Nesta tese propomos o Full Bayesian Significance Test (FBST), apresentado por Pereira e Stern em 1999, para análise de modelos de misturas de normais multivariadas. Estendemos o conceito de modelos de misturas para explorar outro problema clássico em Estatística, o problema de modelos separados. Nas duas propostas, realizamos experimentos numéricos inspirados em problemas biológicos importantes: o problema de classificação não supervisionada de genes baseada em seus níveis de expressão, e o problema da discriminação entre os modelos Weibull e Gompertz - distribuições clássicas em análise de sobrevivência. / In this thesis we propose the Full Bayesian Significance Test (FBST) as a tool for multivariate normal mixture models. We extend the fundamental mixture concepts to another important problem in Statistics, the problem of separate models. In both methods, we perform numerical experiments based on important biological problems: the unsupervised classification of genes based on their expression profiles, and the problem of deciding between the Weibull and Gompertz models - two classical distributions widely used in survival analysis.

About 85% of the world’s energy is derived from non-renewable sources—coal, petroleum, and natural gas. Solar photocatalysis is one way to potentially generate cheap renewable fuels by harnessing energy from the sun and converting it into chemical energy. Photosensitizers serve as materials that absorb and store sunlight in the form of chemical energy. The efficiency of a photosensitizer depends on its capacity to form a prolonged triplet excited state. Zinc dipyrrin complexes have the potential to be efficient sensitizers for reductive photochemistry, but their ability to form long-lived triplet excited states still needs extensive research. The overall aim of this research is to probe the role charge separation plays in the formation of triplet state in metal complexes of dipyrrin photosensitizers. The specific objectives are to synthesize and characterize zinc and boron dipyrrin complexes, analyze their photophysical properties—such as steady-state spectroscopy, low-temperature emission spectroscopy—and quantify their triplet states using time-resolved transient absorption spectroscopy.

Viscous drag reduction plays a vital role in increasing the performance of vehicles. However, there are only so many measurement techniques that can quickly and accurately measure this when compared to pressure drag measurement techniques. The current study makes use of one of the direct and robust measurement techniques that exist, called the Oil Film Interferometry (OFI) to estimate skin friction on the NASA/Virginia Tech BeVERLI (Benchmark Validation Experiment for RANS and LES Investigations) hill. This project aims to develop a detailed database of non-equilibrium, separated turbulent boundary layer flows obtained through wind tunnel experiments for CFD validation. Skin friction measurements are obtained at specific critical locations on the hill and in its close proximity. The challenges involved in obtaining skin friction data from these locations are discussed in detail. Detailed discussions on the experimental setup and data processing methodology are presented. Qualitative and quantitative results from each measurement location are discussed along with uncertainties to explain certain key flow physics. Additionally, skin friction coefficients from selected overlapping measurement locations from another experimental flow measurement technique called Laser Doppler Velocimetry (LDV) are compared with OFI, and a cross-instrument study is performed. Finally, results from well-refined RANS CFD simulations are assessed with the experimental results, and critical improvement areas are identified. / Master of Science / Drag force is a parameter that significantly contributes to the performance efficiency of any vehicle moving in a fluid. This force is categorised into two types - pressure and viscous drag- both of which need to be minimised as much as possible to contribute towards higher vehicle performance. While there are numerous measurement techniques and documentation currently available to measure pressure drag, this is not the case with the measurement of viscous drag. Skin friction measurement directly relates to the estimation of viscous drag, but accurate and quick measurement of this quantity highly challenging with countable measurement techniques currently available. Through this project, BeVERLI (Benchmark Validation Experiment for RANS and LES Investigations), a detailed documentation is developed for accurate measurement of skin friction through Oil Film Interferometry (OFI). The results obtained through this measurement is explained with a detailed experimental procedure as well as using a data processing code. The accuracy of these results are then discussed with the results from another flow measurement technique called Laser Doppler Velocimetry (LDV) and from Computational Fluid Dynamics (CFD).

