A modified structural stress method for fatigue assessment of welded structures

Fatigue life assessment of weldedstructures is commonly based on the nominal stress method, but more flexible and accurate methods have been introduced. In general, the assessment accuracy is improved as more localized information about the weld is incorporated. The structural hot spot stress method includes the influence of macro geometric effects and structural discontinuities on the design stress but excludes the local features of the weld. In this thesis, the limitations of the structural hot spot stress method are discussed and a modified structural stress method with improved accuracy is developed and verified for selected welded details. The fatigue life of structures in the as-welded state consists mainly of crack growth from pre-existing cracks or defects. Crack growth rate depends on crack geometry and the stress state on the crack face plane. This means that the stress level and shape of the stress distribution in the assumed crack path governs thetotal fatigue life. In many structural details the stress distribution is similar and adequate fatigue life estimates can be obtained just by adjusting the stress level based on a single stress value, i.e., the structural hot spot stress. There are, however, cases for which the structural stress approach is less appropriate because the stress distribution differs significantly from the more common cases. Plate edge attachments and plates on elastic foundations are some examples of structures with this type of stress distribution. The importance of fillet weld size and weld load variation on the stress distribution is another central topic in this thesis. Structural hot spot stress determination is generally based on a procedure that involves extrapolation of plate surface stresses. Other possibilities for determining the structural hot spot stress is to extrapolate stresses through the thickness at the weld toe or to use Dong's method which includes through-thickness extrapolation at some distance from the weld toe. Both of these latter methods are less sensitive to the FE mesh used. Structural stress based on surface extrapolation is sensitive to the extrapolation points selected and to the FE mesh used near these points. Rules for proper meshing, however, are well defined and not difficult to apply. To improve the accuracy of the traditional structural hot spot stress, a multi-linear stress distribution is introduced. The magnitude of the weld toe stress after linearization is dependent on the weld size, weld load and plate thickness. Simple equations have been derived by comparing assessment results based on the local linear stress distribution and LEFM based calculations. The proposed method is called the modified structural stress method (MSHS) since the structural hot spot stress (SHS) value is corrected using information on weld size andweld load. The correction procedure is verified using fatigue test results found in the literature. Also, a test case was conducted comparing the proposed method with other local fatigue assessment methods.

Unsupervised visual object categorization

The number of digital images has been increasing exponentially in the last few years. People have problems managing their image collections and finding a specific image. An automatic image categorization system could help them to manage images and find specific images. In this thesis, an unsupervised visual object categorization system was implemented to categorize a set of unknown images. The system is unsupervised, and hence, it does not need known images to train the system which needs to be manually obtained. Therefore, the number of possible categories and images can be huge. The system implemented in the thesis extracts local features from the images. These local features are used to build a codebook. The local features and the codebook are then used to generate a feature vector for an image. Images are categorized based on the feature vectors. The system is able to categorize any given set of images based on the visual appearance of the images. Images that have similar image regions are grouped together in the same category. Thus, for example, images which contain cars are assigned to the same cluster. The unsupervised visual object categorization system can be used in many situations, e.g., in an Internet search engine. The system can categorize images for a user, and the user can then easily find a specific type of image.

Transnationalisation religieuse : un temple d’umbanda à Montréal

Les pratiques religieuses dans les pays de l’Atlantique Nord se transforment et on observe pour une partie de leur population le passage d’un « croire institutionnalisé » à une spiritualité influencée par diverses traditions, dont certaines ont fait leur apparition sur ce territoire au milieu du 20e siècle. Le présent mémoire vise à mettre en lumière une des facettes de la diversité religieuse contemporaine; suite aux questionnements qui ont surgi au long du travail, il aborde aussi certains enjeux sous-jacents à l’analyse d’un groupe religieux en anthropologie, notamment comment aborder le terrain et comment considérer un tel groupe. J’ai choisi pour ce faire de décrire un groupe religieux québécois qui est lié à un culte afro-brésilien – l’umbanda– et qui est membre d’un réseau transnational de temples. J’examinerai d’abord comment l’umbanda s’est développée au Brésil, car cela fournira des indications pertinentes sur, entre autres choses, l’éventuelle perméabilité de cette tradition, une fois le groupe implanté dans un pays de l’Atlantique Nord. J’examinerai ensuite le type de transnationalisation qui a présidé à la naissance du temple à Montréal, car cette analyse offre des indices permettant de déterminer comment un tel groupe s’insère dans le paysage religieux de divers pays. Enfin, en me penchant sur les pratiques religieuses des membres du groupe, je tenterai de dégager certaines particularités locales.

