O crescimento da demanda de energia e das interligações entre os Sistemas Elétricos de Potência, assim como a necessidade de prestação de melhores serviços exigem que as técnicas de análise destes sejam cada vez mais rápidas e refinadas. Dentre estas técnicas, as análises de estabilidade transitória através de métodos energéticos são estudadas e o desenvolvimento de técnicas computacionais que permitam a análise em tempo real da estabilidade transitória é o principal objetivo deste trabalho. Os métodos energéticos são métodos diretos adequados à aplicações em tempo real, pois obtém informações a respeito da estabilidade diretamente das equações diferenciais do sistema, sem a necessidade do conhecimento das soluções das equações diferenciais. A aplicação de métodos energéticos aos estudos de estabilidade transitória evoluiu significativamente nos últimos anos. Esta evolução culminou com o desenvolvimento do método BCU. Este método está fundamentado no conceito de ponto de equilíbrio de controle e portanto, garante apenas a existência ou a estabilidade do primeiro \"swing\". Embora este seja um método eficiente, existe uma grande quantidade de casos que são estáveis no primeiro \"swing\", mas o sistema perde a estabilidade em \"swings\" subseqüentes. Para resolver este problema, desenvolveu-se uma técnica de estimativa do parâmetro amortecimento para o caso de uma máquina versus barramento infinito. Estima-se o amortecimento necessário que a máquina deverá possuir tal que a estabilidade nos \"swings\" subseqüentes seja garantida. Esta metodologia é utilizada em conjunto com o BCU garantindo a satisfação da condição de transversalidade exigida no desenvolvimento teórico deste método. Maiores investigações serão necessárias para estimar os amortecimentos em sistemas multimáquinas. / The increase of load demand and interconnection between power systems as well as the necessity of offering better customer services require fast and refined techniques to analyze these systems. In particular, the transient stability analysis is studied and the development of computational techniques suitable to real-time transient stability analysis is the main aim of this work. The energetic methods are direct methods suitable to real-time applications because they obtain qualitative information about the system without the knowledge of the differential equation solutions. The application of energetic methods to the transient stability analysis developed significantly in the last years. This evolution culminated with the development of the BCU method. This method is based on the concept of controlling equilibrium point, therefore it only guarantees the existence of the first swing. In spite of the efficiency of this method, there exists a high number of cases where the BCU predicts stability to the first swing but the system becomes unstable or loses synchronism in future swings. In order to solve this problem, a methodology to estimate the damping coefficient was developed to the machine versus infinite bus case. The damping coefficient that must be applied to the machine to guarantee the stability in future swings is estimated. This methodology is used together with the BCU method guaranteeing the satisfying of the transversality condition required by the BCU theory. Further investigations must be carried out to extend the methodology to multimachine systems.

