Devant la forte demande mondiale en énergie électrique, les alternateurs à diodes tournantes constituent une solution largement répandue dans les installations de génération d’électricité de fortes puissances (hydraulique, fossile et éolien) ainsi que dans les applications îlotées sous forme de groupes électrogènes ou de petits alternateurs intégrés dans les systèmes embarqués. La renommée de ce type d’alternateur s’est construite sur leurs robustes constitutions mécanique et électrique et sur leur parfaite adaptabilité au type de charge alimentée. Néanmoins, l’utilisation de ces machines dans des conditions de fonctionnement très contraignantes, que cela soit à cause des fortes puissances demandées par les applications industrielles ou des contraintes environnementales dans lesquelles travaillent les alternateurs isolés, engendre une recrudescence de défauts, principalement de types électriques, à l’intérieur du système. L’apparition de ces défaillances est extrêmement délétère pour des applications à haut niveau de service et dont un arrêt intempestif peut engendrer des coûts de maintenance et d’immobilisation très élevés pour les opérateurs. Devant la nécessité de planifier de façon optimisée les opérations de maintenance à effectuer sur les alternateurs, il est possible de mettre en place des stratégies de diagnostic qui surveillent l’apparition des principales défaillances susceptibles de toucher ce type de machine. Bien que les modifications imprévisibles du point de fonctionnement liées à la charge compliquent la tâche, il est envisageable de mettre en lumière la présence de défauts de court-circuit dans les bobinages ainsi que des défaillances de diodes dans le pont redresseur triphasé en étudiant les modifications des formes d’ondes des signaux électriques générés. Ce travail est décrit dans la présente thèse. Face au manque d’antécédents sur le sujet, une grande partie des recherches s’est focalisée sur la conception et la mise en place d’un modèle numérique d’alternateur à diodes tournantes représentatif des formes d’ondes réelles en régimes sain et défaillant, tache non triviale étant donné le caractère saillant des pôles de l’alternateur. Pour répondre à ces attentes, un processus original de co-simulation a été mis en place présentant une identification des inductances de l’alternateur sous Flux2D et une estimation numérique des équations différentielles du système sous Matlab. Cette modélisation fiable a par la suite permis une sélection d’indicateurs de diagnostic par analyse fréquentielle des signaux électriques qui sont capables, sans ajout de capteurs supplémentaires, d’informer l’utilisateur sur la présence de défauts à l’intérieur du système. Afin de s’assurer une bonne compréhension des phénomènes, un grand soin a été apporté à la justification théorique des modifications spectrales introduites par les défauts dans les signaux électriques. Une importante campagne d’essais expérimentaux a permis la validation des modèles sain et défaillant grâce à la réalisation, par la société Nidec Leroy- Somer, d’un alternateur capable de simuler des défauts de court-circuit inter-spires stator. Ces essais ont mis au jour la possibilité de détecter les défauts dans de nombreuses configurations de court-circuit, mais également la difficulté de les prévoir de façon anticipée, ouvrant par là même de nombreuses perspectives de recherche. / Considering human.s huge needs in electrical energy, power alternators equipped with integrated rotating rectifiers represent a widely used solution. Indeed, this type of generator equips most of the industrial electrical power plants (hydro, fossil and wind) as well as isolated applications in the form of Diesel generators or onboard alternators. Due to their robustness and their electrical adaptability, power alternators have acquired a solid reputation and are therefore considered as an obvious solution when dealing with energy production activities. Nonetheless, the hard conditions to which they are constrained imply a dramatic deterioration of their mechanical and electrical components that can lead to serious damages, such as the loss of diodes or the emergence of short-circuits in the alternator.s coils. Such damages can make it impossible for the alternator either to fully deliver the expected power or to simply continue being operated, engendering in both cases a forced outage of the alternator and impacting dramatically the maintenance and exploitation costs. In order to avoid such issues, a setting up of diagnostic methods has been considered in this work. Even though the modifications of the electrical load is a big matter proper to electrical generation applications, a diagnostic method consisting in the monitoring of electrical signal spectra may be undertaken and can inform the user about the apparition of unintended faults. Due to the lack of knowledge on the electrical malfunctions of this kind of alternator, a great part of the thesis has been dedicated to the numerical modelling of the complete system, in healthy as well as in faulty operating conditions. Because of the salience of the rotor poles, an accurate representation of the alternator.s inductances constitutes a key point of a faithful simulation of the signals. waveforms. To offer a more convenient model, an original co-simulation process between Flux2D and Matlab has been developed. It is composed of an easy-to-use numerical model based on differential equations developed in Matlab and conserves the inductances accuracy thanks to an identification with the finite elements in Flux2D. This model could then be used to develop a diagnostic methodology based on the surveillance of several spectral components specific to different electrical signals. These components have proven themselves sensitive to the fault and offer by the way a clear view of the state of health of the system. Furthermore, a theoretical study has been undertaken to demonstrate the modifications of the fault indicators created by the occurrence of an electrical fault. In addition, a thorough test campaign has been realised in order to confirm the validity of both the numerical models and the diagnostic methodology. This campaign has become feasible thanks to the development of a special 27kVA alternator by Nidec Leroy-Somer, in which additional plugs have been inserted to give access to different turns of a stator coil, and thus to enable the realisation of stator inter-turn short-circuits. The campaign has brought valuable information about the opportunities to detect a short-circuit in a primitive phase and let the doors open to further researches in this domain.

