L'auteur n'a pas fourni de résumé en français. / L'auteur n'a pas fourni de résumé en anglais.

Made available in DSpace on 2014-10-09T12:35:10Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:00:55Z (GMT). No. of bitstreams: 0 / Dissertação (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

Made available in DSpace on 2014-10-09T12:54:13Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:09:27Z (GMT). No. of bitstreams: 1 12432.pdf: 12974191 bytes, checksum: ddfc093df6afa20374e446cb5c994b88 (MD5) / Tese (Doutoramento) / IPEN/T / Escola Politecnica, Universidade de São Paulo - POLI/USP

Made available in DSpace on 2014-10-09T12:35:10Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:00:55Z (GMT). No. of bitstreams: 0 / Os tubos de Geradores de Vapor em Reatores Nucleares do tipo PWR são submetidos a diferentes níveis de tensões e carregamento em altas temperaturas, reduzindo sua vida útil devido o surgimento de defeitos e corrosão. A inspeção por Correntes Parasitas é um ensaio não destrutivo usado para diagnosticar defeitos de corrosão e descontinuidades na superfície externa e interna em tubos de trocadores de calor. Esses tubos estão sujeitos a danos por diferentes mecanismos de degradação mecânica e química, tais como trincas por fadiga e corrosão sob tensão. Os sinais de inspeção por Correntes Parasitas são afetados por diferentes ruídos dificultando sua análise pelo inspetor. Esse trabalho apresenta os resultados da análise dos sinais de Correntes Parasitas usando a Transformada de Hilbert-Huang (THH) funcionando como filtro de ruídos (De-noising), como uma técnica alternativa de processamento e análise de sinais. A Transformada de Hilbert-Huang teve esse nome atribuído pela agência espacial norte-americana (NASA) para o resultado da reunião de dois processos, um método de decomposição empiricamente modal (Empirical Mode Decomposition EMD), seguido da análise espectral de Hilbert (Hilbert Spectral Analysis HSA). Os sinais de inspeção por correntes parasitas possuem características de transiente, não estacionário e não linear. A transformada de Hilbert-Huang aplicada neste trabalho forneceu dois recursos alternativos em processamento de sinais; o pré-processamento que funcionou como filtro de ruídos, e outro de análise de sinais, responsável pela identificação das características tempo-frequência-energia do sinal. / Dissertação (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

Made available in DSpace on 2014-10-09T12:54:13Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:09:27Z (GMT). No. of bitstreams: 1 12432.pdf: 12974191 bytes, checksum: ddfc093df6afa20374e446cb5c994b88 (MD5) / Tese (Doutoramento) / IPEN/T / Escola Politecnica, Universidade de São Paulo - POLI/USP

Magnetic induction tomography (MIT) is a non-invasive, soft field imaging modality that has the potential to map the electrical conductivity (σ) distribution inside an object under investigation. In MIT, a number of exciter and receiver coils are distributed around the periphery of the object. A primary magnetic field is emitted by each exciter, and interacts with the object. This induces eddy currents in the object, which in turn create a secondary field. This latter is coupled to the receiver coils and voltages are induced. An image reconstruction algorithm is then used to infer the conductivity map of the object. In this thesis, the application of MIT for volumetric imaging of objects with low conductivity materials (< 5 Sm-1) and dimensions < 1 m is investigated. In particular, two low conductivity applications are approached: imaging cerebral stroke and imaging the saline water in multiphase flows. In low conductivity applications, the measured signals are small and the spatial sensitivity is critically compromised making the associated inverse problem severely non-linear and ill-posed.The main contribution from this study is to investigate three non-linear optimisation techniques for solving the MIT inverse problem. The first two methods, namely regularised Levenberg Marquardt method and trust region Powell's Dog Leg method, employ damping and trust region strategies respectively. The third method is a modification of the Gauss Newton method and utilises a damping regularisation technique. An optimisation in the convergence and stability of the inverse solution was observed with these methods compared to standard Gauss Newton method. For such non linear treatment, re-evaluation of the forward problem is also required. The forward problem is solved numerically using the impedance method and a weakly coupled field approximation is employed to reduce the computation time and memory requirements. For treating the ill-posedness, different regularisation methods are investigated. Results show that the subspace regularisation technique is suitable for absolute imaging of the stroke in a real head model with synthetic data. Tikhonov based smoothing and edge preserving regularisation methods also produced successful results from simulations of oil/water. However, in a practical setup, still large geometrical and positioning noise causes a major problem and only difference imaging was viable to achieve a reasonable reconstruction.

