The main focus of this study is to improve the profitability, availability and condition monitoring of Liquefied Natural Gas (LNG) Floating Production Storage and Offloading platforms (FPSOs). Propane pre-cooled, mixed refrigerant (C3MR) liquefaction is the key process in the production of LNG on FPSOs. LNG liquefaction system equipment has the highest failure rates among the other FPSO equipment, and thus the highest maintenance cost. Improvements in the profitability, availability and condition monitoring were made in two ways: firstly, by making recommendations for the use of redundancy in order to improve system reliability (and hence availability); and secondly, by developing an effective condition-monitoring algorithm that can be used as part of a condition-based maintenance system. C3MR liquefaction system reliability modelling was undertaken using the time-dependent Markov approach. Four different system options were studied, with varying degrees of redundancy. The results of the reliability analysis indicated that the introduction of a standby liquefaction system could be the best option for liquefaction plants in terms of reliability, availability and profitability; this is because the annual profits of medium-sized FPSOs (3MTPA) were estimated to increase by approximately US$296 million, rising from about US$1,190 million to US$1,485.98 million, if redundancy were implemented. The cost-benefit analysis results were based on the average LNG prices (US$500/ton) in 2013 and 2014. Typically, centrifugal turbines, compressors and blowers are the main items of equipment in LNG liquefaction plants. Because centrifugal equipment tops the FPSO equipment failure list, a Condition Monitoring (CM) system for such equipment was proposed and tested to reduce maintenance and shutdown costs, and also to reduce flaring. The proposed CM system was based on a novel FFT-based segmentation, feature selection and fault identification algorithm. A 20 HP industrial air compressor system with a rotational speed of 15,650 RPM was utilised to experimentally emulate five different typical centrifugal equipment machine conditions in the laboratory; this involved training and testing the proposed algorithm with a total of 105 datasets. The fault diagnosis performance of the algorithm was compared with other methods, namely standard FFT classifiers and Neural Network. A sensitivity analysis was performed in order to determine the effect of the time length and position of the signals on the diagnostic performance of the proposed fault identification algorithm. The algorithm was also checked for its ability to identify machine degradation using datasets for which the algorithm was not trained. Moreover, a characterisation table that prioritises the different fault detection techniques and signal features for the diagnosis of centrifugal equipment faults, was introduced to determine the best fault identification technique and signal feature. The results suggested that the proposed automated feature selection and fault identification algorithm is effective and competitive as it yielded a fault identification performance of 100% in 3.5 seconds only in comparison to 57.2 seconds for NN. The sensitivity analysis showed that the algorithm is robust as its fault identification performance was affected by neither the time length nor the position of signals. The characterisation study demonstrated the effectiveness of the AE spectral feature technique over the fault identification techniques and signal features tested in the course of diagnosing centrifugal equipment faults. Moreover, the algorithm performed well in the identification of machine degradation. In summary, the results of this study indicate that the proposed two-pronged approach has the potential to yield a highly reliable LNG liquefaction system with significantly improved availability and profitability profiles.