Made available in DSpace on 2016-06-02T20:18:23Z (GMT). No. of bitstreams: 1 6236.pdf: 2944020 bytes, checksum: dcbf2b2eaca77c3425524a77887f4d6b (MD5) Previous issue date: 2014-06-06 / Universidade Federal de Minas Gerais / The aging process affects many systems of the human body, including: autonomic nervous system, which can be assessed by heart rate variability (HRV); cellular structures, such as telomere length; and mechanisms of regulation of the inflammatory process, which can be evaluated by inflammatory markers such as high-sensitivity C-reactive protein (hsCRP). The combined analysis of these variables enables the study of the aging process in a multidimensional way. Additionally, the effects of hormone replacement therapy (HRT) on HRV are contradictory. In this way, we conducted the study I, which aimed to investigate the effects of HRT on HRV in healthy postmenopausal women. Two groups were evaluated: Group 1 (G1): 20 women who did not use HRT (60 ± 5.89 years) and group 2 (G2): 20 women undergoing HRT (59 ± 5.70 years). The electrocardiogram was recorded in supine position for 10 min. Spectral analysis included low and high frequency in absolute (LF and HF) and normalized (LFnu and HFnu) units. LF/HF ratio was also calculated. Symbolic analysis (0V%, 1V%, 2LV% e 2UV%), Shannon and conditional entropy were calculated. LF, LFnu and LF/HF ratio were higher, whereas HFnu was lower in G2 than in G1. Correlations between complexity indices and HFnu were significant and positive only in G1. We conclude that women undergoing HRT had higher cardiac sympathetic modulation and reduced cardiac vagal modulation compared to women not using HRT. Moreover, the expected positive relationship between cardiac vagal modulation and HRV complexity was found only in the group not undergoing HRT, indicating that vagal modulation in women under therapy drop below a minimum value necessary to the association to become apparent, suggesting an unfavorable cardiac autonomic modulation in spite of HRT. Considering the findings of the study I, we chose to adopt the use of the therapy as an exclusion criterion for the study II. Thus, the study II aimed to examine the aging effect on heart rate variability in supine and standing, on serum hsCRP and leukocyte telomere length, as well as to verify the age at which the changes caused by aging process are accentuated. One hundred and ten volunteers were divided into five groups according to age: G21-30 years, G31-40 years, G41-50 years, G51-60 years, and G61-70 years. Venous blood samples were collected for measurements of serum hsCRP and telomere length. ECG signals were recorded in rest supine and standing (15 min in each posture). HRV was assessed by spectral analysis in low and high frequencies in absolute (LF e HF) and normalized (LFnu e HFnu) units; symbolic analysis (0V%, 1V%, 2LV% e 2UV%); Shannon entropy; and complexity index (CI) and normalized CI (NCI) from conditional entropy. The main results were: 1) HF and 2UV% reduction (vagal modulation) in G51-60, and 0V% increase (sympathetic modulation) and NCI reduction (complexity) in G61-70, in supine; 2) less efficient response to postural change from supine to standing with advancing age; 3) hsCRP increase in G51-60; 4) telomere shortening in G61-70; 5) in supine, HRV indices showed stronger relationship with the principal component of most relevance from the multivariate principal component analysis, compared to hsCRP and telomere length. Considering that HRV indices in supine had a stronger association with the aging process, we can conclude that the decrease in cardiac vagal modulation may have influenced the increase in serum hsCRP (although normal values), in G51-60, since this effect is described by the cholinergic anti-inflammatory pathway. Decreased cardiac vagal modulation and increased hsCRP may have contributed to the telomere shortening identified in the following decade (G61-70). In this way, we must consider the importance of preventive actions prior to the onset of aging effects, particularly in the 41-50 age range, in an attempt to attenuate the natural effects of senescence. / O envelhecimento exerce influência sobre vários sistemas do corpo humano, dentre eles: sistema nervoso autonômico, que pode ser avaliado pela variabilidade