Thesis (M.Sc. (Agric.)) -- University of Pretoria, 2008. / Abstract in English. Includes bibliographical references.

Thèse (Ph. D.)--Université Laval, 2007. / Titre de l'écran-titre (visionné le 5 mai 2008). Bibliogr.

Th. doct.-ing.--Besançon, 1983. N°: 129.

This thesis presents an investigation into the application of optical fibre sensors to a tomographic imaging system for use with gas/water mixtures. Several sensing techniques for measurement of two component flow using non-intrusive techniques are discussed and their relevance to tomographic applications considered. Optical systems are shown to be worthy of investigation. The interaction between a collimated beam of light and a spherical bubble is described. Modelling of different arrangements of projections of optical sensing arrays is carried out to predict the expected sensor output voltage profiles due to different flow regimes represented by four models. The four flow models investigated are: a single pixel flow, two pixels flow, half flow and full flow models. The response of the sensors is based on three models: optical path length, optical attenuation and a combination of optical attenuation model and signal conditioning. In the optical path length model, opaque solids or small bubbles, which are conveyed, may totally or partially interrupt the optical beams within the sensing volume. In the optical attenuation model, the Lambert-Beer's Law is applied to model optical attenuation due to the different optical densities of the fluids being conveyed. The combination of optical attenuation model and signal conditioning is designed to improve the visual contrast of the tomograms compared with those based on the optical attenuation model. Layergram back-projection (LYGBP) is used to reconstruct the image. A hybrid reconstruction algorithm combining knowledge of sensors reading zero flow with LYGBP is tested and shown to improve the image reconstruction. The combination of a two orthogonal and two rectilinear projections system based on optical fibres is used to obtain the concentration profiles and velocity of gas bubbles in a vertical column. The optical fibre lens is modelled to determine the relationships between fibre parameters and collimation of light into the receiver circuit. Modelling of the flow pipe is also carried out to investigate which method of mounting the fibres minimises refraction of the collimated light entering the pipe and the measurement cross-section. The preparation of the ends of the optical fibre and design of the electronics, which process the tomographic data, are described. Concentration profiles obtained from experiments on small bubbles and large bubbles flowing in a hydraulic conveyor are presented. Concentration profiles are generated using the hybrid reconstruction algorithm. The optical tomographic system is shown to be sensitive to small bubbles in water of diameter 1-10 mm and volumetric flow rates up to 1 1/min, and large bubbles in water of diameter 15-20 mm and volumetric flow rates up to 3 1/min. Velocity measurements are obtained directly from cross correlation of upstream and downstream sensors' signals as well as from upstream and downstream pixel concentration values. Suggestions for further work on optical tomographic measurements are made.

Ces travaux concernent le développement d’un amplificateur à fibre optique souple, microstructurée, double-gaine, dopée ytterbium (Yb), et monomode à large coeur, dans la gamme d’impulsion nanoseconde, multi-kiloHertz et milliJoule, pour l’injection de chaînes lasers de puissance. L’architecture amplificatrice est mise en oeuvre dans une configuration MOPA (Master Oscillator Power Amplifier) à plusieurs étages. Un modèle numérique de l’amplification sur fibre double-gaine dopée Yb, incluant l’émission spontanée amplifiée, a été développé pour étudier le comportement de ce type d’amplificateur fibré et procéder au dimensionnement du dispositif expérimental. Afin de s’affranchir du processus de saturation par le gain, un algorithme de contre-réaction permettant de déterminer numériquement la forme temporelle optimale a été associé au modèle. Nous avons obtenu des résultats expérimentaux en bon accord avec les simulations numériques, et avec les performances suivantes : une énergie de 0.5 mJ par impulsion à une fréquence de répétition dans la gamme de 1 kHz à 10 kHz, sur des impulsions à spectre étroit centré à la longueur d’onde 1053 nm, à profil temporel super-gaussien d’ordre 20 de durée 10 ns, avec un rapport signal-sur-bruit optique supérieur à 50 dB et un taux de maintien de la polarisation à 20 dB. Le profil spatial en sortie de système est monomode (M²=1.1). Ce dispositif peut également délivrer des énergies jusqu’à 1.5 mJ. Nous avons ensuite mis à profit ces performances pour l’amplification d’impulsions à dérive de fréquence, et avons obtenu une énergie par impulsion de 0.7 mJ sur une durée de 570 fs, à une fréquence de répétition de 10 kHz. / This work concerns the development of a double-clad ytterbium-doped single-mode microstructured flexible fiber-based amplifier, in the nanosecond, multi-kiloHertz and milliJoule regime, for large-scale laser facilities seeding. We have used a multi-stage master oscillator power amplifier fibered architecture. A numerical model of ytterbium-doped double-clad fiber-based amplification, including amplified spontaneous emission, was developed in order to study the behaviour of such amplifier and to correctly design the experimental set-up. This model was completed by a feed-back algorithm to numerically predict the optimal temporal shape to compensate the gain saturation process. We demonstrated experimental results in good agreement with numerical simulations, with the following performances: 0.5 mJ pulse energy, at a frequency repetition from 1 kHz to 10 kHz, with a narrow bandwidth spectrum centred at 1053 nm wavelength, with 10 ns pulse duration on a perfect super-Gaussian temporal profile, an optical signal-to-noise ratio better than 50 dB and a polarization extinction ratio of 20 dB. We checked that the beam quality was diffraction limited, with an M² measurement of 1.1. Moreover, the system can deliver energies up to 1.5 mJ. Then, we took the advantage of such results to amplify chirped pulses. We demonstrated 0.7 mJ pulse energy, with 570 fs duration at 10 kHz repetition frequency.

