No description available.

No description available.

No / Extrusion is one of the major methods for processing polymeric materials and the thermal homogeneity of the process output is a major concern for manufacture of high quality extruded products. Therefore, accurate process thermal monitoring and control are important for product quality control. However, most industrial extruders use single point thermocouples for the temperature monitoring/control although their measurements are highly affected by the barrel metal wall temperature. Currently, no industrially established thermal profile measurement technique is available. Furthermore, it has been shown that the melt temperature changes considerably with the die radial position and hence point/bulk measurements are not sufficient for monitoring and control of the temperature across the melt flow. The majority of process thermal control methods are based on linear models which are not capable of dealing with process nonlinearities. In this work, the die melt temperature profile of a single screw extruder was monitored by a thermocouple mesh technique. The data obtained was used to develop a novel approach of modelling the extruder die melt temperature profile under dynamic conditions (i.e. for predicting the die melt temperature profile in real-time). These newly proposed models were in good agreement with the measured unseen data. They were then used to explore the effects of process settings, material and screw geometry on the die melt temperature profile. The results showed that the process thermal homogeneity was affected in a complex manner by changing the process settings, screw geometry and material. (C) 2013 Elsevier Inc. All rights reserved.

I explore five topics in topology using set-theoretic techniques. The first of these is a generalization of 2-point sets called slice sets. I show that, for any small-in-cardinality subset A of the real line, there is a subset of the plane meeting every line in a topological copy of A. Under Martin's Axiom, I show how to improve this result to any totally disconnected A. Secondly, I show that it is consistent with and independent of ZFC to have a topologically rigid subset of the real line that is smaller than the continuum. Thirdly, I define and examine a new cardinal function related to cleavability. Fourthly, I explore the Toronto Problem and prove that any uncountable, Hausdorff, non-discrete Toronto space that is not regular falls into one of two strictly-defined classes. I also prove that for every infinite cardinality there are precisely 3 non-T1 Toronto spaces up to homeomorphism. Lastly, I examine a notion of dimension called the "neight", and prove several theorems that give a lower bound for this cardinal function.

L’objet de ce travail de thèse est le regroupement des essences bois du bassin du Congo en quatre groupes en fonction de la similitude de leurs propriétés mécaniques essentielles. Le module d’élasticité (MOE), propriété indicatrice, est estimé par une méthode non destructive : l’essai vibratoire. Pour chaque classe, le fractile à 5% d’exclusion supérieur d’une propriété mécanique est calculé à partir d’une théorie clairement exposée. Le bassin du Congo est un patrimoine en danger à cause des pressions anthropiques au premier rang desquelles figure la coupe hyper sélective des essences bois. Le résultat de ce regroupement permet de proposer comme substitut aux essences menacées d’extinction des essences aux propriétés mécaniques similaires plus disponibles en terme de ressources sur pied. D’où la contribution à la gestion durable et au maintien de la biodiversité du bassin du Congo. On montre que le MOE vibratoire est quasi-invariant avec la teneur en eau et qu’il est bien corrélé au MOE statique, à la masse volumique et à la contrainte à la rupture. C’est donc un bon prédicteur du module de rupture (MOR) à 12% du taux d’humidité du bois vert. Ceci valide l’organisation en groupe d’essences réalisée dans ce travail. Les valeurs caractéristiques des propriétés mesurées sont calculées pour chaque classe. Parce que l’Eurocode 5 exige que les bois utilisés en structure soient classés, nous proposons, in fine, un système de classement en structure des bois du bassin du Congo voué à une meilleure utilisation des forêts pour la conservation des écosystèmes forestiers et la satisfaction des besoins fondamentaux de l’homme. / This Ph.D. work deals with a non-destructive experimental approach organizing the species of the Congo basin in four groups according to the likeness of their main mechanical properties, and to guarantee for each group the 5th percentile characteristic value of mechanical properties such as the modulus of elasticity (MOE). For the sustainable management of forest and conservation of exotic biodiversity of the Congo basin, it is necessary to promote the exploitation of less-consumed species with equivalent mechanical characteristics. The aim of the present work is to propose a scheme for grouping species with similar values of modulus of elasticity (MOE) obtained through vibratory method. Only small clear specimens are tested. The assignment of a given specie to one of the four groups takes place in return for a homogeneity test of comparison of the random variables of species to the random variable of the group. Results of the present grading show that some species can be interchangeably consumed in the construction industry and that the procedure must include more species of the Congo basin. This property grouping will facilitate substitution of underutilized species for ones that are being more exploited. Thus, wood designers and constructors might have a wide variety of choices in their decision-making while promoting less-consumed species and reducing the demand of traditionally most wanted extinguishable species. The MOE obtained by intrinsic vibration of wood versus is not influenced by moisture content. So, the non destructive method bases on longitudinal vibrations can be used in industrial production of timber to certify the modulus of rupture (MOR) at 12% moisture content of green wood by direct correlation. Finally, we propose a grading system of tropical wood.