da frequência cardíaca (VFC); estruturas celulares, como o comprimento de telômeros; e mecanismos reguladores de processos inflamatórios, que podem ser avaliados por marcadores inflamatórios, como a proteína C-reativa ultra sensível (PCRus). A análise conjunta dessas variáveis permitiria o estudo do processo de envelhecimento de forma multidimensional. Adicionalmente, são controversos os efeitos da terapia de reposição hormonal (TRH) sobre a VFC. Assim, foi realizado o estudo I, o qual teve por objetivo investigar os efeitos da TRH na VFC em mulheres pós-menopáusicas saudáveis. Foram avaliados dois grupos: grupo 1 (G1): 20 mulheres que não faziam uso de TRH (60 ± 5,89 anos) e grupo 2 (G2): 20 mulheres submetidas à TRH (59 ± 5,70 anos). O eletrocardiograma foi registrado na posição supina por 10 min. A análise espectral incluiu a baixa e a alta frequência em unidades absolutas (BF e AF) e normalizadas (BFun e AFun). A relação BF/AF também foi calculada. A análise simbólica (0V%, 1V%, 2LV% e 2UV%), e entropias de Shannon e condicional também foram calculadas. BF, BFun e a razão BF/AF foram maiores, enquanto AFun foi menor no G2 do que no G1. As correlações entre índices de complexidade e AFun foram significativos e positivos apenas no G1. Concluímos que mulheres submetidas à TRH apresentaram maior modulação cardíaca simpática e menor modulação cardíaca vagal em comparação às que não faziam a terapia. Além disso, a relação positiva esperada entre modulação cardíaca vagal e a complexidade da VFC foi encontrada apenas no grupo não submetido à TRH, indicando que a modulação vagal em mulheres sob a terapia não atinge um valor mínimo necessário para a associação se tornar aparente, sugerindo uma modulação autonômica cardíaca desfavorável, apesar da TRH. A partir dos achados do estudo I, optou-se por adotar, como critério de exclusão para o estudo II, o uso da terapia. Assim, o estudo II teve por objetivo analisar o efeito do envelhecimento sobre a VFC nas posições supina e ortostática, os níveis séricos da PCRus e o comprimento de telômeros leucocitários, além de verificar em qual faixa etária se acentuam as alterações provocadas pelo processo de envelhecimento. Foram avaliados 110 voluntários, divididos em cinco grupos, de acordo com a idade: G21-30 anos, G31-40 anos, G41-50 anos, G51-60 anos e G61-70 anos. Amostras de sangue venoso foram coletadas para medidas de PCRus e comprimento de telômeros. Os sinais eletrocardiográficos foram registrados em repouso nas posições supina e ortostática (15 min em cada postura). A VFC foi avaliada por índices de baixa e alta frequências em unidades absolutas (BF e AF) e normalizadas (BFun e AFun) da análise espectral; índices 0V%, 1V%, 2LV% e 2UV% da análise simbólica; entropia de Shannon; e índice de complexidade (IC) e IC normalizado (ICN) da entropia condicional. Os principais resultados foram: 1) redução de AF e 2UV% (modulação vagal) em G51-60, além de aumento de 0V% (modulação simpática) e diminuição de ICN (complexidade) em G61-70 na posição supina; 2) resposta menos eficiente à manobra de mudança postural de supino para ortostatismo com o avanço da idade; 3) aumento da PCRus em G51-60; 4) encurtamento do comprimento de telômeros em G61-70; 5) na posição supina, os índices da VFC apresentaram relação mais alta com o componente principal de maior relevância, proveniente da análise multivariada por componentes principais, em comparação à PCRus e ao comprimento de telômeros. Considerando-se que os índices da VFC na posição supina apresentaram uma associação mais forte com o envelhecimento, podemos concluir que a diminuição da modulação cardíaca vagal possa ter contribuído para o aumento dos níveis séricos de PCRus (apesar dos valores estarem dentro de faixa de normalidade), na faixa etária de 51 a 60 anos, uma vez que este efeito é descrito pela via anti-inflamatória colinérgica. A diminuição da modulação cardíaca vagal e o aumento da PCRus podem ter contribuído para o encurtamento de telômeros, identificado na década seguinte, de 61 a 70 anos. Dessa maneira, torna-se importante a proposição de ações preventivas em faixas etárias anteriores ao início das alterações provocadas pelo envelhecimento, especialmente na década de 41 a 50 anos, na tentativa de atenuar os efeitos naturais da senescência.