Le thème central de la thèse est l'interférométrie atomique et la réduction du bruit de phase lié aux vibrations de l'environnement. Les interféromètres atomiques sont des instruments pouvant permettre des mesures fondamentales de grande précision et cette précision est fortement liée à la vitesse des atomes. L'interféromètre qui était utilisé à Toulouse travaillait alors avec des atomes de lithium aux énergies thermiques et le projet était de pouvoir utiliser des atomes fortement ralentis. Les vibrations du sol (bruit sismique) deviennent alors un inconvénient majeur et il est indispensable de s'en affranchir le plus possible. La première partie du travail fut de prévoir une isolation vis à vis du bruit sismique. Un premier filtre est réalisé en plaçant l'ensemble de la manipulation sur un pendule suspendu par 3 fils. Celui ci atténue les vibrations de fréquence supérieure à sa fréquence propre mais amplifie celles de fréquences voisines de celle-ci. Il faut donc un pendule asservi et cela implique un sismomètre sensible pouvant fonctionner sous ultravide. Nous avons développé une stabilisation des mouvements horizontaux en nous appuyant tout d'abord sur des sismomètres simples mais peu sensibles, puis nous avons cherché à améliorer les performances en réalisant un capteur de déplacement basé sur un interféromètre de Michelson à coins de cube. Nous avons suivi en cela les travaux de l'équipe de M.Zumberge qui utilise une détection de deux signaux en quadrature ce qui permet une mesure de déplacement avec une sensibilité meilleure que 4.10-13 m/vHz à 1 Hz et entrepris d'adapter cette technologie à un fonctionnement sous ultravide. Mais les difficultés rencontrées et l'abandon de l'interféromètre tel qu'il était pour en developper un nouveau ne nous ont pas permis d'atteindre complètement notre but et de pouvoir tester le sismomètre in-situ.Cependant, la mise au point et l'optimisation du sismomètre nous a amené à nous pencher sur la théorie des pendules à lame élastique, lesquels sont largement utilisés dans ce genre de capteurs. Il nous est apparu que cette théorie était très incomplète et nous avons entrepris une étude plus systématique de tels pendules ce qui a donné lieu à une publication et fait l'objet de la seconde partie de la thèse. [...] / The main theme of my thesis is atomic interferometry and in particular the reduction of the phase noise induced by vibrations. Atomic interferometers are good devices to achieve accurate and fundamental measuring. The sensibility of these devices is related to the flying time of the atoms inside the apparatus. At Toulouse, our interferometer worked with atoms at thermal velocity and to increase the sensibility we wanted slower atoms. However, this will at the same time increase the effect of vibrations, witch result in a larger phase noise and a jamming of the fringes. In order to reduce this effect, I've put the core of the interferometer on a 3 wires pendulum. A pendulum attenuates the vibrations of frequencies much higher than its resonant frequencies but amplifies those with frequencies close to its resonances. To avoid this phenomenom, we have to enslave the pendulum on the signal given by seismometers. With a first realisation, I was able to stabilize 2 horizontal movements with 2 low sensibility seismometers. To increase the performances, I needed high sensibility seismometer and the possibility to operate under ultra vacuum. I've made a deplacement sensor based on the Michelson interferometer with cube corners, following the works of Zumberge's team. By choosing cleverly the polarisation of the laser beam, we can detect 2 signals in quadrature and the sensibility achieved is better than 4x10-13 m/vHz at 1 Hz. The next step was to migrate this seismometer in ultra vacuum but the retirement of the interferometer using slow down lithium atoms at the benefit of an atomic fountain of rubidium stopped this project. However, this work on the seismometer led me to think about elastic blade pendulums, widely used in such sensors. I've complete the theory, showing the presence of 2 resonance frequencies and, as a test, I've build a such pendulum, for witch I've measured the caracteristics with some position and velocity sensors I've developped for this purpose. I was able to measure precisely the damping of the oscillations of the pendulum and study more precisely the different origins of the damping. Two of them have given some additionnal work : a)the coupling with the resonances of the frame witch support the pendulum may have an effect on the quality factor of the pendulum. [...]

Electrical substations perform a key role in electrical transmission and distribution; the ability for a substation to remain functional during and after a seismic event contributes significantly to the resilience of the clients supplied. Many legacy components currently installed in the main grid substations were designed with minimal consideration of lateral loads and are not qualified per IEEE693. One of the more critical high-voltage substation components that are vulnerable to earthquake damage is the 500kV freestanding current transformer (CT). The CT is particularly vulnerable due to the slenderness and mass distribution of the component. Current transformers are typically constructed from a combination of aluminum and brittle porcelain. Two novel retrofit measures were investigated utilizing base rocking and supplemental damping to reduce the seismic amplification in the CT while also potentially providing post-earthquake self-centering capability. The retrofit measures utilize both shift in system frequency and energy dissipation through supplemental damping to reduce seismic demands on the CT. The purpose of the research was to conceptually develop, detail design, analyze and experimentally validate the retrofit measures. A desired feature of the retrofit measures was for minimal or no residual displacement following the seismic event, which was implemented in the retrofit through a preloaded centering mechanism. Based on the analyses and experiments, the proposed retrofit measures exhibited significantly decreased demands on the CT and true self-centering.