Understanding deep architectures and the effect of unsupervised pre-training

Cette thèse porte sur une classe d'algorithmes d'apprentissage appelés architectures profondes. Il existe des résultats qui indiquent que les représentations peu profondes et locales ne sont pas suffisantes pour la modélisation des fonctions comportant plusieurs facteurs de variation. Nous sommes particulièrement intéressés par ce genre de données car nous espérons qu'un agent intelligent sera en mesure d'apprendre à les modéliser automatiquement; l'hypothèse est que les architectures profondes sont mieux adaptées pour les modéliser. Les travaux de Hinton (2006) furent une véritable percée, car l'idée d'utiliser un algorithme d'apprentissage non-supervisé, les machines de Boltzmann restreintes, pour l'initialisation des poids d'un réseau de neurones supervisé a été cruciale pour entraîner l'architecture profonde la plus populaire, soit les réseaux de neurones artificiels avec des poids totalement connectés. Cette idée a été reprise et reproduite avec succès dans plusieurs contextes et avec une variété de modèles. Dans le cadre de cette thèse, nous considérons les architectures profondes comme des biais inductifs. Ces biais sont représentés non seulement par les modèles eux-mêmes, mais aussi par les méthodes d'entraînement qui sont souvent utilisés en conjonction avec ceux-ci. Nous désirons définir les raisons pour lesquelles cette classe de fonctions généralise bien, les situations auxquelles ces fonctions pourront être appliquées, ainsi que les descriptions qualitatives de telles fonctions. L'objectif de cette thèse est d'obtenir une meilleure compréhension du succès des architectures profondes. Dans le premier article, nous testons la concordance entre nos intuitions---que les réseaux profonds sont nécessaires pour mieux apprendre avec des données comportant plusieurs facteurs de variation---et les résultats empiriques. Le second article est une étude approfondie de la question: pourquoi l'apprentissage non-supervisé aide à mieux généraliser dans un réseau profond? Nous explorons et évaluons plusieurs hypothèses tentant d'élucider le fonctionnement de ces modèles. Finalement, le troisième article cherche à définir de façon qualitative les fonctions modélisées par un réseau profond. Ces visualisations facilitent l'interprétation des représentations et invariances modélisées par une architecture profonde.

Spectral clustering independent component analysis for tissue classification from brain MRI

A spectral angle based feature extraction method, Spectral Clustering Independent Component Analysis (SC-ICA), is proposed in this work to improve the brain tissue classification from Magnetic Resonance Images (MRI). SC-ICA provides equal priority to global and local features; thereby it tries to resolve the inefficiency of conventional approaches in abnormal tissue extraction. First, input multispectral MRI is divided into different clusters by a spectral distance based clustering. Then, Independent Component Analysis (ICA) is applied on the clustered data, in conjunction with Support Vector Machines (SVM) for brain tissue analysis. Normal and abnormal datasets, consisting of real and synthetic T1-weighted, T2-weighted and proton density/fluid-attenuated inversion recovery images, were used to evaluate the performance of the new method. Comparative analysis with ICA based SVM and other conventional classifiers established the stability and efficiency of SC-ICA based classification, especially in reproduction of small abnormalities. Clinical abnormal case analysis demonstrated it through the highest Tanimoto Index/accuracy values, 0.75/98.8%, observed against ICA based SVM results, 0.17/96.1%, for reproduced lesions. Experimental results recommend the proposed method as a promising approach in clinical and pathological studies of brain diseases