Two surface wave testing methods, namely, (i) the spectral analysis of surface waves (SASW), and (ii) the multi-channel analysis of surface waves (MASW), form non-destructive and non-intrusive techniques for predicting the shear wave velocity profile of different layers of ground and pavement. These field testing tools are based on the dispersive characteristics of Rayleigh waves, that is, different frequency components of the surface wave travel at different velocities in layered media. The SASW and MASW testing procedure basically comprises of three different components: (i) field measurements by employing geophones/accelerometers, (ii) generating dispersion plots, and (iii) predicting the shear wave velocity profile based on an inversion analysis. For generating the field dispersion plot, the complexities involved while doing the phase unwrapping calculations for the SASW technique, while performing the spectral calculations on the basis of two receivers’ data, makes it difficult to automate since it requires frequent manual judgment. In the present thesis, a new method, based on the sliding Fourier transform, has been introduced. The proposed method has been noted to be quite accurate, computationally economical and it generally overcomes the difficulties associated with the unwrapping of the phase difference between the two sensors’ data. In this approach, the unwrapping of the phase can be carried out without any manual intervention. As a result, an automation of the entire computational process to generate the dispersion plot becomes feasible. The method has been thoroughly validated by including a number of examples on the basis of surface wave field tests as well as synthetic test data. While obtaining the dispersion image by using the MASW method, three different transformation techniques, namely, (i) the Park’s wavefield transform, (ii) the frequency (f) -wavenumber ( ) transform and (iii) the time intercept ( -phase slowness (p) transform have been utilized for generating the multimodal dispersion plots. The performance of these three different methods has been assessed by using synthetic as well as field data records obtained from a ground site by means of 48 geophones. Two-dimensional as well as three-dimensional dispersion plots were generated. The Park’s wavefield transformation method has been found to be especially advantageous since it neither requires a very high sampling rate nor an inclusion of the zero padding of the data in a wavenumber (distance) domain. In the case of an irregular dispersive media, a proper analysis of the higher modes existing in the dispersion plots becomes essential for predicting the shear wave velocity profile of ground on the basis of surface wave tests. In such cases, the establishment of the predominant mode becomes quite significant. In the current investigation for Rayleigh wave propagation, the predominant mode has been computed by maximizing the normalized vertical displacements along the free surface. Eigenvectors computed from the thin layer approach (TLM) approach are analyzed to predict the corresponding predominant mode. It is noted that the establishment of the predominant mode becomes quite important where only two to six sensors are employed and the governing (predominant) modal dispersion curve is usually observed rather than several multiple modes which can otherwise be identified by using around 24 to 48 multiple sensors. By using the TLM, it is, however, not possible to account for the exact contribution of the elastic half space in the dynamic stiffness matrix (DSM) approach. A method is suggested to incorporate the exact contribution of the elastic half space in the TLM. The numerical formulation is finally framed as a quadratic eigenvalue problem which can be easily solved by using the subroutine polyeig in MATLAB. The dispersion plots were generated for several chosen different ground profiles. The numerical results were found to match quite well with the data available from literature. In order to address all the three different aspects of SASW and MASW techniques, a series of field tests were performed on five different ground sites. The ground vibrations were induced by means of (i) a 65 kg mass dropped freely from a height of 5 m, and (ii) by using a 20 pound sledge hammer. It was found that by using a 65 kg mass dropped from a height of 5 m, for stiffer sites, ground exploration becomes feasible even up to a depth of 50-80 m whereas for the softer sites the exploration depth is reduced to about 30 m. By using a 20 lb sledge hammer, the exploration depth is restricted to only 8-10 m due to its low impact energy. Overall, it is expected that the work reported in the thesis will furnish useful guidelines for (i) performing the SASW and MASW field tests, (ii) generating dispersion plots/images, and (iii) predicting the shear wave velocity profile of the site based on an inversion analysis.