Afin de réduire les vibrations d'une structure, on utilise des éléments piézoélectriques connectés à des circuits électriques passifs. L'objectif est de se rapprocher de l'efficacité du contrôle actif sans en supporter la complexité et la consommation. On considère d'abord l'association d'une résistance (qui a un effet similaire à un amortissement visqueux) et éventuellement d'une inductance (permettant de réaliser un oscillateur accordé) aux éléments piézoélectriques. Ces systèmes ont des propriétés intéressantes, mais sont peu efficaces à moins d'un accord très précis de l'inductance. Afin d'obtenir des performances élevées sans accord précis, on étudie un circuit à commutation, qui se ferme et s'ouvre à des instants bien précis. L'effet de la charge, qui freine la structure, s'apparente à un frottement sec. En synchronisant les commutations sur les vibrations, le système est auto-adaptatif et peut être auto-alimenté. Les fortes non-linéarités entraînent une excitation haute fréquence de la structure qui peut rendre la synchronisation problématique. Deux modèles électromécaniques (analytique et éléments finis) réduits sont proposés, permettant de décrire la dynamique du système complet de manière précise et de mettre en valeur le couplage entre un mode de vibration et le circuit électrique. Ce couplage est déterminant pour la réduction de vibrations. Une étude de l'influence de divers paramètres permet d'optimiser les éléments piézoélectriques, les circuits, et les instants de commutation. Ces résultats sont vérifiés expérimentalement. On constate un bon accord avec la théorie ; la difficulté de synchroniser correctement la commutation est aussi constatée. / In order to reduce the vibrations of a mechanical structure, one can use piezoelectric elements connected to passive electrical circuits. The goal is to achieve the same efficiency as active vibration control without the associated complexity and energy consumption. First the use of a resistor (with an effect similar to viscous damping) and eventually of an inductor (allowing the creation of a tuned resonator) for the circuit is considered. These systems have interesting properties, but are not very efficient, except in the case of a finely tuned inductor. In order to obtain good performance without requiring a precise tuning, a switching circuit is considered. The switching process is synchronized on the vibrations, and the effect of the free electric charge (similar to a dry damping) reduces the vibrations. This system is self-adaptive and can be self-powered. However, the strong non-linearities create a high frequency excitation which may disturb the switch timing. Two different reduced electromechanical models (analytical and finite element) are proposed, allowing a description of the whole system dynamics with accuracy and to emphasize the coupling between one vibration mode and the circuit. This coupling is found to be decisive for the performance in vibration reduction. A study of the influence of various parameters allow the optimisation of the piezoelectric elements, electric circuits and switch timing. These results are experimentally tested and a good agreement with the predictions is obtained ; the difficulty of switch timing is also noticed.

The purpose of this research is to develop MEMS based acoustic emission sensors for structural health monitoring. Acoustic emission (AE) is a well-established nondestructive testing technique that is typically used to monitor for fatigue cracks in structures, leaks in pressurized systems, damages in composite materials or impacts. This technology can offer a precise evaluation of structural conditions and allow identification of imminent failures or minor failures that can be addressed by planned maintenances routines. AE causes a burst of ultrasonic energy that is measured as high frequency surface vibrations (30 kHz to 1 MHz) generated by transient elastic waves that are typically emitted from growing cracks at the interior of the structure. The AE sensor marketplace is currently dominated by bulky and expensive piezoelectric transducers that are wired to massive multichannel data acquisition systems. These systems are complex to operate with the need of signal conditioning units and near proximity pre-amplifiers for each sensor that demands a fairly complicated wiring requirements. Furthermore, due to the high prices of conventional AE sensors and associated instrumentation, and the current requirements in sensor volumes for smart transportation infrastructure, it is undeniable that new AE technology is required for affordable structural health monitoring. The new AE technology must deliver comparable performance at one or two orders of magnitude lower cost, size and weight. MEMS acoustic emission (AE) sensors technology has the potential to resolve several of these traditional sensor’s shortcomings with the advantage of possible integration of on-chip preamplifier while allowing substantially cost reduction due to the batch processing nature of MEMS technology. This study will focus on filling some of the major existing gaps between current developments in MEMS acoustic emission sensors and commercial piezoelectric sensors, such as sensor size, signal-to-noise ratio (SNR), cost and the possibility to conform to sharply curved surfaces. Basically, it is proposed to develop a new class of micro-machined AE sensors or sensor arrays through strategic design of capacitive and piezoelectric MEMS sensors, which will focus on optimizing the following performance aspects: Creating geometric designs to manipulate the sensor resonant frequency and to optimize Q factor under atmospheric pressure and ambient environment. Developing a strategic selection of materials according to its acoustic impedance as insulator, structure and backing material. Developing strategies to improve the signal to noise ratio SNR with and without integrated amplification/signal processing. Performing a comparison between MEMS and commercial piezoelectric sensors.

