Todos os anos, milhões de reais são gastos para cobrir os custos de garantia de caixas de direção que apresentam problemas de ruído e inconvenientes são causados nas linhas de montagem das grandes montadoras devido a este mesmo problema. Parte destas ocorrências se devem a um ruído específico chamado de Rattle Noise originado por uma operação de regulagem incorreta do bujão de ajuste. Em caixas de direção cujo “design” não permite uma medição direta da regulagem do sistema de ajuste, não existe um meio de avaliar a qualidade da operação e garantir que todas as peças estejam dentro das especificações. Baseado neste problema, este trabalho visa propor uma metodologia de avaliação da qualidade da operação de ajuste da caixa de direção de uma maneira indireta, porém com uma representação muito próxima do resultado que seria apresentado por uma medição direta, se o “design” do produto permitisse. Para tornar isso possível, foram realizados ensaios com caixas de direção em uma bancada experimental de testes aplicando-se a cremalheira cargas perpendiculares a seu eixo e momentos torsores. Foram avaliadas as respostas da amplitude do movimento angular descrito e também do deslocamento da cremalheira no sentido do seu raio observando-se a correlação que as respostas tinham com a real folga do mancal no interior do sistema de ajuste. Estes ensaios mostram que existe uma correlação muito forte entre as variáveis medidas externamente e a real flutuação do mancal no interior do sistema de ajuste e, através da regressão linear, uma equação foi definida para expressar este correlacionamento. Uma máquina foi construída para que a verificação fosse feita em 100% dos mecanismos produzidos na linha de montagem de um modelo de caixa de direção. Depois de construída a máquina, validou-se o seu funcionamento através dos estudos de R&R (Repetitividade e Reprodutibilidade), que ficaram abaixo de 20%, e da calibração da mesma comparando-se os resultados fornecidos pela máquina com as leituras realizadas de forma direta através de peças com o “design” alterado de forma a tornar isto possível, mostrando uma correlação de 99%. Hoje, o sistema está rodando perfeitamente em uma linha de montagem de um grande fabricante de sistemas de direção nacional. / Every year, millions of dollars are spent to cover the insurance costs of steering systems that have noise problems and inconveniences are caused in assembly lines of major car industry because of this same problem. Most of these occurrences are due to a specific noise called Rattle Noise caused by an incorrect operation of the adjustment screw to adjust the steering box. In cases where the direction of "design" does not allow a direct measurement of the values of the adjustment system, there is no way to assess the quality of operation and thus ensure that all parts are within specifications. Based on this problem, this work proposes a methodology to assess the quality of the adjust operation of the steering system in an indirect way, but with a very close representation of the results to be presented by a direct measurement if the "product design" had allowed. To make this possible, tests on steering systems were conducted in a test bench each exciting to rack with loads perpendicular to its axis and torsion loads. We studied responses of the magnitude of angular momentum described by the system and also the displacement of the rack along its radius by observing the correlation within the measured values to the actual bearing clearance into the system. These tests showed that there was a very strong correlation between the measured variables and externally real bearing fluctuation within the system of adjustment and using a linear regression, an equation was defined to express this correlation. A machine was built in order to verify 100% of the products in the assembly line of a vehicle steering system. Once built the machine and validated its operation through R & R studies, which has shown results below of 20%, the system was calibrated comparing the output of the machine with the readings taken directly in parts with the altered "design" to make this possible, showing a correlation of 99%. Today, the system is running perfectly on an assembly line of a major national manufacturer of steering systems.

In the late 19th and early 20th centuries famous scientists like Boltzmann, Loschmidt, Maxwell and Einstein tried, unsuccessfully, to find the link between the time-reversible equations of motion of individual molecules and irreversible thermodynamics. The solution to this puzzle was found in 1993, and the link is now known as the Fluctuation Theorem (FT). In the decade that followed theory and computer simulation tested the FT and, in 2002, an experiment indirectly demonstrated the FT.¶ This thesis describes original experiments that demonstrate the FT directly using Optical Tweezers. A related expression, known as the Kawasaki Identity, is also experimentally demonstrated. These experimental results provide a rigorous demonstration that irreversible dynamics can be obtained from a system with time-reversible dynamics.