O presente texto descreve a tese de doutorado intitulada Análise de Formas usando Wavelets em Grafos. O tema está relacionado à área de Visão Computacional, particularmente aos tópicos de Caracterização, Descrição e Classificação de Formas. Dentre os métodos da extensa literatura em Análise de Formas 2D, percebe-se uma presença menor daqueles baseados em grafos com topologia arbitrária e irregular. As contribuições desta tese procuram preencher esta lacuna. É proposta uma metodologia baseada no seguinte pipeline : (i) Amostragem da forma, (ii) Estruturação das amostras em grafos, (iii) Função-base definida nos vértices, (iv) Análise multiescala de grafos por meio da Transformada Wavelet Espectral em grafos, (v) Extração de Características da Transformada Wavelet e (vi) Discriminação. Para cada uma das etapas (i), (ii), (iii), (v) e (vi), são inúmeras as abordagens possíveis. Um dos desafios é encontrar uma combinação de abordagens, dentre as muitas alternativas, que resulte em um pipeline eficaz para nossos propósitos. Em particular, para a etapa (iii), dado um grafo que representa uma forma, o desafio é identificar uma característica associada às amostras que possa ser definida sobre os vértices do grafo. Esta característica deve capturar a influência subjacente da estrutura combinatória de toda a rede sobre cada vértice, em diversas escalas. A Transformada Wavelet Espectral sobre os Grafos revelará esta influência subjacente em cada vértice. São apresentados resultados obtidos de experimentos usando formas 2D de benchmarks conhecidos na literatura, bem como de experimentos de aplicações em astronomia para análise de formas de galáxias do Sloan Digital Sky Survey não-rotuladas e rotuladas pelo projeto Galaxy Zoo 2 , demonstrando o sucesso da técnica proposta, comparada a abordagens clássicas como Transformada de Fourier e Transformada Wavelet Contínua 2D. / This document describes the PhD thesis entitled Shape Analysis by using Wavelets on Graphs. The addressed theme is related to Computer Vision, particularly to the Characterization, Description and Classication topics. Amongst the methods presented in an extensive literature on Shape Analysis 2D, it is perceived a smaller presence of graph-based methods with arbitrary and irregular topologies. The contributions of this thesis aim at fullling this gap. A methodology based on the following pipeline is proposed: (i) Shape sampling, (ii) Samples structuring in graphs, (iii) Function dened on vertices, (iv) Multiscale analysis of graphs through the Spectral Wavelet Transform, (v) Features extraction from the Wavelet Transforms and (vi) Classication. For the stages (i), (ii), (iii), (v) and (vi), there are numerous possible approaches. One great challenge is to nd a proper combination of approaches from the several available alternatives, which may be able to yield an eective pipeline for our purposes. In particular, for the stage (iii), given a graph representing a shape, the challenge is to identify a feature, which may be dened over the graph vertices. This feature should capture the underlying inuence from the combinatorial structure of the entire network over each vertex, in multiple scales. The Spectral Graph Wavelet Transform will reveal such an underpining inuence over each vertex. Yielded results from experiments on 2D benchmarks shapes widely known in literature, as well as results from astronomy applications to the analysis of unlabeled galaxies shapes from the Sloan Digital Sky Survey and labeled galaxies shapes by the Galaxy Zoo 2 Project are presented, demonstrating the achievements of the proposed technique, in comparison to classic approaches such as the 2D Fourier Transform and the 2D Continuous Wavelet Transform.

Este estudo foi realizado para testar técnicas de ressonância magnética funcional (RMf) para investigar aspectos funcionais da pineal humana. Foram coletadas imagens funcionais e amostras de sangue total para a dosagem da melatonina plasmática antes, durante e após a apresentação de estímulo com luz azul quase monocromática em indivíduos saudáveis. Os participantes realizaram o exame de tomografia computadorizada do crânio sem o uso de contraste endovenoso (TC), para a avaliação qualitativa do grau de calcificação pineal. As conclusões foram: a) não houve ativação da pineal em resposta à aplicação da luz; b) não houve diferenças estatisticamente significativas entre as condições pré-estímulo, durante o estímulo ou pós-estímulo usando diferentes métodos de análise dos dados de RMf, porém, observou-se tendência de maior poder espectral na pineal durante a aplicação do estímulo luminoso do que nas condições pré e pós-estímulo; c) foi identificada a conectividade funcional da pineal, que poderá ser melhor avaliada em estudo futuro. / This study aimed to test the functional magnetic resonance imaging (fMRI) techniques in order to investigate the functional aspects of human pineal gland. Some functional images and total blood samples for dosing the plasmatic melatonin concentration were collected before, during and after the presentation of a monochromatic blue light stimulation in healthy individuals. All subjects were examined by a brain CT scan, with no the administration of endovenous contrast, for the qualitative assessment of the pineal calcification level. The conclusions were the following: a) there was no pineal activation in response to the application of light; b) there were no statistically significant differences between the pre-, during and poststimulation conditions with different analysis methods of fMRI data, however, there was a trend of greater spectral power in the pineal gland during the luminous stimulation application than under the other conditions; c) the functional connectivity of the pineal could be identified, which should be better assessed in a future study.