Abstract not available

[Formulae and special characters can only be approximated here. Please see the pdf version of the Abstract for an accurate reproduction.] Since the damping functions employed by most of the low-Reynolds-number models are related to the non-dimensional distance y+[ special character ??? near-wall non-dimensional distance in y direction], which is based on local wall shear stress, these models become invalid for separated flows, because the wall shear stress is zero at the reattachment point. In addition, the pressure-velocity correlation term is neglected in most of these models, although this term is shown in this thesis to be important in the near-wall region for simple flows and large pressure gradient flows. In this thesis, two main efforts are made to improve the k ??? [special character - dissipation rate] model. First, based on Myong and Kasagi???s (1990) low-Reynolds-number model (hereafter referred to as MK model), a more general damping function [special character - turbulent viscosity damping function in LRN turbulent model] is postulated which only depends on the Reynolds numbers [formula ??? near-wall turbulence Reynolds number]. Second, a form for the pressure-velocity correlation term is postulated based on the Poisson equation for pressure fluctuations. This modified model predicts the turbulent flow over a flat plate very well. It is found that the inclusion of the pressure-velocity correlation term leads to significant improvement of the prediction of near-wall turbulence kinetic energy. When the model is applied to turbulent flow over a backward-facing step, it produces better predictions than the traditional k ??? [special character - dissipation rate] model, FLUENT???s two-layer model and the MK model. Again, the pressure-velocity correlation term improves the turbulence kinetic energy prediction in the separated region over that of other models investigated here. The studies of numerical methods concerning computational domain size and grid spacing reveal that a very large domain size is required for accurate flat plate flow computation. They also show that a fine grid distribution in the near-wall region upstream of the step is necessary for acceptable flow prediction accuracy in the downstream separated region.

The modern game of tennis has changed in recent years as a result of lightweight, stiffer racquets. The evolution of the tennis racquet, with respect to both design and materials, has increased the speed of the game but also the levels of stress placed on the player's bodies. Many believe that injuries such as lateral epicondylitis (tennis elbow) are caused and aggravated by the absorption of racquet energy by the player, in the form of shock and vibration. This thesis presents an experimental investigation into the absorption of racquet vibration to the player's hand and forearm. Quantification of the tennis grip has been achieved in this research using different experimental techniques to analyse different aspects of the tennis grip. Grip pressure distribution profiles during impact have been established using both pressure sensitive film and real-time data acquisition methods. Quantification of grip tightness during impact, together with gripping times, has also been quantified using a strain gauge cantilever system manufactured specifically for this research. The experimental data acquired in this research has provided the base for grip pressure distribution profiles to be established for three stroke types (e.g. Forehand, service and the problematic backhand). The profiles depict the distribution of pressure in the tennis grip in relation to the ball impact, in the time domain. Based on these grip profiles, the research hypothesises hand movements in an attempt to establish muscle contractions (and moreover locations of vibration absorption) specific to stroke types. The research investigates the absorption of racquet vibrations by the player's hand in the time domain. Filtering of accelerometer data allows for the isolation of specific frequencies of interest (i.e. below 200Hz). Logarithmic decrement of racquet vibration has been calculated and related to the grip pressure distributions in the time domain, and the relationship between grip pressure and vibration damping has been modelled. The correlation between grip pressure and the logarithmic decrement has been show to be significant (p less than 0.005) and non-linear. The relationship between the tennis grip and the damping of racquet vibrations has been found to be dependant on both grip pressure and the proximity of grip pressure application in proximity to the handle node. Grip pressure applied to the racquet close to the handle node has a greater damping effect than a similar pressure further away. In addition to these key research findings, the effectiveness of a piezoelectric racquet damping system is also investigated. A comprehensive modal analysis of two tennis racquets is given with further ball impact tests. The ball impact tests showed that the damping system has a 28% difference in racquet vibrations during freely suspended grip conditions. However, under hand-held grip conditions the inclusion of grip damping into the system provides a much greater damping entity (880% greater). Therefore, the effect of the piezoelectric system was deemed to be negligible.