The flow over sharp-edged wings is almost always separated. The control of separated flows is possible and benefits can be achieved but only in a time average sense. A new design of an actuator was designed and tested which can achieve a wide range velocity of without frequency dependence, is free of oscillating components as well as free of secondary frequencies and therefore can be scaled up easily, unlike a traditional synthetic jet. The actuator can achieve a considerable amount of jet vectoring, thus aligning the disturbance with the leading edge shear layer. Results indicate that unsteady mini-jet actuation is an effective actuation device capable of increasing the lift in the stall region of the airfoil. Moreover, pressure measurements showed that two parameters could be altered to maximize the lift. The momentum coefficient needed a minimum value to exert influence and the actuating frequency need not be at exact the natural shedding frequency to improve the lift and can be operated at harmonics of the natural shedding frequency and obtain improvements. / Master of Science

No description available.

No description available.

No description available.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第22416号 / 工博第4677号 / 新制||工||1730(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 八木 知己, 教授 清野 純史, 教授 高橋 良和 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Der vorliegende Bericht gibt einen zusammenfassenden Überblick der im Vorhaben erreichten Ergebnisse. Ziel war die Qualifikation von CFD-Methoden für Zweiphasenströmungen mit Phasenüber¬gang. Dafür werden neuartige experimentelle Daten benötigt. Diese können an der TOPFLOW-Anlage des HZDR generiert werden, da die Anlage Experimente in für die Reaktorsicher-heits¬forschung relevanten Skalen und Parametern mit innovativen Messtechniken verbindet. Die experimentellen Arbeiten umfassen Untersuchungen zu Strömungen in vertikalen Rohren mit Hilfe der ultraschnellen Röntgentomographie, zu Strömungen mit und ohne Phasenübergang in einem Testbassin sowie zur Gegenstrombegrenzung in einem Heißstrangmodell. Diese werden im vorliegenden Bericht nur kurz dargestellt, da es zu allen 3 Versuchsserien ausführliche Dokumentationen in separaten Berichten gibt. Ein wichtiges Ergebnis der Arbeiten zur CFD-Qualifizierung ist der Erstellung des Baseline-Modellkonzepts sowie die Erstellung des Baseline-Modells für polydisperse Blasenströmungen. Damit wird ein wesentlicher Beitrag zur Erhöhung der Vorhersagefähigkeit von CFD-Codes auf Basis des Zwei- oder Mehr-Fluid-Modells erreicht. Das innovative Generalized Two-Phase Flow Konzept (GENTOP) zielt hingegen auf eine Erweiterung der Einsatzmöglichkeiten der Zweiphasen-CFD. In vielen Strömungen treten unterschiedlicher Morphologien der Phasen bzw. Strömungsformen parallel in einer Strömungsdomäne auf. Außerdem gibt es Übergänge zwischen diesen Morphologien. Mit dem GENTOP-Konzept wurde erstmals ein Rahmen geschaffen der die Simulation solcher Strömungen auf konsistente Art und Weise ermöglicht. Spezielle Modellentwicklungen erfolgten mit dem Ziel einer besseren Modellierung des Phasenübergangs.

Master of Science / Department of Mechanical and Nuclear Engineering / Steven Eckels / Both cavitating and flashing flows are important phenomena in fluid flow. Cavitating flow, a common consideration in valves, orifices, and metering devices, is also a concern in loss of coolant accidents for liquid water in power plants when saturation pressures are below atmospheric pressure. Flashing flow is a common consideration for devices such as relief and expansion valves and fluid injectors as well as for loss of coolant accidents in which the coolant’s saturation pressure is above atmospheric. Of the two phenomena, flashing flow has received greater interest due to its applicability to safety concerns, though cavitating flow is perhaps of greater interest in terms of energy efficiency. It is possible for cavitating and flashing flow to actually become sonic. That is, the local velocity of a fluid can exceed the local speed of sound due to the unique properties of two-phase mixtures. When a flow becomes sonic, it is possible for the flow to accelerate and impose additional energy losses that would not otherwise occur. Models of this aspect of two-phase flow are not well developed, typically only being presented for the case of constant area ducts. In this paper two models for cavitating sonic flow are developed and described by applying the integral forms of the mass, momentum, and energy equations to a control volume of variable cross-sectional area. These models, based on the homogeneous equilibrium model (HEM) and separated flow model, are then applied to experimental data taken by the author with R-134a as the fluid of interest. Experimental data were taken with four instrumented converging-diverging nozzles of various geometries using a custom testing rig that allowed for precise control and measurement of flow parameters such as mass flow, temperature, and pressure. The resultant data from the models are then examined, focusing on the resultant velocities, Mach numbers, quality, and shear stresses.