Made available in DSpace on 2016-12-12T20:15:48Z (GMT). No. of bitstreams: 1 Introducao.pdf: 113158 bytes, checksum: b51f1d71911c67cb5ee9a76e2f64e8be (MD5) Previous issue date: 2012-03-02 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / In this work we studied the dynamical behavior of driven oscillators on the parameter spaces. To characterize the behavior on parameter spaces, we use two numerical methods: one to calculate the most positive Lyapunov exponent and one to calculate the spectrum of the exponents. The dynamical systems studied in this work are: three complex driven oscillators and a four-dimensional Chua circuit. With the help of the parameter spaces, it was possible to observe various dynamical behaviors of the systems. We also use others techniques, as bifurcation diagrams and trajectories on phase space (attractors) to characterize the systems dynamics. However, between the two numerical methods used, the best one was that calculate the Lyapunov exponent spectrum, because it is possible to construct the parameter spaces for the .rst and the second largest Lyapunov exponent. / Nesta dissertação estudamos a dinâmica dos osciladores forçados, via estudo do espaço (plano) de parâmetros. Para caracterizar o comportamento no espaço de parâmetros, usamos dois métodos: um que calcula somente o maior expoente de Lyapunov e o outro que calcula o espectro de Lyapunov. Os sistemas estudados nessa dissertação são: três osciladores complexos forçados e um circuito de Chua forçado no espaço quadridimensional. Com a construção dos espaços de parâmetros, foi possível observar diversos comportamentos dinâmicos. Usamos também, outras técnicas conhecidas, como a construção de diagramas de bifurcação e trajetórias no espaço de fase, para caracterizar a dinâmica dos sistemas. Porém, o método que apresentou mais recursos para caracterizar a dinâmica de um sistema, foi o que calcula o espectro de Lyapunov, pois, a partir daí, é possível construir os espaços de parâmetros para o primeiro e para o segundo maiores expoentes de Lyapunov.

Nas últimas décadas uma considerável parcela das pesquisas nas áreas de Física e Matemática tem sido dedicada ao estudo de fenômenos não lineares. Uma possível explicação para isso foi o rápido desenvolvimento de sistemas computacionais, tanto em nível de hardware quanta em nível de software, algoritmos e técnicas de programação que propiciaram ao homem maiores facilidades no tratamento de sistemas não lineares, o que levou a um maior grau de entendimento de sua complexidade. Geralmente, aos sistemas não lineares esta associada uma geometria irregular, onde comum o aparecimento de regimes caóticos, com um conjunto atrator de órbitas cuja dimensão não é um inteiro positivo, mas sim um número real positivo. Por esta razão, tais atratores, são denominados estranhos e ditos possuírem uma geometria fractal. É possível, através de métodos cuidadosamente desenvolvidos, estimar-se as dimensões associadas à dinâmica de séries temporais. Uma das séries de maior dificuldade de análise através do computador, e de particular interesse na medicina, são as séries de sons pulmonares humanos. Desde quando o estetoscópio foi inventado até os dias de hoje não há uma ferramenta plenamente confiável para a análise destas séries. Recentemente, temos trabalhado com estas séries e verificamos que há uma geometria fractal. Esta tese propõe a utilização da análise não-linear para identificação de padrões sonoros. Além da geometria fractal, a análise por wavelets tem sido utilizada no estudo de sinais complexos, sobretudo naqueles que apresentam estruturas fractais. O conjunto de filtros construído através da translação, expansão ou compressão de uma função wavelet mãe tem uma estrutura auto-similar, mostrando-se particularmente apropriado para a verificação da auto similaridade dos sons. A técnica da estimativa dos expoentes de Lyapunov dependente do tempo, a qual e desenvolvida na tese, tem se mostrado bastante adequada para identificação de padrões sonoros de origem pulmonar. / It has been observed that in the last decades, considerable amount of the research in the areas of Physics and Mathematics have been dedicated to the study of nonlinear phenomena. A possible explanation for this fact is the fast development of computational systems occurring in the level of the hardware as in computer languages, algorithms and programming techniques. These developments propitiated to the researchers a broader contact with nonlinear systems, which led to a better understanding of their complexity. In general, for nonlinear systems an irregular geometry is associated, where the appearance of chaotic regimes has an associated attractor set of orbits whose dimension is not a positive integer number, but a real one. Such attractors are called strange and said to possess fractal geometry. It is possible, through carefully developed methods, to estimate the dimension associated to the dynamics of time series. One of the series with high difficulty to be analyzed through a computer and of particular interest in medicine, is the time series generated out of human pulmonary sounds. Since the creation of the stethoscope, there is not yet a fully trustworthy tool for the lung sound analysis. Recently, we have studied these series and verified that they have a fractal geometry nature. The purpose of this thesis is to investigate non-linear analysis as a tool for pattern recognition in lung sounds. In addition to fractal geometry, the wavelet analysis has been used in the study of complex signs, in particular for those presenting a fractal structure. The set of filters constructed through the translation, expansion or compression of a function wavelet mother has an auto-similar structure, being particularly useful for the verification of self similarity of pulmonary sounds. The largest time dependent Lyapunov exponent estimation technique that has been proposed in this thesis has shown a high degree of confidence for the identification of lung sound patterns.