Une technique de corrélation d’images numériques (DIC) a été adaptée à l’étuded’échantillons de glace polycristalline colonnaire, afin de caractériser le développementdes hétérogénéités de déformation à une échelle intragranulaire au cours du fluage transitoireen compression. Ces hétérogénéités sont dues à la grande anisotropie viscoplastiquedu cristal de glace (avec un seul plan de glissement facile pour les dislocations), qui estresponsable d’interactions fortes entre les grains voisins lorsque l’échantillon se déforme,et de la mise en place de fortes contraintes internes. L’orientation particulière des plans deglissement et la structure colonnaire des échantillons, sans gradient de microstructure dansl’épaisseur, permet de considérer la déformation plastique essentiellement dans le plan.Les champs de déformation mesurés en surface peuvent donc être considérés en premièreapproximation comme représentatifs de tout le volume de l’échantillon. Les orientationscristallographiques, et désorientations locales à l’échelle intragranulaire, ont également étémesurées avec un analyseur optique de texture, avant et après déformation.De très fortes localisations de la déformation sont mises en évidence par la DIC;les déformations équivalentes locales peuvent atteindre plus de dix fois la valeur macroscopique.La structure des motifs de localisation évolue peu durant le régime de fluagetransitoire et elle est caractérisée par des bandes de localisation qui traversent l’échantillonobliquement, sur des distances allant jusqu’à dix fois la taille des grains. La corrélationentre le facteur de Schmid du plan de glissement d’un grain et sa déformation locale esttrès faible : l’agencement local des grains et les interactions intergranulaires jouent doncun rôle primordial dans la distribution spatiale des déformations. A l’échelle du grain, lesfortes désorientations de réseau ont été associées aux composantes locales du déplacement.On a montré en particulier que les distorsions sous forme de ”bandes en genou” (ou kinkbands) étaient corrélées à un déplacement le long de l’axe c du grain, mesuré par DIC.Les microstructures expérimentales ont été implémentées dans un modèle plein champFFT qui simule le comportement élasto-viscoplastique des polycristaux (code CraFT).Les champs de déformation simulés présentent globalement les mêmes caractéristiquesque les champs expérimentaux. En particulier, la majorité des zones de localisation dela déformation sont bien reproduites, spatialement et en intensité. Cependant, il n’a pasété possible de reproduire le comportement macroscopique des échantillons à la décharge.La déformation retardée effective est d’un ordre de grandeur inférieure à celle mesuréeexpérimentalement. Il est probable que cette limitation soit liée à l’e↵et du champ descontraintes internes, généré par la structure des champs de dislocations, et qui n’a pas étépris en compte dans la loi de comportement. / A Digital Image Correlation (DIC) technique has been adapted to polycrystalline icespecimens in order to characterize the development of strain heterogeneities at an intra-granular scale during transient creep deformation. These heterogeneities are due to thestrong viscoplastic anisotropy of the ice single crystal (with only one easy glide planefor dislocations), inducing strong mechanical interactions between adjacent grains duringdeformation, and the establishment of strong internal stresses. Specimens exhibit a co-lumnar microstructure so that plastic deformation is essentially 2-D with few in-depthgradients, and therefore surface DIC analyses are representative for the whole specimenvolume. Local misorientations at the intragranular scale were also extracted from mi-crostructure analyses carried out with an automatic texture analyzer before and afterdeformation. Highly localized strain patterns are evidenced by the DIC technique. Localequivalent strain can reach values as high as one order of magnitude larger than the ma-croscopic average. The structure of the strain pattern does not evolve signicantly withstrain during the transient creep regime. Almost no correlation was measured betweenthe local strain and the Schmid factor of the slip plane of the underlying grain ; thishighlights the importance of the mechanical interactions between neighboring grains. Atthe grain scale, the strong lattice misorientations have been associated to the local com-ponents of the displacement. In particular, kink band distortions could be correlated to adisplacement along the c-axis measured by DIC. The experimental microstructures wereintroduced into a full-eld FFT model that simulates the behaviour of elasto-viscoplasticpolycrystals (CraFT code). The simulated deformation elds present globally the samecharacteristics as the experimental ones. Most of the localization bands are well repro-duced, spatially and quantitatively. However, we could not reproduce the macroscopicbehaviour of the specimens upon unloading. The predicted recovery eective strain is oneorder of magnitude lower than the measured one. This limitation might be linked withthe internal stress eld induced by the dislocation structure, which is not explicitly takeninto account in the local constitutive relation.

In this thesis we evaluate the prospects of performing real-time digital signal processing on a graphics processing unit (GPU) when linked together with a high-performance digitizer. A graphics card is acquired and an implementation developed that address issues such as transportation of data and capability of coping with the throughput of the data stream. Furthermore, it consists of an algorithm for executing consecutive fast Fourier transforms on the digitized signal together with averaging and visualization of the output spectrum. An empirical approach has been used when researching different available options for streaming data. For better performance, an analysis of the introduced noise of using single-precision over double-precision has been performed to decide on the required precision in the context of this thesis. The choice of graphics card is based on an empirical investigation coupled with a measurement-based approach. An implementation in single-precision with streaming from the digitizer, by means of double buffering in CPU RAM, capable of speeds up to 3.0 GB/s is presented. Measurements indicate that even higher bandwidths are possible without overflowing the GPU. Tests show that the implementation is capable of computing the spectrum for transform sizes of <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?2%5E%7B21%7D" />, however measurements indicate that higher and lower transform sizes are possible. The results of the computations are visualized in real-time.

GPGPU is a collective term for research involving general computation on graphics cards. A modern graphics card typically provides more than ten times the computational power of an ordinary PC processor. This is a result of the high demands for speed and image quality in computer games. This thesis investigates the possibility of exploiting this computational power for image processing purposes. Three well known methods where implemented on a graphics card: FFT (Fast Fourier Transform), KLT (Kanade Lucas Tomasi point tracking) and the generation of scale pyramids. All algorithms where successfully implemented and they are tree to ten times faster than correspondning optimized CPU implementation.

