1,721,176 research outputs found
Moment independent and reliability-based importance measures
No full textThis chapter discusses the class of moment independent importance measures. This class comprises density-based, cumulative distribution function based and value of information based sensitivity measures. The chapter illustrates the definition and properties of these importance measures as they have been proposed in the literature, reviewing a common rationale that envelops them, as well as recent results that concern the general properties of global sensitivity measures. The final part of the chapter reviews importance measures developed in the context of reliability theory
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Uncertainty quantification in multi-physics model for wind turbine asset management
No full textL'éolien en mer est l'un des moyens de réduire la part des énergies fossiles dans le mix électrique mondial. Cette technologie bénéficie de vents plus réguliers que l'éolien terrestre, principalement en raison de l'absence de relief. L'exploitation en mer permet également d'installer des éoliennes plus grandes et plus puissantes, ce qui pose plusieurs problèmes concernant la logistique portuaire, les besoins en ressources naturelles rares et les processus de recyclage en fin de vie.Les éoliennes en mer sont des systèmes dynamiques qui interagissent avec un environnement fortement incertain. Le traitement des incertitudes autour de leurs modèles de simulation numérique est donc essentiel pour proposer une conception et une exploitation fiable. Cependant, l'application d'une quantification d'incertitudes à ces systèmes soulève de nombreuses problématiques et nécessite le couplage de données avec des modèles numériques multiphysiques.Cette thèse traite d'abord des problèmes liés à la modélisation probabiliste multivariée des conditions environnementales en mer. Une approche semi-paramétrique est proposée, mélangeant des méthodes paramétriques pour l'ajustement des marginales avec la copule empirique de Bernstein pour approximer la structure de dépendance complexe entre les variables environnementales. Après avoir défini un modèle probabiliste des conditions environnementales, les perturbations causées par l'effet de sillage des turbines sont étudiées à l'échelle de la ferme en créant des groupes de turbines présentant des perturbations similaires. Ce regroupement préliminaire vise à réduire le nombre d'études de chargement à l'échelle d'une ferme (par exemple, pour l'évaluation de la fatigue).Le deuxième axe méthodologique de ce travail concerne la propagation d'incertitudes pour l'estimation de tendances centrales ou d'événements rares. Comme alternative aux cas de charge recommandés par les normes internationales pour le calcul du dommage en fatigue moyen, la méthode du "kernel herding" s'est avérée être une solution efficace et flexible pour réaliser ce calcul à partir de données (c'est-à-dire en sous-échantillonnant des données environnementales sans inférence). Cette méthode de quadrature Bayésienne est également adaptée à la construction de bases de test pour l'estimation de la précision moyenne de modèles d'apprentissage statistique.Pour estimer des événements rares, une nouvelle méthode intégrant une copule non paramétrique dans un mécanisme d'échantillonnage préférentiel adaptatif est proposée. Cette approche donne des résultats équivalents à la méthode de "subset simulation" tout en évitant un échantillonnage de Monte Carlo par chaîne de Markov. Par la suite, la fiabilité en fatigue d'une éolienne en mer est étudiée en tenant compte des incertitudes liées à l'environnement, mais aussi celles d'autres variables comme la rigidité du sol, l'erreur d'alignement de la nacelle, la courbe de Wöhler et la résistance critique à la fatigue. La robustesse de la fiabilité estimée est ensuite étudiée à l'aide des "perturbed-law based indices". Enfin, pour assurer la reproductibilité des résultats numériques, la plupart des développements présentés dans ce travail sont open source et documentés.Offshore wind energy is one of the ways to reduce the share of fossil fuels in the global electricity mix. This technology benefits from more consistent winds than the onshore one, mainly due to the absence of terrain roughness. Operating offshore also allows the installation of larger and more powerful wind turbines, which poses several scaling issues about port logistics, the demand for critical natural resources, and sustainable end-of-life processes.Offshore wind turbines are dynamic systems interacting with a highly uncertain environment.Uncertainty quantification of the multi-physics numerical models used to simulate them is therefore essential to propose risk-informed design and operation. However, developing a dedicated uncertainty quantification strategy for these systems raises numerous questions and requires coupling data with