Cette thèse concerne les propriétés dynamiques des endomorphismes holomorphes du plan projectif complexe. La première partie introduit et minore les dimensions directionnelles du courant de Green. Nos résultats mènent une analyse multifractale des tranches de ce courant par des coordonnées locales, relativement aux mesures ergodiques dilatantes. Une première application montre que, relativement à toute mesure ergodique de grande entropie, tout courant positif fermé possède une dimension directionnelle strictement plus grande que deux, ce qui répond à une question de de Thélin-Vigny. Comme deuxième application, nous décrivons les dimensions directionnelles du courant de Green des endomorphismes semi-extrémaux de Dujardin, c'est à dire ceux dont la mesure d'équilibre est absolument continue par rapport à la mesure trace du courant de Green. Dans la deuxième partie, nous majorons les dimensions directionnelles du courant de Green en utilisant des techniques de Théorie du pluripotentiel. En combinant ces résultats à ceux de la première partie, nous montrons une propriété de séparation des dimensions directionnelles du courant de Green relativement à la mesure d'équilibre. Dans la dernière partie, nous étudions la régularité des tranches du courant de Green dans deux situations semi-extrémales. Nous montrons que la dérivée de Radon-Nikodym des tranches stables est bornée presque partout. Cette propriété, proche de l'absolue continuité par rapport à la mesure de Lebesgue, apporte une précision à nos résultats précédents. Les techniques utilisées ont également permis d'obtenir une nouvelle majoration de la dimension locale des mesures ergodiques dilatantes. Cette majoration nous rapproche de la conjecture de Binder-DeMarco concernant la dimension de la mesure d'équilibre. / This thesis studies the dynamical properties of holomorphic endomorphisms of the complex projective plane. The first part introduces and proves lower bounds for the directional dimensions of the Green current. We give there a multifractal analysis of the slices of that current by local coordinates, with respect to dilating ergodic measures. A first application shows that, with respect to every measure of large entropy, every closed positive current has a directional dimension strictly larger than two, which answers a question by de Thélin and Vigny. A second application describes the directional dimensions of the Green current of Dujardin's semi-extremal endomorphisms, which have an equilibrium measure absolutely continuous with respect to the trace measure of the Green current. The second part provides upper bounds for the directional dimensions of the Green current by using Pluripotential Theory. Combining these results with those of the first part, we obtain a separation property of the directional dimensions of the Green current with respect to the equilibrium measure. In the last part, we focus on the regularity of one-dimensional slices of the Green current in two semi-extremal situations. We show that the Radon-Nikodym derivative of the stable slices is bounded almost everywhere. This property is close to the absolute continuity with respect to the Lebesgue measure, and specifies our previous results. Our methods also allow to prove an upper bound for the local dimension of dilating ergodic measures, which is a new step towards Binder-DeMarco's conjecture concerning the dimension of the equilibrium measure.

We consider diffeomorphisms of a compact Riemann Surface. A development of Oseledec's Multiplicative Ergodic Theorem is given, along with a development of measure theoretic entropy and dimension. The main result, due to L.S. Young, is that for certain diffeomorphisms of a surface, there is a beautiful relationship between these three concepts; namely that the entropy equals dimension times expansion.

No description available.