3GPP LTE (Long Term Evolution) is a project of the Third Generation Partnership Project to improve the UMTS (Universal Mobile Telecommunications System) mobile phone standard to cope with future technology evolutions. Two duplex schemes FDD and TDD are investigated in this thesis. Several computational intensive components of the baseband processing for LTE uplink such as synchronization, channel estimation, equalization, soft demapping, turbo decoding is analyzed. Cost analysis is hardware independent so that only computational complexity is considered in this thesis. Hardware dependent discussion for LTE baseband SDR platform is given according the analysis results.

The advanced signal processing systems of today require extreme data throughput and low power consumption.The only way to accomplish this is to use parallel processor architecture with efficient algorithms.The aim of this thesis was to evaluate the use of parallel processor architecture in Radar signal processingapplications where the processor has to compute complex calculations. This has been done by implementingdemanding algorithms on Ambric Am2000 family Massively Parallel Processor Array (MPPA). The Ambricplatform evaluated in terms of Latency, Cycle Count per output sample and Efficiency of the developmenttools.The two algorithms chosen for implementation are Fast Fourier Transform (FFT) and Finite Impulse Response(FIR) algorithms. We have implemented parameterized versions of FFT and FIR. The FFT algorithmimplemented for N-point input for the range of 8 point to 32 point for complex input variables. It works for anygiven number of inputs within the range for given parameter values and mapped on Ambric processor withfixed point radix - 2. Another one is FIR algorithm for the range of 12 Taps to 64 Taps for complex inputvariables. The Implementation of algorithms shows that high level of parallelism can be achieved in MassivelyParallel Processing Arrays (MPPA) especially on complex algorithms like FFT and FIR.

Abstract Functional magnetic resonance imaging (fMRI) has become a leading tool in the evaluation of the human brain function. In fMRI the activation induced blood oxygenation changes in the brain can be detected with an inherent blood oxygen level dependent (BOLD) contrast. Even small blood oxygen fluctuations in a resting brain can be depicted with the BOLD contrast. This thesis focuses on characterizing spontaneous oxygenation fluctuations of the brain by using BOLD fMRI. The effects of anesthetics on blood oxygen fluctuations were assessed in 38 children and 12 adults. The spatial distribution, frequency, synchrony, and statistical independence of the spontaneous oxygenation changes were analyzed. The role of imaging artifacts in the generation of BOLD signal fluctuations was investigated. The study aimed to develop and compare methods of detecting the nondeterministic oxygenation fluctuations of the brain. VLF BOLD signal fluctuation in the brain cortex is induced by physiological oscillation instead of imaging artifacts. This study shows for the first time how the power and synchrony of very low frequency (VLF &lt;  0.05 Hz) blood oxygen fluctuation significantly increases after sedation. In deeper anesthesia, the VLF fluctuation overpowers other sources of blood oxygen variation as a sign of reduced blood flow and altered hemodynamic control. Regional hemodynamic mechanisms induce non-Gaussian features on the VLF blood oxygen fluctuation that can be depicted effectively with independent component analysis. Combined use of frequency, time, and spatial domain analysis guarantees a more complete picture of brain oxygenation fluctuations. The results of this thesis have a dualistic impact on fMRI research. First of all, VLF fluctuation alters the BOLD activation and connectivity results after sedation. Thus it has to be accounted for in the fMRI of sedated subjects. Secondly, by using the methods developed in this thesis, VLF fluctuation and other physiological BOLD signal sources can now be used in characterizing physiological alterations and pathology of the brain.