Within the context of the assessment of laryngeal function, acoustic analysis has an important place because the speech signal may be recorded non-invasively and it forms the base on which the perceptual assessment of voice is founded. Given the limitations of perceptual ratings, one has investigated vocal cues of disordered voices that are clinically relevant, summarize properties of speech signals and report on a speaker's phonation in general and voice in particular. Ideally, the acoustic descriptors should also be correlates of auditory-perceptual ratings of voice. Generally speaking, the goal of acoustic analysis is to document quantitatively the degree of severity of a voice disorder and monitor the evolution of the voice of dysphonic speakers.<p><p><p>The first part of this thesis is devoted to the analysis of disordered connected speech. The aim is to investigate vocal cues that are clinically relevant and correlated with auditory-perceptual ratings. Two approaches are investigated. The variogram-based method in the temporal domain is addressed first. The second approach is in the cepstral domain. In particular, the first rahmonic amplitude is used as an acoustic cue to describe voice quality. A multi-dimensional approach combining temporal and spectral aspects is also investigated. The goal is to check whether acoustic cues in both domains report complementary information when predicting perceptual scores.<p><p><p>Both methods are tested first on a corpus of synthetic sound stimuli that has been obtained by means of a synthesizer of disordered voices. The purpose is to learn about the link between the signal properties (fixed by the synthesis parameters) and acoustic cues.<p>In this study, we had the opportunity to use two large natural speech corpora. One of them has been perceptually rated. <p><p><p>The final part of the text is devoted to the automatic classification of voice with regard to perceived voice quality. Many studies have proposed a binary (normal/pathological) classification of voice samples. An automatic categorization according to perceived degrees of hoarseness appears, however, to be more attractive to both clinicians and technologists and more likely to be clinically relevant. Indeed, one way to reduce inter-rater variability of an auditory-perceptual evaluation is to ask several experts to participate and then to average the perceptual scores. However, auditory-perceptual evaluation of a corpus by several judges is a very laborious, time-consuming and costly task. Making this perceptual evaluation task automatic is therefore desirable. <p>The aim of this study is to exploit the support vector machine classifier that has become, over the last years, a popular tool for classification, to carry out categorization of voices according to perceived degrees of hoarseness. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

Cette thèse est consacrée à l’étude de la stabilisation de certains systèmes distribués avec contrôle frontière de type dynamique. Nous considérons, d’abord, la stabilisation de l’équation de la poutre de Rayleigh avec un seul contrôle frontière dynamique moment ou force. Nous montrons que le système n’est pas uniformément (autrement dit exponentiellement) stable; mais par une méthode spectrale, nous établissons le taux polynomial optimal de décroissance de l’énergie du système. Ensuite, nous étudions la stabilisation indirecte de l’équation des ondes avec un amortissement frontière de type dynamique fractionnel. Nous montrons que le taux de décroissance de l’énergie dépend de la nature géométrique du domaine. En utilisant la méthode fréquentielle et une méthode spectrale, nous montrons la non stabilité exponentielle et nous établissons, plusieurs résultats de stabilité polynomiale. Enfin, nous considérons l’approximation de l’équation des ondes mono-dimensionnelle avec un seul amortissement frontière de type dynamique par un schéma de différence finie. Par une méthode spectrale, nous montrons que l’énergie discrétisée ne décroit pas uniformément (par rapport au pas du maillage) polynomialement vers zéro comme l’énergie du système continu. Nous introduisons, alors, un terme de viscosité numérique et nous montrons la décroissance polynomiale uniforme de l’énergie de notre schéma discret avec ce terme de viscosité. / This thesis is devoted to the study of the stabilization of some distributed systems with dynamic boundary control. First, we consider the stabilization of the Rayleigh beam equation with only one dynamic boundary control moment or force. We show that the system is not uniformly (exponentially) stable. However, using a spectral method, we establish the optimal polynomial decay rate of the energy of the system. Next, we study the indirect stability of the wave equation with a fractional dynamic boundary control. We show that the decay rate of the energy depends on the nature of the geometry of the domain. Using a frequency approach and a spectral method, we show the non exponential stability of the system and we establish, different polynomial stability results. Finally, we consider the finite difference space discretization of the 1-d wave equation with dynamic boundary control. First, using a spectral approach, we show that the polynomial decay of the discretized energy is not uniform with respect to the mesh size, as the energy of the continuous system. Next, we introduce a viscosity term and we establish the uniform (with respect to the mesh size) polynomial energy decay of our discrete scheme.