Cette thèse se situe dans le cadre de la conception du réacteur nucléaire à caloporteur sodium ASTRID, actuellement à l'étude au CEA Cadarache. La mesure de vitesse du sodium liquide dans les boucles primaire et secondaire, ainsi que la détection de bulles, en cas de fuite par exemple, représentent un enjeu majeur pour le contrôle et la surveillance en fonctionnement du réacteur. Le caractère conducteur électrique du sodium liquide permet d'envisager l'utilisation de débitmètres à courants de Foucault, ou ECFM (pour Eddy Current Flow Meter) pour répondre à ces fonctions. Une précédente thèse réalisée par Mitlesh Kumar a permis de découpler le signal mesuré à l'ECFM relatif à la vitesse du métal, de celui relatif à la présence d'hétérogénéités dans le métal (telles que des bulles). L'objet de la présente thèse est de caractériser expérimentalement et analytiquement les effets de la vitesse et de la présence d'une inclusion isolante sur le signal mesuré, à partir d'écoulements modèles. Cette approche, complémentaire avec un certain nombre d'études en écoulement réel recensées dans la littérature, a pour objectif d'apporter un outil de compréhension physique au système. Deux dispositifs expérimentaux en métal liquide (galinstan) sont réalisés. Le premier dispositif représente un écoulement de galinstan en conduite cylindrique avec un champ de vitesse uniforme (écoulement piston), contenant une inclusion rigide électriquement isolante simulant une bulle. Le second dispositif expérimental est un écoulement en conduite de galinstan réel en l'absence d'inclusion. Les paramètres contrôlés sont la vitesse débitante du métal liquide (de 0.01 à 1 m/s), la taille et la position de l'inclusion (rayon entre 1 et 2.5 mm, profondeur entre 3 et 6 mm) et la fréquence (de 0.5 à 20 kHz). Le rayon de la conduite vaut 12.5 mm, et l'épaisseur de peau varie entre 2.4 et 15.3 mm sur cette plage de fréquence. Deux modèles théoriques, basés sur la résolution de l'équation d'induction du potentiel vecteur, sont en outre développés pour déterminer les effets de la vitesse et les effets de la présence de l'inclusion sur le signal mesuré. Dans les deux expériences, il est montré que le signal mesuré relatif à la vitesse du métal varie linéairement avec la vitesse et est maximal à une certaine fréquence (f = 2 kHz ici). Ces résultats correspondent à ceux du premier modèle théorique et montrent un bon accord avec la littérature. Dans la première expérience par ailleurs, le passage de l'inclusion à travers l'ECFM se manifeste par une oscillation du signal mesuré. Le comportement de l'oscillation est correctement décrit par le second modèle théorique dans la limite des basses fréquences (jusqu'à 2 kHz) : l'amplitude de l'oscillation est alors proportionnelle au volume de l'inclusion et suit une loi de puissance en f^2. Aux hautes fréquences, il est montré que l'amplitude et le déphasage du signal mesuré relatif à la présence de l'inclusion sont impactés de manière importante par la taille et la profondeur de l'inclusion. Un début de méthode inverse est développé à partir de ce résultat, pour déterminer la taille et la position de l'inclusion / This thesis is included in the conception of the prototype of Sodium Fast Reactor (SFR) ASTRID, currently studied in the CEA Cadarache. Velocimetry of liquid sodium in the primary andsecondary loops, and bubble detection in sodium (e.g. in case of leaks) represent a major issue for the control and oversight of the reactor. The electrical conductive property of liquid sodium allows to consider the use of Eddy Current Flow Meters (ECFM) for this purpose. A previous thesis realized by Mithlesh Kumar highlighted a decoupling of the signal measured with the ECFMrelative to the velocity, to the one relative to the presence of heterogeneities (such as bubbles). The object of the present thesis is to caracterize