In der hier vorliegenden Arbeit wurde die Validität von Aufmerksamkeits- und Konzentrationstests untersucht. Im Vordergrund stand dabei die Frage nach dem Einfluss verschiedener kritischer Variablen auf die prädiktive Validität in diesen Tests, insbesondere der Itemschwierigkeit und Itemhomogenität, der Testlänge bzw. des Testverlaufs, der Testdiversifikation und der Validität im Kontext einer echten Personalauslese. In insgesamt fünf Studien wurden die genannten Variablen systematisch variiert und auf ihre prädiktive Validität zur (retrograden und konkurrenten) Vorhersage von schulischen und akademischen Leistungen (Realschule, Abitur, Vordiplom/Bachelor) hin analysiert. Aufgrund der studentischen (d. h. relativ leistungshomogenen) Stichprobe bestand die Erwartung, dass die Korrelationen etwas unterschätzt werden. Da die Validität in dieser Arbeit jedoch „vergleichend“ für bestimmte Tests bzw. experimentelle Bedingungen bestimmt wurde, sollte dies keine Rolle spielen. In Studie 1 (N = 106) wurde zunächst untersucht, wie schwierig die Items in einem Rechenkonzentrationstest sein sollten, um gute Vorhersagen zu gewährleisten. Dazu wurden leichte und schwierigere Items vergleichend auf ihre Korrelation zum Kriterium hin untersucht. Im Ergebnis waren sowohl leichte als auch schwierigere Testvarianten ungefähr gleich prädiktiv. In Studie 2 (N = 103) wurde die Rolle der Testlänge untersucht, wobei die prädiktive Validität von Kurzversion und Langversion in einem Rechenkonzentrationstest vergleichend untersucht wurde. Im Ergebnis zeigte sich, dass die Kurzversion valider war als die Langversion und dass die Validität in der Langversion im Verlauf abnimmt. In Studie 3 (N = 388) stand der Aspekt der Testdiversifikation im Vordergrund, wobei untersucht wurde, ob Intelligenz besser mit einem einzelnen Matrizentest (Wiener Matrizen-Test, WMT) oder mit einer Testbatterie (Intelligenz-Struktur-Test, I-S-T 2000 R) erfasst werden sollte, um gute prädiktive Validität zu gewährleisten. Die Ergebnisse sprechen klar für den Matrizentest, welcher ungefähr gleich valide war wie die Testbatterie, aber dafür testökonomischer ist. In den Studien 4 (N = 105) und 5 (N =97) wurde die prädiktive Validität zur Vorhersage von Schulleistungen im Kontext einer realen Personalauswahlsituation untersucht. Während die großen Testbatterien, Wilde-Intelligenz-Test 2 (WIT-2) und Intelligenz-Struktur-Test 2000R (I-S-T 2000 R), nur mäßig gut vorhersagen konnten, war der Komplexe Konzentrationstest (KKT), insbesondere der KKT-Rechentest ein hervorragender Prädiktor für schulische und akademische Leistungen. Auf Basis dieser Befunde wurden schließlich Empfehlungen und Anwendungshilfen für den strategischen Einsatz von Testinstrumenten in der diagnostischen Berufspraxis ausgesprochen.