Wavelet based Independent Component Analysis for Multispectral Brain Tissue Classification

Multispectral analysis is a promising approach in tissue classification and abnormality detection from Magnetic Resonance (MR) images. But instability in accuracy and reproducibility of the classification results from conventional techniques keeps it far from clinical applications. Recent studies proposed Independent Component Analysis (ICA) as an effective method for source signals separation from multispectral MR data. However, it often fails to extract the local features like small abnormalities, especially from dependent real data. A multisignal wavelet analysis prior to ICA is proposed in this work to resolve these issues. Best de-correlated detail coefficients are combined with input images to give better classification results. Performance improvement of the proposed method over conventional ICA is effectively demonstrated by segmentation and classification using k-means clustering. Experimental results from synthetic and real data strongly confirm the positive effect of the new method with an improved Tanimoto index/Sensitivity values, 0.884/93.605, for reproduced small white matter lesions

State-of-the-Art: Transformation Invariant Descriptors

As the popularity of digital videos increases, a large number illegal videos are being generated and getting published. Video copies are generated by performing various sorts of transformations on the original video data. For effectively identifying such illegal videos, the image features that are invariant to various transformations must be extracted for performing similarity matching. An image feature can be its local feature or global feature. Among them, local features are powerful and have been applied in a wide variety of computer vision aplications .This paper focuses on various recently proposed local detectors and descriptors that are invariant to a number of image transformations.

Understanding projections of sea level rise in a Hadley Centre coupled climate model

Sea level changes resulting from CO2-induced climate changes in ocean density and circulation have been investigated in a series of idealised experiments with the Hadley Centre HadCM3 AOGCM. Changes in the mass of the ocean were not included. In the global mean, salinity changes have a negligible effect compared with the thermal expansion of the ocean. Regionally, sea level changes are projected to deviate greatly from the global mean (standard deviation is 40% of the mean). Changes in surface fluxes of heat, freshwater and wind stress are all found to produce significant and distinct regional sea level changes, wind stress changes being the most important and the cause of several pronounced local features, while heat and freshwater flux changes affect large parts of the North Atlantic and Southern Ocean. Regional change is related mainly to density changes, with a relatively small contribution in mid and high latitudes from change in the barotropic circulation. Regional density change has an important contribution from redistribution of ocean heat content. In general, unlike in the global mean, the regional pattern of sea level change due to density change appears to be influenced almost as much by salinity changes as by temperature changes, often in opposition. Such compensation is particularly marked in the North Atlantic, where it is consistent with recent observed changes. We suggest that density compensation is not a property of climate change specifically, but a general behavior of the ocean.

Multiscale Analysis of Topographic Surface Roughness in the Midland Valley, Scotland

Surface roughness is an important geomorphological variable which has been used in the Earth and planetary sciences to infer material properties, current/past processes, and the time elapsed since formation. No single definition exists; however, within the context of geomorphometry, we use surface roughness as an expression of the variability of a topographic surface at a given scale, where the scale of analysis is determined by the size of the landforms or geomorphic features of interest. Six techniques for the calculation of surface roughness were selected for an assessment of the parameter`s behavior at different spatial scales and data-set resolutions. Area ratio operated independently of scale, providing consistent results across spatial resolutions. Vector dispersion produced results with increasing roughness and homogenization of terrain at coarser resolutions and larger window sizes. Standard deviation of residual topography highlighted local features and did not detect regional relief. Standard deviation of elevation correctly identified breaks of slope and was good at detecting regional relief. Standard deviation of slope (SD(slope)) also correctly identified smooth sloping areas and breaks of slope, providing the best results for geomorphological analysis. Standard deviation of profile curvature identified the breaks of slope, although not as strongly as SD(slope), and it is sensitive to noise and spurious data. In general, SD(slope) offered good performance at a variety of scales, while the simplicity of calculation is perhaps its single greatest benefit.