experimentally and analytically the effects of velocity and the effects of the presence of an insulating inclusion on the measured signal, thanks to modeled flows. This approach, complementary with most of the studies of real flows existing in the litterature, aims to give a tool for a physical comprehension of the system. Two experimental set-ups with liquid metal (galinstan) have been developed. The first set-up represent a galinstan flow in a cylindrical duct at uniform velocity (plug flow), advecting electrically insulating rigidinclusion simulating a bubble. The second experimental set-up is a galinstan flow in a cylindrical duct without inclusion. The controled parameters are the flow velocity (from 0.01 to 1 m/s), the size and location of the inclusion (radius from 1 to 2.5 mm, depth of 3 and 6 mm) and the frequency (from 0.5 to 20 kHz). The radius of the duct is equal to 12.5 mm, and the skin depth varies between 2.4 and 15.3 mm for this frequency range. Two theoretical models, based on the resolution of the induction equation of the vector potential, are moreover developed to determine velocity effects and inclusion effects on the measured signal. In both experimental studies, it is shown that the measured signal relative to the liquid metal velocity varies linearly with velocity and is maximal at a given frequency (f = 2 kHz here). These results are corresponding well with those of the first theoretical model and show a good agreement with litterature. Besides, in the first experimental study, the passage of the inclusion through the ECFM manifests itself by an oscillation of the measured signal. The behaviour of the oscillation is well described by the second theoretical model within the limit of low frequencies (up to 2 kHz) : the amplitude of the oscillation is then proportionnal to the inclusion volume and follows a power law in f^2. At high frequencies, it is shown that amplitude and phaseshift of the measured signal relative to the presence of an inclusion are highly impacted by inclusion size and depth. A first step of inverse method is developed from this result, in order to determine size and location of an inclusion.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23511号 / 工博第4923号 / 新制||工||1769(附属図書館) / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 松尾 哲司, 教授 雨宮 尚之, 准教授 久門 尚史 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Práce se zabývá problematikou měření momentu u velmi vysokootáčkových strojů. Z důvodů vysokých rychlostí jsou tradiční metody měření momentu nepoužitelné, z tohoto důvodu je nutno vyvinout novou bezkontaktní metodu. Pro měření je zvolena metoda zatěžování vířivou brzdou, pro kterou je proveden analytický výpočet jejích vlastností pomocí Maxwellových rovnic a je vytvořen model vířivé brzdy v programu FEMM a Ansys/Maxwell 2D. Na závěr jsou výsledky simulací a výpočtů porovnány s měřením na prototypech brzdy.

The diploma thesis deals with the design of a magnetorheological damper with a short response time of the damping force for the bogie of a rail vehicle. The largest component, which slows down the response, is the eddy currents generated in the piston group of the MR damper. A rapid onset of damping force is achieved by using a shape approach in magnetic circuit design. The shape approach increases the distance that the eddy currents must travel, and therefore it helps to reduce them. The shape approach is realized by grooving the core and the shell of the MR damper. The proposed fast version of the MR damper is measured on a hydraulic pulsator and its results show a sufficient time response of the damping force. This response does not exceed 8 ms for the primary onset of the damping force. The new fast design of the MR damper combines the greatest advantages of the technology and enables to meet new requirements for damping elements in the rolling stock chassis.