Time varying flow structures are involved in a large percentage of fluid flows although there is still much unknown regarding their behavior. With the development of high spatiotemporal resolution measurement systems it is becoming more feasible to measure these complex flow structures, which in turn will lead to a better understanding of their impact. One method that has been developed for studying these flow structures is finite time Lyapunov exponents (FTLEs). These exponents can reveal regions in the fluid, referred to as Lagragnian coherent structures (LCSs), where fluid elements diverge or attract. Better knowledge of how these time varying structures behave can greatly impact a wide range of applications, from aircraft design and performance, to an improved understanding of mixing and transport in the human body. This work provides the development of new methodologies for measuring and studying three-dimensional time varying structures. Provided herein is a method to improve replacement of erroneous measurements in particle image velocimetry data, which leads to increased accuracy in the data. Also, a method for directly measuring the finite time Lyapunov exponents from particle images is developed, as well as an experimental demonstration in a three-dimensional flow field. This method takes advantage of the information inherently contained in these images to improve accuracy and reduce computational requirements. Lastly, this work provides an in depth look at the flow field for developing wall jets across a wide range of Reynolds numbers investigating the mechanisms that contribute to their development. / Ph. D.

Many chemical reactions can be described as the crossing of an energetic barrier. This process is mediated by an invariant object in phase space. One can construct a normally hyperbolic invariant manifold (NHIM) of the reactive dynamical system which is an invariant sphere that can be considered as the geometric representation of the transition state itself. The NHIM has invariant cylinders (reaction channels) attached to it. This invariant geometric structure survives as long as the invariant sphere is normally hyperbolic. We applied this theory to the hydrogen exchange reaction in three degrees of freedom in order to figure out the reason of the transition state theory (TST) failure. Energies high above the reaction threshold, the dynamics within the transition state becomes partially chaotic. We have found that the invariant sphere first ceases to be normally hyperbolic at fairly low energies. Surprisingly normal hyperbolicity is then restored and the invariant sphere remains normally hyperbolic even at very high energies. This observation shows two different energy values for the breakdown of the TST and the breakdown of the NHIM. This leads to seek another phase space object that is related to the breakdown of the TST. Using theory of the dividing surface including reactive islands (RIs), we can investigate such an object. We found out that the first nonreactive trajectory has been found at the same energy values for both collinear and full systems, and coincides with the first bifurcation of periodic orbit dividing surface (PODS) at the collinear configuration. The bifurcation creates the unstable periodic orbit (UPO). Indeed, the new PODS (UPO) is the reason for the TST failure. The manifolds (stable and centre-stable) of the UPO clarify these expectations by intersecting the dividing surface at the boundary of the reactive island (on the collinear and the three (full) systems, respectively).

Friction compensation is challenging but imperative for control engineers. For high-performance engineering systems, a friction-model-based controller is typically required to accommodate the nonlinearities arisen from the friction model employed. It is well known that the parameters of the friction model used in the friction compensation are nearly impossible to be accurately identified. Therefore, the objective of this research is to study the effect of these parametric uncertainties on the stability of a set-point position control system. With the above goal in mind, a variety of aspects are investigated in this work. Firstly, several common friction features and friction models are surveyed to provide background knowledge which helps select the friction model with the highest accuracy for our study. Secondly, an experimental setup is proposed and fabricated to validate the levels of accuracy given by the candidate friction models. The comparisons between the numerical and experimental results confirm that the LuGre friction model is the best approximation of the observed friction behaviours among all models selected. Moreover, a series of profound discussions addressing the relation between the candidate models’ structures and their numerical friction feature predictions are provided and followed by a summary table that recapitulates the properties of the candidate friction models. Last but not least, the state space models of the proposed setup formulated by the improved version of the LuGre model and the two controllers of interest, namely input-output linearization controller and nominal characteristic trajectory following (NCTF) controller, are derived for the stability analysis under the parametric uncertainties. Two parameters in the friction model used, σ_0 and σ_1, are perturbed for the stability analysis in which the results applying the concept of Lyapunov exponents (LEs) clearly illustrate the significant effect of the varying σ_0 and σ_1 values on the system stability. The effect of parametric uncertainties can depend quite significantly on the incorporated controller, and the stability results obtained here are applicable to the design and analysis of other systems that are inherently similar to our setup. The stability analysis conducted is this work is recommended for other control systems to avoid unwanted qualitative behaviours under parametric perturbations. / October 2016