A global, multi-level, wave-mean flow model based on an approximate version of the primitive equations is developed to investigate the development of a baroclinic wave field initially confined to a single zonal wavenumber. The effects of physical processes (surface drag and thermal damping) and internal diffusion on the evolution have been examined. The nature of the mean flow adjustment by the nonlinear baroclinic waves is also studied. For a simulation with a relatively strong internal diffusion it is found that a single life cycle characterized by baroclinic growth and barotropic decay is obtained (as in Simmons and Hoskins, 1978), whereas with weaker diffusion the wave undergoes secondary life cycles before a nearly wave-free state is reached (as in Barnes and Young, 1991). In an experiment with weak 4th order diffusion secondary life cycles occur with little net decay. Relatively strong barotropic growth follows the initial life cycle. In experiments with surface drag (Rayleigh friction) and thermal damping (Newtonian cooling), repeated life cycles of baroclinic growth and barotropic decay can be obtained. It is found that in the complete absence of surface drag, the flow evolves to a nearly wave-free state after one secondary cycle. This demonstrates that surface drag plays an important role in nonlinear baroclinic instability. With relatively strong surface drag multiple life cycle behavior is found for sufficiently strong thermal damping. Such a behavior strengthens for very strong thermal damping. A steady wave state in which the wave amplitude equilibrates at an essentially constant level has only been obtained with very strong "potential vorticity damping". Both the "barotropic governor" process (James and Gray, 1986) and the baroclinic adjusment process are responsible for major parts of the stabilization of the mean flow in simulations with and without surface drag and thermal damping. However, the "barotropic governor" process dominates the flow evolution in the model simulations without surface drag and thermal damping. The "barotropic governor" modifies the meridional gradient of zonal mean potential vorticity, which influences the baroclinic adjustment. / Graduation date: 1992

The aerospace industry is utilizing low cost miniature inertial measurement units (IMUs) that employ Micro Electro-Mechanical Systems (MEMS) technology in an effort to reduce size, weight, and cost of systems. A drawback of these MEMS devices is they are sensitive to vibration, shock and acoustic environments, which limits the usefulness of such devices in the severe environments imposed by many aerospace applications. In an effort to reduce the vibration, shock and acoustic environments experienced by these MEMS devices, the desire to develop passive damping treatments to structural components used to mount these devices. The damping treatments can be applied at the printed circuit board (PCB) level, the component level, the component interface, or at the airframe level. The purpose is to reduce the overall environment and improve the usefulness and performance of the MEMS based sensors. The primary technique to introduce damping into metallic parts and PCBs is to provide a viscoelastic coating or layer. The ability to analyze structures with this configuration requires a thorough understanding of the dynamic properties. Hooke’s law of elasticity is one of the most fundamental relationships governing dynamic properties. Metals typically have a low damping coefficient, and Hooke’s law of elasticity represents a linear relationship between the ratio of stress and strain, known as the modulus of elasticity. But for viscous materials the modulus of elasticity becomes a complex value since the stress and strain are not in phase. The complex modulus of elasticity is a complex function of frequency. The complex modulus can be established via frequency response function measurements of compliance, mobility, and accelerance, and the dimensions of the block of material under test. At low frequencies (less than resonance of the block) the results are relatively straight forward, but at higher frequencies where resonances of the block occur the inertial forces begin to influence the FRF results. This thesis effort establishes techniques for measuring the complex moduli associated with viscoelastic materials, and presents methods and results from modulus tests conducted for this thesis.