Ce travail de thèse s'inscrit dans le cadre de la description de la rupture ductile des métaux. L'influence couplée de la plasticité cristalline et de la porosité est abordée à l'aide d'approches micromécaniques. A partir des travaux existants en rupture ductile, notamment pour les matériaux isotropes poreux, et en nous appuyant sur les études traitant des monocristaux poreux, nous proposons une approche originale de la prédiction du comportement de divers matériaux cristallins poreux.Une approche numérique de cette question a tout d'abord était menée. Des outils d'homogénéisation numérique ont été développés et mis en œuvre pour prédire la résistance de monocristaux présentant une répartition périodique de cavités. Une étude de l'influence de l'orientation cristalline sur la surface de plasticité, sur l'écrouissage et l'évolution de la porosité a été effectuée pour différentes structures cristallines. Ces résultats numériques ont servi de référence pour évaluer la pertinence d'un modèle analytique approché que nous avons proposé. Celui-ci se fonde sur une approche par analyse limite intégrant l'anisotropie plastique cristalline. Notre critère a pu être ainsi validé pour différentes situations d'anisotropie du monocristal. Une extension permettant de décrire l'influence de l'écrouissage, en tenant compte de l'hétérogénéité de la déformation plastique, a été proposée. Notre critère a ensuite été utilisé dans un contexte polycristallin afin de décrire la plasticité de métaux présentant de la porosité intragranulaire.Enfin, le cas particulier de monocristaux n'obéissant pas à la loi de Schmid (plasticité cristalline non associée) est considéré. Celle-ci n'entrant pas dans le cadre classique de l'analyse limite, une approche basée sur le bipotentiel proposée par de Saxcé a été choisie. Un bipotentiel spécifique au modèle de plasticité cristalline non associée de Qin et Bassani pour le cristal cubique à faces centrées a été développé. Celui-ci est utilisé dans le cadre de l'homogénéisation par bipotentiel afin d'obtenir un critère de plasticité et une loi d'écoulement. Les outils numériques développés pour la plasticité associée ont été adaptés à la plasticité non associée, et des simulations numériques ont été réalisées pour confronter la loi de comportement obtenue. / This work is a part of the description of ductile rupture of metals. The coupled influences of the crystalline plasticity and the porosity growth is studied by micromecanichal approach. Following existing works in ductile rupture, especially for isotropic porous materials, and using works on porous single crystals, we propose an original approach to predict the behaviour of several crystalline porous materials.First, a numerical approach has been performed. Numerical homogenization tools has been developped and used to study the resistance of single crystals with periodic microstructures. The influence of the crystalline orientation on the plastic yield limit, the hardening and the evolution of the porosity has been evaluated for different crystalline structures.These numerical results has been used to assess an approximate yield criterion we proposed. This one is derived from a limit analysis process taking into consideration the plastic crystalline anisotropy. Thus, this criterion has been assessed for different plastic crystalline anisotropies. An extension to take into account the hardening has been proposed using an evaluation of the heterogeneity of the strain field. An exemple of use of this criterion for numerical computation of polycrystal with intragranular porosity is also proposed. Finally, the particular case of single crystals obeying non Schmid plasticity (non associated plasticity) is considered. As it does not fall in the classical scope of limit analysis, an approach based on the bipotential proposed by de Saxcé has been used. A specific bipotential for the non associated crystalline plasticity proposed by Qin and Bassani has been derived. Then, a macroscopic yield criterion and a macroscopic flow rule is derived by homogenization by bipotential. Numerical tools developped for the associated plasticity have been adaptated for the non associated one. Numerical simulations are performed to assess the behaviour law obtained.

Contact-less human detection and monitoring using radar technology has been recently applied in many areas including search-and-rescue for earthquake victims, fall detection, gait analysis and detection of other human activities. Radars can also provide important information about a persons state of health by monitoring the level of activities, heart and breathing rate. Also it can be used to generate warnings if some of the monitored parameters are outside of predefined limits. The major application of this work is for monitoring in-mates and their activities. This thesis deals with characterizing the performance of the radar system used for monitoring a single person in a contained environment. This thesis is experimentally based and during the thesis a large number of experiments were performed in order to monitor subjects in realistic conditions. The thesis explores feasibility of using the radar with a single radio-frequency channel input and two algorithms for breathing and heart rate estimation when the subject is at different relative orientation towards the radar as well as in different postures. Algorithm one is using Fast Fourier Transformation (FFT) and algorithm two is using Empirical Mode Decomposition (EMD) with Minkowski distance. We also detect the zones where the subject is when the subject is moving. Since this exploratory analysis provides initial features for classifications and algorithms for breathing and heart beat estimation, it can represent a foundation for future works on designing systems that track subjects and their breathing in real-time.

The doctoral thesis proposes a synchronization event detection method suitable for establishing a common time base across devices using power-line communication. The technique facilitates the creation of a synchronization procedure for smart grid terminal devices, thus bringing to the low-voltage segment new functionalities already known from the higher voltage levels of the distribution network. High-precision, zero-cross line voltage detection was employed as the synchronization event. The use of the PLL (phase-locked loop) was considered and analyzed as a potentially applicable option; however, such a solution proved to be insufficiently robust against interference. Further, an FFT-based approach was designed and reviewed. The technique performs zero-cross detection, but only for the fundamental harmonic component that remains the same across the entire network. The influence of incoherent sampling on the FFT, too, was tested during the development of the method; in this context, coherent sampling was found to constitute an essential prerequisite for accurate detection as it allows us to eliminate the spectral leakage error. The impact of an AD converter on the detection accuracy was also evaluated. The proposed approach ensures the accuracy required for all common measurements and operations to improve the production, distribution, and consumption of electricity through smart grids.