Ce travail traite de la dynamique des flammes turbulentes prémélangées confinées et swirlées soumises à des perturbations de vitesses acoustiques. L'objectif général est d'acquérir une compréhension des mécanismes régissant la réponse de ces flammes et d'en tirer des méthodes de prévision des instabilités de combustion. Les écoulements swirlés sont d'abord examinés en termes de nombre de swirl et de nouvelles expressions sont données pour cette quantité. On traite notamment des effets de perturbations de vitesse et une expression est proposée qui tient compte des fluctuations de vitesses dans l'écoulement. Le système utilisé pour l'étude expérimentale comprend une cavité amont, un injecteur équipé d'un swirler et un tube à flamme transparent permettant la visualisation directe du mouvement de la flamme. Deux points de fonctionnement sont étudiés correspondant à des vitesses débitantes différentes. La cavité amont et le tube à flamme du brûleur peuvent être facilement changés pour étudier plusieurs configurations différentes. L'acoustique du brûleur est également analysée au moyen d'une approche de cavités couplées pour déterminer les fréquences de résonance du système en configuration non-réactive. Des expériences sont menées pour mesurer les fréquences propres du système et l'estimation du coefficient d'amortissement est réalisée à partir de la réponse du système à une modulation externe. Un critère de découplage des mode acoustiques est proposé. La dynamique de l'écoulement est examinée en termes de conversion de modes au niveau de la vrille (swirler) ou dans une grille d'aubes. Cette partie du travail, effectuée au moyen de simulations numériques montre que lorsqu'une grille ou une vrille sont soumis à une onde acoustique, le swirler donne naissance à une onde azimutale convective en plus de l'onde acoustique axiale transmise. Les deux types de swirlers, axial et radial, donnent lieu à ce mécanisme, un fait confirmé par des expériences. Il est montré que ce processus de conversion de mode a un impact important sur la dynamique de la flamme swirlée. La dynamique de la combustion est ensuite analysée en mesurant la fonction de transfert généralisée ainsi que les distributions de taux de dégagement de chaleur au cours du cycle d'oscillation. La fonction de transfert est utilisée pour déterminer la réponse de la flamme à des perturbations acoustiques se propageant dans l'écoulement en amont de la flamme. Il est aussi montré que le nombre de Strouhal est un groupe sans dimensions qui permet de caractériser la réponse de la flamme. La dynamique est également analysée au moyen d'un ensemble de diagnostics comprenant des sondes de pression, un photomultiplicateur et un vélocimètre laser Doppler. Un modèle pour la fonction de transfert linéaire de la flamme est dérivé théoriquement à partir d'une description de la flamme au moyen de l'équation pour une variable de champ G. Les mécanismes physiques de la réponse de la flamme sont identifiés : enroulement tourbillonnaire et fluctuations du nombre de swirl. L'enroulement tourbillonnaire est associé à l'onde acoustique transmise en aval du swirler et qui pénètre dans la chambre de combustion. Tandis que les fluctuations du nombre de swirl sont directement liées aux mécanismes de conversion de mode au swirler qui induit différentes vitesses pour les perturbations axiales et azimutales. L'enroulement tourbillonnaire enroule l'extrémité de la flamme tandis que les fluctuations du nombre de swirl agissent sur l'angle de la flamme. Ces deux mécanismes en compétition se combinent de manière constructive ou destructive conduisant à des gains faibles ou élevés dans la réponse de la flamme en fonction de la fréquence. Ces mécanismes sont retrouvés par simulation aux grandes échelles (LES). Enfin, une analyse d'instabilité est réalisée en combinant la fonction de transfert généralisée expérimentale et un modèle acoustique du brûleur afin de déterminer la fréquence et l'amplitude des perturbations de vitesse au cycle limite.

Children and adolescents with autism spectrum disorder (ASD) were compared to age-matched typically developing controls in the context of phase synchrony statistics acquired through magnetoencephalography. Subjects performed a “low-attention” task, pressing a button to a single, repeated auditory tone, and this was contrasted to an “oddball” task where they were required to button press only to deviant tones. The differences in task outcomes were primarily thought to relate to differences in sensory and executive attention. Regardless of task condition, ASD subjects showed greater magnitudes of synchrony between the parietal hemispheres which variably extended across frequencies. Additionally, for ASD subjects, increases in synchrony from the “low-attention” task to the “oddball” task were observed; whereas an opposite tendency occurred for controls. Furthermore, rates of fluctuation of phase differences showed more stable parietal synchrony within ASD across the hemispheric midline. Implications including the possibility of coordination as an indicator of ASD are discussed.