La première partie de la thèse traite d’inférence combinatoire en Analyse Géométrique des Données (AGD). Nous proposons des tests multidimensionnels sans hypothèse sur le processus d’obtention des données ou les distributions. Nous nous intéressons ici aux problèmes de typicalité (comparaison d’un point moyen à un point de référence ou d’un groupe d’observations à une population de référence) et d’homogénéité (comparaison de plusieurs groupes). Nous utilisons des procédures combinatoires pour construire un ensemble de référence par rapport auquel nous situons les données. Les statistiques de test choisies mènent à des prolongements originaux : interprétation géométrique du seuil observé et construction d’une zone de compatibilité.La seconde partie présente l’étude de l’absentéisme dans les Industries Electriques et Gazières de 1995 à 2011 (avec construction d’une cohorte épidémiologique). Des méthodes d’AGD sont utilisées afin d’identifier des pathologies émergentes et des groupes d’agents sensibles. / The first part of this PhD thesis deals with combinatorial inference methods forGeometric Data Analysis (GDA). We propose multidimensional tests that make no assumption on the process of generating data or distributions. We focus particularly on problems of typicality (comparison of a mean point to a reference point or comparison of a group of observations to a reference population) and on problems of homogeneity (comparison of several groups). These methods consist in using combinatorial procedures to build a reference set with respect to which we situate the data. The chosen test statistics lead to original extensions: geometric interpretation of the observed level and construction of a compatibilityzone.The second part of this thesis presents the study of absenteeism in the French Electricity and Gas Industries from 1995 to 2011 (with construction of an epidemiological cohort). GDA methods are used to identify emerging diseases and sensitive groups of agents.

