple load levels (corresponding to apparent strain of 2,000-25,000 με). The results show that: the time-dependent behaviour of trabecular bone comprises of both recoverable and irrecoverable strains; the strain response is nonlinearly related to applied load levels; and the response is associated with bone volume fraction. It was found that bone with low porosity demonstrates elastic stiffening followed by elastic softening, while elastic softening is demonstrated by porous bone at relatively low loads. Linear, nonlinear viscoelastic and nonlinear viscoelastic-viscoplastic constitutive models were developed to predict trabecular bone's time-dependent behaviour. Nonlinear viscoelastic constitutive model was found to predict the recovery behaviour well, while nonlinear viscoelastic-viscoplastic model predicts the full creep-recovery behaviour reasonably well. Depending on the requirements all these models can be used to incorporate time-dependent behaviour in finite element models. To evaluate the contribution of the key constituents of trabecular bone and its microstructure, tests were conducted on demineralised and deproteinised samples. Reversed cyclic loading experiments (tension to compression) were conducted on demineralised trabecular bone samples. It was found that demineralised bone exhibits asymmetric mechanical response - elastic stiffening in tension and softening in compression. This tension to compression transition was found to be smooth. Tensile multiple-load-creep-unload-recovery experiments on demineralised trabecular samples show irrecoverable strain (or residual strain) even at the low stress levels. Demineralised trabecular bone samples demonstrate elastic stiffening with increasing load levels in tension, and their time-dependent behaviour is nonlinear with respect to applied loads . Nonlinear viscoelastic constitutive model was developed which can predict its recovery behaviour well. Experiments on deproteinised samples showed that their modulus and strength are reasonably well related to bone volume fraction. The study considers an application of time-dependent behaviour of trabecular bone. Time-dependent properties are assigned to trabecular bone in a bone-screw system, in which the screw is subjected to cyclic loading. It is found that separation between bone and the screw at the interface can increase with increasing number of cycles which can accentuate loosening. The relative larger deformation occurs when this system to be loaded at the higher loading frequency. The deformation at the bone-screw interface is related to trabecular bone's bone volume fraction; screws in a more porous bone are at a higher risk of loosening.

Este trabalho mostra a comparação dos efeitos das falhas provocadas pelos Single-Event Effects em dispositivos 28nm FDSOI, 28nm FDSOI High-K e 32nm Bulk CMOS e células de memória 6T SRAM feitas com estes dispositivos. Para conseguir isso, foram usadas ferramentas TCAD para simular falhas transientes devido a impacto de íons pesados a nível dispositivo e nível circuito. As simulações neste ambiente tem como vantagem a simulação dos fatos e mecanismos que produz as falhas transientes e seus efeitos nos dispositivos, além de também servir para projetar virtualmente estes dispositivos e caraterizar eles para estas simulações. Neste caso, foram projetados três dispositivos para simulação: um transistor NMOS de 32nm Bulk, um transistor NMOS de 28nm FDSOI e um transistor NMOS de 28nm FDSOI High-K para fazer comparações entre eles. Estes dispositivos foram projetados, caraterizados e testados contra o impacto de íons pesados a níveis dispositivo e circuito. Como resultado obtido, transistor Bulk de 32nm teve, no pior caso, uma carga coletada de 7.57 e 7.19 vezes maior que a carga coletada pelo dispositivo FDSOI de 28nm e FDSOI High-K de 28nm respectivamente atingido pelo mesmo íon pesado de 100MeV-cm2/mg. Com estes dados foi possível modelar o comportamento da carga coletada de ambos dispositivos usando este íon pesado, atingindo os terminais de Fonte e Dreno em distintos lugares e ângulos. Usando a mesma ferramenta e os dados obtidos de carga coletada pelos testes anteriores, foram projetadas células de memória SRAM de 6 transistores. Isso foi para testar elas contra os efeitos do impacto de íons pesados nos transistores NMOS de armazenagem da dados. Neste caso, a Transferência Linear de Energia (LET) do íon necessária para fazer que o dado armazenado na SRAM Bulk mude é 12.8 vezes maior que no caso da SRAM FDSOI e 10 maior no caso da SRAM FDSOI High-K, embora a quantidade de carga coletada necessária para que o dado mude em ambas células seja quase a mesma. Com estes dados foi possível modelar os efeitos dos íons pesados em ambos circuitos, descobrir a Carga Crítica destes e qual é o mínimo LET necessário para que o dado armazenado nestas SRAMs mude. / This work shows a comparison of faults due to Single-Event Effects in 28nm Fully Depleted SOI (FDSOI), 28nm FDSOI High-K and 32nm Bulk CMOS devices, and in 6T SRAM memory cells made with these devices. To provide this, was used TCAD tools to simulate transient faults due to heavy ion impacts on device and circuit levels. The simulations in that environment have the advantage to simulate the facts and mechanisms which produce the transient faults and this effects on the electronic devices, it also allow to simulate the virtual device fabrication and to characterize them. In this case, two devices were created for the simulations: a 32nm Bulk NMOS transistor and a 28nm FDSOI NMOS transistor for compare them. These devices were created, characterized and tested against heavy ion impacts at device and circuit levels. The results show that 32nm Bulk transistor has, in the worst case, a collected charge 7.57 and 7.19 times greater than the 28nm FDSOI and 28nm FDSOI High-K respectively collected charge with the same 100MeV-cm2/mg heavy ion. With these data it was possible to model the behavior of the collected charge in both devices with the same heavy-ion, reach the Source and Drain Terminal in different places and angles. Using the same tools and the obtained collected charge data of previous simulations, it was designed 6 transistors SRAM Memory Cells. That is done to test these circuits against the heavy ion effects on the data-storage NMOS transistor. In this case, the necessary Ion Linear Energy Transfer (LET) to flip the Bulk SRAM is 12.8 greater than the FDSOI SRAM and 10 times greater than the FDSOI High- K SRAM case, although the amount of charge to flip the cells is almost the same in both cases. With these data it was possible to model the heavy-ion effects in both circuits, discover the Critical Charge of them and the minimum LET to flips these SRAMs.