Neste trabalho nós estudamos os expoentes de Lyapunov de aplicações f : Td → Td homotópicas a uma aplicação Anosov linear e a continuidade absoluta de folheações. Nós mostramos para algumas classes de homotopia de aplicações que a soma dos expoentes de Lyapunov está limitado pela soma dos expoentes de Lyapunov da aplicação Anosov linear. Além disso, admitindo uma propriedade conhecida como densidade uniformemente limitada (UBD) nas folheações, mostramos uma igualdade entre a soma dos expoentes de Lyapunov de f e do Anosov linear. Também construímos um conjunto C1 aberto de difeomorfismos parcialmente hiperbólicos do toro T4, preservando volume, com folheação central bidimensional não compacta e não absolutamente contínua. Ainda construímos um exemplo parcialmente hiperbólico com folhas centrais bidimensionais, não compactas onde a desintegração do volume ao longo da folheação central não é nem Lebesgue nem atômica. / In this work we study the Lyapunov exponents of maps f : Td → Td homotopic to a linear Anosov map. We proof for some homotopic classes of maps which the sum of Lyapunov exponents is bounded by the sum of the Lyapunov exponents of the linear Anosov map. Moreover, by assuming a property known as uniformly bounded density (UBD) in the foliations, we show an equality between the sum of the Lyapunov exponents of f and the linear Anosov. We also construct an C1 open set of volume preserving partially hyperbolic diffeomorphisms with non compact two dimensional center foliation and non absolutely continuous. We still build an example of partially hyperbolic diffeomorphism with non compact bidimensional center leaves where the disintegration of volume along the center foliation is neither Lebesgue nor atomic.

Nas últimas décadas, diversos sistemas de comunicação baseados em caos foram propostos. Dentre eles, vários utilizam uma função para codificar uma mensagem em um sinal caótico, que é caracterizado como um sinal de banda larga. Dado que o canal de transmissão é limitado em banda por natureza, é necessário determinar e controlar o espectro do sinal caótico transmitido por esse sistema. Nesse sentido, um sistema de comunicação em banda limitada, baseado em sincronismo caótico, foi proposto recentemente utilizando-se filtros digitais para controlar a largura de banda dos sinais transmitidos. Esses filtros, inseridos no sistema de comunicação, modificam o sistema original gerador do sinal caótico, tornando-se necessário analisar como essa inserção afeta o sincronismo caótico. Nessa tese, apresenta-se uma análise desse sistema de comunicação digital de tempo discreto, baseado em sincronismo caótico, considerando-se um canal com ruído aditivo branco gaussiano. As condições necessárias para a sincronização desse sistema são obtidas analiticamente, por meio de um teorema, para um mapa gerador de caos qualquer. O desempenho desse sistema é avaliado em termos da taxa de erro de bit, e, para melhorar seu desempenho, propõe-se filtrar o ruído fora da banda do sinal na entrada do receptor. Apesar das condições de sincronismo terem sido determinadas, a inserção dos filtros também pode modificar a natureza caótica dos sinais, e não há garantia que os sinais transmitidos sejam caóticos. Para analisar a natureza caótica dos sinais transmitidos pelo sistema de comunicação, o maior expoente de Lyapunov é obtido numericamente em função dos coeficientes dos filtros, dos parâmetros do mapa e da função de codificação da mensagem. / In recent decades, several chaos-based communication systems have been proposed. Many of them use a function to encode a message into a chaotic signal, which is characterized as wideband. Since every transmission channel is bandlimited in nature, it is necessary to determine and to control the spectrum of the chaotic signal transmitted by this system. This way, a bandlimited chaos-based communication system was recently proposed using digital filters and chaotic synchronization. These filters, inserted in the communication system, modify the original chaotic generator system, becoming necessary to study how their insertion affect chaotic synchronization. In this work, we present an analysis of this discrete-time chaos-based digital communication system considering an additive white Gaussian noise channel. The synchronization conditions of this system is analytically obtained, through a theorem, for a generic chaos generator map. The system performance is evaluated in terms of bit error rate, and, to obtain a performance improvement, it is also proposed to filter the out-of-band noise in the receiver. Although the conditions for chaotic synchronization have been determined, the filters insertion can also modify the chaotic nature of the signals, and there is no guarantee that the transmitted signals remain chaotic. To analyze the chaotic nature of the communication system transmitted signals, the largest Lyapunov exponent is numerically accessed as a function of the filters coefficients, the parameters of the map and the message coding function.