Children and adolescents with autism spectrum disorder (ASD) were compared to age-matched typically developing controls in the context of phase synchrony statistics acquired through magnetoencephalography. Subjects performed a “low-attention” task, pressing a button to a single, repeated auditory tone, and this was contrasted to an “oddball” task where they were required to button press only to deviant tones. The differences in task outcomes were primarily thought to relate to differences in sensory and executive attention. Regardless of task condition, ASD subjects showed greater magnitudes of synchrony between the parietal hemispheres which variably extended across frequencies. Additionally, for ASD subjects, increases in synchrony from the “low-attention” task to the “oddball” task were observed; whereas an opposite tendency occurred for controls. Furthermore, rates of fluctuation of phase differences showed more stable parietal synchrony within ASD across the hemispheric midline. Implications including the possibility of coordination as an indicator of ASD are discussed.

L'évaluation des conséquences environnementales des rejets de polluants atmosphériques est notamment possible grâce à l'utilisation de modèles numériques. Des modèles de dispersion spécifiques simulent les processus atmosphériques caractéristiques des échelles étudiées. Afin de représenter la dispersion de contaminants sur de très courtes périodes, soit de l'ordre de quelques secondes, des modèles de fluctuations de concentration (FC) sont nécessaires. La caractérisation des FC est essentielle pour des applications telles l'évaluation en air ambiant de la toxicité, inflammabilité et explosibilité de gaz ainsi que l'identification de nuisances olfactives. Ce projet traite du développement d'un modèle de FC. Les facteurs influençant les FC sont discutés, soit la période de moyennage, la stabilité atmosphérique, le temps de transport ainsi que les caractéristiques des sources émettrices. Une revue de la littérature des approches de modélisation des FC a permis de comparer les avantages et inconvénients de chacune en vue de sélectionner la plus adéquate. L'approche consistant à séparer la dispersion du panache en une composante de serpentement et de dispersion relative a été retenue. Le modèle a été développé pour qu'il s'applique à l'ensemble des régimes de turbulence atmosphérique. Une méthode basée sur une approche gaussienne a été développée afin de représenter les FC en conditions neutres et stables. Dans le cas de conditions convectives où les fluctuations verticales du vent sont asymétriques et le profil vertical de dispersion est non-gaussien, l'emploi d'un modèle lagrangien stochastique permet de simuler les trajectoires des centroïdes du panache mus par serpentement. On a également identifié des paramétrages représentant les profils verticaux des fluctuations turbulentes tridimensionnelles du vent, du taux de dissipation de l'énergie cinétique turbulente et des coefficients de dispersion. Un paramétrage a été développé afin de simuler les intensités relatives des FC en conditions neutres et stables. Les simulations réalisées avec le modèle développé montrent des comportements similaires aux résultats expérimentaux et aux autres modèles de FC. En effet, les intensités des FC, définies comme l'écart-type des FC sur la concentration moyenne, sont maximales près de la source et diminuent selon la distance dû à l'homogénéisation du panache. Latéralement par rapport à la direction du vent, les intensités simulées sont minimales dans l'axe central du panache et augmentent en s'approchant des bords, ce qui concorde avec les mesures et est dû à l'effet d'entraînement par mélange du panache avec de l'air ambiant non-contaminé. Une évaluation quantitative avec des bases de données expérimentales de FC permettrait d'évaluer les performances du modèle et d'améliorer la qualité des paramétrages. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : Fluctuations de concentration, Dispersion atmosphérique, Turbulence, Modèle lagrangien stochastique.

The effect of variable inertia flywheel (VIF) on the driving speed fluctuation, and residual vibration of high speed machine systems is investigated in this thesis. Different variable inertia flywheels are proposed to an experimental purpose roller gear cam system and a commercial super high speed paper box folding machine. The effects of time varying inertia and intermittent cam motion on the dynamic responses of different high speed cam droved mechanism systems are simulated numerically. The nonlinear time varied system models are derived by applying the Lagrange¡¦s equation and torque-equilibrium equations. The dynamic responses of these two nonlinear systems under different operating speed are simulated by employing the 4th order Runge-Kutta method. The effects of VIF parameters on the dynamic responses, i.e. the output precision, variation of motor speed, and torque, during the active and dwell periods for these two systems are studied and discussed. The difference between the dynamic responses of constant inertia and variable inertia flywheel systems are also compared. The feasibility and effectiveness of depression of driving speed and torque fluctuations by analying variable inertia flywheel has also been demonstrated.