A disponibilidade de recursos, tanto de ordem financeira quanto de mão-de-obra, é escassa. Sendo assim, deve-se incentivar o planejamento regional que minimize a utilização de recursos. A previsão de safra por intermédio de técnicas de modelagem deve ser realizada anteriormente com base nas características regionais, indicando assim as diretrizes básicas da pesquisa, bem como o planejamento regional. Dessa forma, os objetivos deste trabalho são: (i) caracterizar as variáveis do clima por intermédio de diferentes distribuições de probabilidade; (ii) verificar a homogeneidade espacial e temporal para as variáveis do clima; (iii) utilizar a distribuição normal bivariada para simular parâmetros utilizados na estimação de produtividade da cultura de soja; e (iv) propor um modelo para estimar a ordem de magnitude da produtividade potencial (dependente da interação genótipo, temperatura, radiação fotossinteticamente ativa e fotoperíodo) e da produtividade deplecionada (dependente da podutividade potencial, da chuva e do armazenamento de água no solo) de grãos de soja, baseados nos valores diários de temperatura, insolação e chuva, para o estado de São Paulo. As variáveis utilizadas neste estudo foram: temperatura média, insolação, radiação solar fotossinteticamente ativa e precipitação pluvial, em escala diária, obtidas em 27 estações localizadas no Estado de São Paulo e seis estações localizadas em Estados vizinhos. Primeiramente, verificou-se a aderência das variáveis a cinco distribuições de probabilidade (normal, log-normal, exponencial, gama e weibull), por intermédio do teste de Kolmogorov-Smirnov. Verificou-se a homogeneidade espacial e temporal dos dados por intermédio da análise de agrupamento pelo método de Ward e estimou-se o tamanho de amostra (número de anos) para as variáveis. A geração de números aleatórios foi realizada por intermédio do método Monte Carlo. A simulação dos dados de radiação fotossinteticamente ativa e temperatura foram realizadas por intermédio de três casos (i) distribuição triangular assimétrica (ii) distribuição normal truncada a 1,96 desvio padrão da média e (iii) distribuição normal bivariada. Os dados simulados foram avaliados por intermédio do teste de homogeneidade de variância de Bartlett e do teste F, teste t, índice de concordância de Willmott, coeficiente angular da reta, o índice de desempenho de Camargo (C) e aderência à distribuição normal (univariada). O modelo utilizado para calcular a produtividade potencial da cultura de soja foi desenvolvido com base no modelo de De Wit, incluindo contribuições de Van Heenst, Driessen, Konijn, de Vries, dentre outros. O cálculo da produtividade deplecionada foi dependente da evapotranspiração potencial, da cultura e real e coeficiente de sensibilidade a deficiência hídrica. Os dados de precipitação pluvial foram amostrados por intermédio da distribuição normal. Sendo assim, a produção diária de carboidrato foi deplecionada em função do estresse hídrico e número de horas diárias de insolação. A interpolação dos dados, de modo a englobar todo o Estado de São Paulo, foi realizada por intermédio do método da Krigagem. Foi verificado que a maior parte das variáveis segue a distribuição normal de probabilidade. Além disso, as variáveis apresentam variabilidade espacial e temporal e o número de anos necessários (tamanho de amostra) para cada uma delas é bastante variável. A simulação utilizando a distribuição normal bivariada é a mais apropriada por representar melhor as variáveis do clima. E o modelo de estimação das produtividades potencial e deplecionada para a cultura de soja produz resultados coerentes com outros resultados obtidos na literatura. / The availability of resources, as much of financial order and human labor, is scarse. Therefore, it must stimulates the regional planning that minimizes the use of resources. Then, the forecast of harvests through modelling techniques must previously on the basis of be carried through the regional characteristics, thus indicating the routes of the research, as well as the regional planning. Then, the aims of this work are: (i) to characterize the climatic variables through different probability distributions; (ii) to verify the spatial and temporal homogeneity of the climatic variables; (iii) to verify the bivaried normal distribution to simulate parameters used to estimate soybean crop productivity; (iv) to propose a model of estimating the magnitud order of soybean crop potential productivity (it depends on the genotype, air temperature, photosynthetic active radiation; and photoperiod) and the depleted soybean crop productivity (it pedends on the potential productivity, rainfall and soil watter availability) based on daily values of temperature, insolation and rain, for the State of São Paulo. The variable used in this study had been the minimum, maximum and average air temperature, insolation, solar radiation, fotosynthetic active radiation and pluvial precipitation, in daily scale, gotten in 27 stations located in the State of São Paulo and six stations located in neighboring States. First, it was verified tack of seven variables in five probability distributions (normal, log-normal, exponential, gamma and weibull), through of Kolmogorov-Smirnov. The spatial and temporal verified through the analysis of grouping by Ward method and estimating the sample size (number of years) for the variable. The generation of random numbers was carried through the Monte Carlo Method. The simulation of the data of photosyntetic active radiation and temperature had been carried through three cases: (i) nonsymetric triangular distribution (ii) normal distribution truncated at 1.96 shunting line standard of the average and (iii) bivaried normal distribution. The simulated data had been evaluated through the test of homogeneity of variance of Bartlett and the F test, t test, agreement index of Willmott, angular coefficient of the straight line, the index of performance index of Camargo (C) and tack the normal distribution (univarieted). The proposed model to simulate the potential productivity of soybean crop was based on the de Wit concepts, including Van Heenst, Driessen, Konijn, Vries, and others researchers. The computation of the depleted productivity was dependent of the potential, crop and real evapotranspirations and the sensitivity hydric deficiency coefficient. The insolation and pluvial precipitation data had been showed through the normal distribution. Being thus, the daily production of carbohydrate was depleted as function of hydric stress and insolation. The interpolation of the data, in order to consider the whole State of Sao Paulo, was carried through the Kriging method. The results were gotten that most of the variable can follow the normal distribution. Moreover, the variable presents spatial and temporal variability and the number of necessary years (sample size) for each one of them is sufficiently changeable. The simulation using the bivaried normal distribution is most appropriate for better representation of climate variable. The model of estimating potential and depleted soybean crop productivities produces coherent values with the literature results.

The thesis aims to present Leibniz's monadic system as the simplest conceivable structure. To this end, the thesis employs both contemporary literature concerned with formal ontology and its logic, semantics and the nature of reference; and Leibniz's own writings coupled with correspondent commentaries, including articles transcending selected interpretive issues into present-day discussions on actualised versions of one of the essential Leibnizian principles: the identity of indiscernibles. The exposition proceeds in three steps: (i) developing concepts of structure, determination and reference in critical confrontation with contemporary approaches, then (ii) applying those concepts to the interpretation of Leibniz's principle of identity of indiscenibles and illuminating therewith links between ontology and semantics, and eventually (iii) utilizing those findings in clarifying the mirroring relation constitutive of the monadic structure as represented a) in Leibniz's own spatiotemporal illustrations and b) in formal models in secondary literature. Acknowledging that the simplest conceivable structure, which is the monadic structure, can neither be satisfactorily expressed by spatiotemporal illustrations, nor can it be represented in formal system, itself forming the limit of extrinsic...