multi-physics numerical models.This thesis first addresses the problems related to the multivariate probabilistic modeling of offshore environmental conditions. A semiparametric approach is suggested, mixing parametric methods for marginals fitting with the empirical Bernstein copula to fit the complex dependence structure among environmental variables. After defining a probabilistic model of the ambient metocean conditions, the perturbations caused by the turbines' wake effect are studied at the farm scale by creating clusters of similarly perturbed turbines. This preliminary clustering aims to reduce the number of loading studies at the farm scale (e.g., for fatigue assessment).The second methodological axis of this work concerns uncertainty propagation for both central study and rare event estimation. As an alternative to the design load cases recommended by international standards for the mean cumulative damage estimation, the kernel herding method proved to be an efficient and flexible solution for given-data uncertainty propagation (i.e., directly subsampling from a large dataset without inference). This Bayesian quadrature method is also well-suited for the construction of test samples for the estimation of the mean predictivity of statistical learning models.For rare event estimation, a new method incorporating a nonparametric copula into an adaptive importance sampling mechanism is proposed. This method displays equivalent results to splitting methods as the subset simulation while avoiding any Markov Chain Monte Carlo sampling and thus generating independent and identically distributed samples. Then, the fatigue reliability of an offshore wind turbine is studied with respect to uncertain environmental variables while considering other variables related to soil stiffness, yaw misalignment, stress-number of cycles curve, and critical damage resistance. The robustness of the estimated reliability is then studied using perturbed-law based indices. Finally, to ensure the reproducibility of the numerical results, most of the developments presented in this work are open source and documented
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Study of the interaction between data acquisition (in situ and in laboratory) and the statistical/geostatistical modelisation : application to radioactive caracterization
No full textDans les projets d’assainissement/démantèlement de sites nucléaires, l’étape de caractérisation radiologique initiale a pour objectif d’estimer la quantité et la répartition spatiale de la contamination en différents radionucléides. Pour réaliser cette estimation, des mesures sont réalisées sur site et en laboratoire. Cependant compte-tenu des environnements et de la nature des mesures, le nombre de mesures peut être réduit et/ou ces mêmes mesures peuvent êtres censurées. Avec ce type de jeu de données, les approches classiques de géostatistique n’offrent pas de solutions pratiques pour les traiter. Dans cette thèse, plusieurs méthodes permettant de traiter les problématiques posées par ces jeux de données sont étudiées. Parmi ces méthodes, le krigeage bayésien sera particulièrement approfondi, puisqu’il permet de construire des modèles efficaces lorsque peu d’observations sont disponibles. Ce krigeage étant singulier, de nouveaux outils de validation seront introduits, et le krigeage bayésien sera comparé à d’autres méthodes usuelles comme le krigeage ordinaire. Nous proposons également une variante d’un algorithme MCMC (Méthode de Monte-Carlo par Chaînes de Markov) pour la résolution des équations du krigeage bayésien ainsi que pour l’augmentation de données, méthode permettant le traitement de données censurées.In decommissioning projects of nuclear facilities, the radiological characterisation step aims to estimate the quantity and spatial distribution of radionucleides. Statistical tools such as ordinary kriging (stemming from geostatistics) is part of usual industrial methods for soils sanitizing. To carry out the estimation, measurements are performed in situ or in laborator. However, due to the constrained environment and the nature of measurements, the data set’s size can be reduced and/or measurement results can be censored. With these types of data sets, usual geostatistical practices do not offer practical solutions. In this thesis we study several methods allowing the treatment of these specific problematics. We will especially dwell on one of these methods, called Bayesian kriging, since it allows to treat efficiently these kind of data sets. This kriging type being singular, new validation criterion will be introduced and Bayesian kriging will be compared to other usual methods like ordinary kriging. We also suggest a variant of a MCMC algorithm to solve Bayesian kriging equations and data augmentation (which allows the processing of censored data
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