Noise and time-dependent fluctuations are usually undesirable signals. However, they have many applications. This dissertation deals with two kinds of sensing applications of fluctuation and noise: soil bulk density assessment and bacterium sensing. The measurement of Vibration-Induced Conductivity Fluctuations (VICOF) provides information about the bulk density and other parameters of soils. Bulk density is the physical property of soils that is important to both the agriculture and construction industries. The traditional measurements of soil bulk density are often time-consuming, expensive or destructive. To determine the soil bulk density without the above drawbacks, the VICOF measurement scheme was proposed. The research of VICOF in this dissertation includes two parts: the initial phase of study and the new methods and their theory. In the initial phase of study, the simple experiments, theory, and simulations of VICOF were tested for relations between the soil bulk density, wetness, salinity, and the VICOF data. Then, new measurement arrangements and their theoretical models were proposed to improve the weaknesses of the initial approach (such as large scattering of data due to loose and heavy contacts) and to calculate the relationship between the measured signals and the electromechanical transport parameters of the soils. The bacterium sensing study in this dissertation was proposed to explore simple, practical, rapid, sensitive, specific, portable, and inexpensive ways to detect and recognize bacteria by Fluctuation-Enhanced Sensing (FES). One such potential way of bacterium sensing is to analyze their odor. The research of bacterium sensing also includes two parts: the initial phase of study and the new methods and their theory. The initial phase study was proposed to explore the possibility of detecting and identifying bacteria by sensing their odor via FES with commercial Taguchi sensors. Then the subsequently developed new methods and their theory provide a simple way to generate binary patterns with perfect reproducibility based on the spectral slopes in different frequency ranges at FES. This new type of signal processing and pattern recognition is implemented at the block diagram level using the building elements of analog circuitries and a few logic gates with total power consumption in the microWatts range.

Fluorescent lamps with ac current generate alternating lamp power and thus the light fluctuates at twice the operating frequency. To observe the light fluctuation from a fluorescent lamp, a light detector is built by using high sensitivity phototransistors in this thesis. The test results show that the light output waveform is very similar to that of the lamp power. It is also found that the light output from the whole lamp tube is not identical because the light output fluctuation becomes significant when close to the end of the lamp. When the lamp comes to the life-end, its light output is different from those produced by lamps of good conditions. In attempts to further discuss the features of light fluctuation, an electronic ballast with balanced multi-phase outputs is designed and built to reduce the variation in the light output. With a balanced multi-phase operation, the resultant light output from lamps¡¦ multi-phase currents should be a constant. An experimental 3-phase electronic ballast circuit is built to test this theoretical prediction. Experimental tests confirm that the light fluctuation can be effectively reduced by operating lamps with balanced multi-phase currents.

Nanostructures are nanometer scale structures (characteristic length less than 100 nm) such as nanowires, ultra-small junctions, etc. Since nanostructures are less stable, their characteristic volume is much smaller compared to defect sizes and their characteristic length is close to acoustical phonon wavelength. Moreover, because nanostructures include significantly fewer charge carriers than microscale structures, electronic noise in nanostructures is enhanced compared to microscale structures. Additionally, in microprocessors, due to the small gate capacitance and reduced noise margin (due to reduced supply voltage to keep the electrical field at a reasonable level), the electronic noise results in bit errors. On the other hand, the enhanced noise is useful for advanced sensing applications which are called fluctuation-enhanced sensing. In this dissertation, we first survey our earlier results about the limitation of noise posed on specific nano processors. Here, single electron logic is considered for voltage controlled logic with thermal excitations and generic shot noise is considered for current-controlled logic. Secondly, we discuss our recent results on the electronic noise in nanoscale sensors for SEnsing of Phage-Triggered Ion Cascade (SEPTIC, for instant bacterial detection) and for silicon nanowires for viral sensing. In the sensing of the phage-triggered ion cascade sensor, bacteriophage-infected bacteria release potassium ions and move randomly at the same time; therefore, electronic noise (i.e., stochastic signals) are generated. As an advanced model, the electrophoretic effect in the SEPTIC sensor is discussed. In the viral sensor, since the combination of the analyte and a specific receptor located at the surface of the silicon nanowire occurs randomly in space and time, a stochastic signal is obtained. A mathematical model for a pH silicon nanowire nanosensor is developed and the size quantization effect in the nanosensor is also discussed. The calculation results are in excellent agreement with the experimental results in the literature.