Cette thèse porte sur l'étude des groupes polonais vus comme groupes d'automorphismes de structures métriques. L'observation que tout groupe polonais non archimédien est le groupe d'automorphismes d'une structure dénombrable ultra homogène a en effet mené à des interactions fructueuses entre la théorie des groupes et la théorie des modèles. Dans le cadre de la théorie des modèles métriques, introduite par Ben Yaacov, Henson et Usvyatsov, cette correspondance a été étendue par Melleray à tous les groupes polonais. Dans cette thèse, nous étudions diverses facettes de cette correspondance. Le lien entre une structure et son groupe d automorphismes est particulièrement étroit dans le cadre des structures ℵ0-categoriques. En effet, le théorème de reconstruction d'Ahlbrandt-Ziegler permet de retrouver une structure ℵ0-categorique, à bi-interprètabilité près, à partir de son groupe d'automorphismes. Dans un travail en commun avec Itai Ben Yaacov, nous généralisons ce résultat aux structures métriques separablement catégoriques. Les structures dénombrables ultra homogènes ont de plus l avantage d'être complètement déterminées par leurs sous-structures finiment engendrées. Cela a notamment permis a Moore de donner une caractérisation combinatoire de la moyennabilité des groupes polonais non archimédiens. Nous étendons cette caractérisation à tous les groupes polonais et nous en déduisons que la moyennabilite est une condition Gδ. Toujours dans une optique de reconstruction, nous nous intéressons à la propriété de continuité automatique pour les groupes polonais. Sabok et Malicki ont introduit des conditions de nature combinatoire sur une structure métrique ultra homogène qui impliquent la propriété de continuité automatique pour son groupe d'automorphismes. Nous montrons que ces conditions passent à la puissance dénombrable, ce qui a pour conséquence que les groupes Aut(μ)N, U(l2)N et Iso(U)N satisfont la propriété de continuité automatique. Ces conditions sont un affaiblissement du fait d'avoir des amples génériques. Dans un travail en commun avec Francois Le Maitre, nous exhibons les premiers exemples de groupes connexes qui ont des amples génériques, ce qui répond à une question de Kechris et Rosendal / This thesis focuses on the study of Polish groups seen as automorphism groups of metric structures. The observation that every non-archimedean Polish group is the automorphism group of an ultrahomogeneous countable structure has indeed led to fruitful interactions between group theory and model theory. In the framework of metric model theory, introduced by Ben Yaacov, Henson and Usvyastov, this correspondence has been extended to all Polish groups by Melleray. In this thesis, we study various facets of this correspondence. The relationship between a structure and its automorphism group is particularly close in the setting of ℵ0-categorical structures. Indeed, the Ahlbrandt-Ziegler reconstruction theorem allows one to recover an ℵ0-categorical structure, up to bi-interpretability, from its automorphism group. In a joint work with Itai Ben Yaacov, we generalize this result to separably categorical metric structures. Besides, ultrahomogeneous countable structures have the advantage of being completely determined by their finitely generated substructures. In particular, this enabled Moore to give a combinatorial characterization of amenability for nonarchimedean Polish groups. We extend this characterization to all Polish groups and we deduce that amenability is a Gδ condition. Still in a reconstruction perspective, we are interested in the automatic continuity property for Polish groups. Sabok and Malicki introduced conditions of a combinatorial nature on an ultrahomogeneous metric structure that imply the automatic continuity property for its automorphism group. We show that these conditions carry to countable powers, which leads to the groups Aut(μ)N, U(l2)N and Iso(U)N satisfying the automatic continuity property. Those conditions are a weakening of the property of having ample generics. In a joint work with Francois Le Maitre, we exhibit the first examples of connected groups with ample generics, which answers a question of Kechris and Rosendal. Finally, in a joint work with Isabel Muller and Aristotelis Panagiotopoulos, we study the relative homogeneity of substructures in an ultrahomogeneous countable structure. We characterize it completely by a property of the types over the substructures: being determined by a finite set