This diploma thesis deals with automatic clasification of vowel R. Purpose of this thesis is to made program for detection of pronounciation of speech defects at vowel R in children. In thesis are processed parts as speech creation, speech therapy, dyslalia and subsequently speech signal processing and analysis methods. In the last part is designed software for automatic detection of pronounciation of vowel R. For recognition of pronounciation is used algorithm MFCC for extracting features. This features are subsequently classified by neural network to the group of correct or incorrect pronounciation and is evaluated classification success.

In this diploma thesis the seismic load on two different models is solved – the first model is a beam hall and the second model is a multistory building. The calculation of seismic load was performed according to standard ČSN EN 1998-1 (Eurocode 8) using the response spectrum analysis and the method of equivalent static forces (substi-tute load). For the combination of seismic responses in different directions are used SRSS and CQC rules. Models and calculations were solved in a student’s version of program RFEM 4.10 (Ing. Software Dlubal, s.r.o.). The comparison of mentioned methods in each models, also the comparison of combinative rules in calculation and the final evaluation of seismic responses on both of models are the subject of this thesis.

Cette thèse vise à étudier certains modèles biologiques dans le réseau neuronal et dans la chimiotaxie avec la méthode d’analyse spectrale. Afin de traiter les principaux problèmes, tels que l’existence et l’unicité des solutions et des états stationnaires ainsi que les comportements asymptotiques, le modèle linéaire ou linéarisé associé est considéré par l’aspect du spectre et des semi-groupes dans les espaces appropriés, puis la stabilité de modèle non linéaire suit. Plus précisément, nous commençons par une équation de courses-et-chutes linéaire dans la dimension d≥1 pour établir l’existence d’un état stationnaire unique, positif et normalisé et la stabilité exponentielle asymptotique dans l’espace L¹ pondéré basé sur la théorie de Kerin-Rutman avec quelques estimations du moment de la théorie cinétique. Ensuite, nous considérons le modèle du temps écoulé sous les hypothèses générales sur le taux de tir et nous prouvons l’unicité de l’état stationnaire et sa stabilité exponentielle non linéaire en cas sans ou avec délai au régime de connectivité faible de la théorie de l’analyse spectrale pour les semi-groupes. Enfin, nous étudions le modèle sous une hypothèse de régularité plus faible sur le taux de tir et l’existence de la solution ainsi que la même stabilité exponentielle sont généralement établies n’importe la prise en compte du délai ou non, au régime de connectivité faible ou forte. / This thesis is aimed to study some biological models in neuronal network and chemotaxis with the spectral analysis method. In order to deal with the main concerning problems, such as the existence and uniqueness of the solutions and steady states as well as the asymptotic behaviors, the associated linear or linearized model is considered from the aspect of spectrum and semigroups in appropriate spaces then the nonlinear stability follows. More precisely, we start with a linear runs-and-tumbles equation in dimension d≥1 to establish the existence of a unique positive and normalized steady state and the exponential asymptotic stability in weighted L¹ space based on the Krein-Rutman theory together with some moment estimates from kinetic theory. Then, we consider time elapsed model under general assumptions on the firing rate and prove the uniqueness of the steady state and its nonlinear exponential stability in case without or with delay in the weak connectivity regime from the spectral analysis theory for semigroups. Finally, we study the model under weaker regularity assumption on the firing rate and the existence of the solution as well as the same exponential stability are established generally no matter taking delay into account or not and no matter in weak or strong connectivity regime.