The state of the youth: prisons, drugs and car crashes

By virtue of the volume and nature of their attributions, including secondary school as well as problem-areas such as security and traffic, the Brazilian states are the ultimate responsible entities for young people. This study argues in favour of granting greater freedom for the states to define their own public policy parameters to deal with local features and to increase the degree of learning about such actions at the national level. In empirical terms, the study assesses the impacts of new laws, such as the new traffic code (from the joint work with Leandro Kume, EPGE/FGV doctor’s degree student) and traces the statistics for specific questions like drugs, violence and car accidents. The findings show that these questions produce different results for young men and women.The main characters in these dramas are young single males, suggesting the need for more distinguished public policies according not only to age, but also by gender. The study also reveals that the magnitude of these problems changes according to the youth’s social class. Prisons concern poorer men (except for the functional illiterate) while fatal car accidents and the confessed use of drugs concern upper-class boys.

Assesment of port governance in Brazil: a managerial perspective on the port of Santos

The role of maritime transportation within international trade was drastically revamped during the inception of the globalization process, which enhanced the contribution of ports in world economy as main logistics gateways for global production and trade. As a result, the relationship between ports and governments has changed. Devolution ideologies that had been applied in other industries decades ago were now being considered by governments for the port industry. Many central governments sought to extract themselves from commercial activities of ports and devolving this responsibility to local governments, communities or private entities. The institution of devolution programs also changed the governance structures of ports further influencing port performance. Consequently, the recent worldwide trend towards devolution in the port industry has spawned considerable variety of governance models that are now set in place around the world. While some countries opt for more decentralized structures others prefer to retain a centralization of powers. In this way some governments consider local features and national integration more than others, which ultimately influence the success of a port reform implementation. Nevertheless, the prime intent of governments is now to maximize the efficiency and performance of their domestic ports. This issue intends to examine the changed port governance environment in Brazil by determining how and why imposed port reforms of the Brazilian federal government have been affecting the overall performance of the national port system, over the last decades, using the Port of Santos as a sample upon an exploratory study. For that, the study will use a contingency theory-based framework – the Matching Framework - that views port performance as a function of the fit among the dimensions of external operating environment, strategy and structure of a port organization. In essence, the greater the fit among these dimensions the better the expected performance of a port will be, and vice-versa. Port managers, government officials and academics alike shall be interested in this document.

3-D reconstruction by extended depth-of-field in failure analysis - Case study II: Fractal analysis of interlaminar fracture in carbon/epoxy composites

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Population and reproductive biology of the crab Uca burgersi (Crustacea : Ocypodidae) in three subtropical mangrove forests

Population and reproductive biology were studied in three populations of the crab Uca burgersi Holthuis, 1967, in the Indaia, Cavalo and Ubatumirim mangrove forests (Ubatuba, São Paulo State, Brazil). Crabs were collected during low tide (August 2001 through July 2002), by digging the sediment, with a standard capture effort (two persons for 30 min.). Carapace width was measured, and gonad developmental stage was recorded from all specimens. U. burgersi was most abundant in the Cavalo mangrove, where the largest mate was found. Juvenile crabs were found year-round at all three sites. However, the ratio of ovigerous females was very low, even null in the Cavalo mangrove. The gonad development rate indicated that U. burgersi was reproducing continuously, but more intensively during spring and summer, with recruitment occurring in winter. The synchrony between the populational and reproductive biology in the three areas showed that local features were not the limiting factors. It is suggested that this species is a habitat generalist.

Aspectos epidemiológicos do tracoma em crianças do ensino fundamental do minicípio de Bauru-SP: a utilização do geoprocessamento na priorização de recursos do setor saúde

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Ontogenetic distribution of Callinectes ornatus (Decapoda, Portunoidea) in southeastern Brazil

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)