No description available.

Structural elements in buildings and civil engineering infrastructure can often be vulnerable to various kinds of impact actions during their service life. These actions could result from various sources e.g. collision of vehicles, ships and vessels or falling masses in industrial buildings. Since, for various reasons, such accidental actions have not always been considered in the existing engineering design of buildings and civil engineering structures such as bridges etc., investigation of effective structural strengthening techniques is justified. As fibre reinforced polymer (FRP) composites have commonly been employed efficiently to strengthen steel members against static and fatigue loads, examining the FRP strengthening technique to enhance structural steelwork in impact situations is the main focus of this study. The research aims to experimentally investigate the dynamic behavioural response of axially loaded steel columns and steel beams strengthened with various carbon fibre reinforced polymer (CFRP) configurations. To achieve this goal, a series of experimental tests was implemented including testing a number of CFRP strengthened and unstrengthened steel beams and columns under static and impact loads. The experimental results show that CFRP can improve the global and local behaviour of steel members subjected to impact loads. This improvement varied depending on the CFRP configuration, the amount of CFRP and the pre-existing axial load value in the member. In order to examine all the parameters that can affect the dynamic behaviour of CFRP strengthened steel members in addition to those not included in the experimental programme, a comprehensive numerical simulation of the experimental work was carried out using a validated finite element model. Afterwards, an extensive parametric study was conducted to provide a comprehensive understanding of the behaviour of CFRP strengthened steel members subjected to impact load. The simulation results illustrate that the effectiveness of CFRP increases with high impact energies. The parametric study results have also revealed that the configurations and distributions of CFRP have a major influence on the effectiveness of the reinforcement. A detailed numerical assessment has also been performed to find the CFRP effectiveness when applied to full-scale steel columns. It has been found that strengthening with CFRP in practical quantities and configurations could prevent steel columns from failure under transverse impact loading. The strengthening effectiveness was found to be dependent on boundary conditions, impact velocity, impact mass, impact location, preloading level, impact direction, CFRP configuration, and the length and thickness of the CFRP. Based on the results obtained from the full-scale simulation, it has been found that the CFRP strengthening technique can be used efficiently and effectively at the scale of elements common in everyday building and infrastructure. This study also provides a useful database for different kinds of strengthening configurations, impact velocities and masses, boundary conditions, etc.

This thesis introduces low-cost implementations for the next generation distributed antenna system (DAS) using analogue radio over fibre. A multiple-input-multiple-output (MIMO) enabled radio over fibre (RoF) system using double sideband (DSB) frequency translation system is proposed. In such a system, the 2x2 MIMO signals can be transmitted to the remote antenna units (RAUs) from the base station via a single optical link. By using the DSB frequency translation, the original single-input-single-output (SISO) DAS can be upgraded into the MIMO DAS without implementing parallel optical links. Experimentally, the DSB frequency translation 2x2 MIMO RoF system transmits 2x2 LTE MIMO signals with 20MHz bandwidth in each channel via a 300m MMF link. The condition number of the system is < 10dB within the power equaliser bandwidth which means the MIMO system is well-conditioned and the crosstalk between the channels can be compensated by the MIMO signal processing. To install the DSB frequency translation system in a wideband service-agnostic DAS, the original MIMO signals need to be translated into unoccupied frequency bands over the DAS, which are usually occupied by specific applications that are not to be transmitted over the DAS. The frequency spectrum allocation of the wireless services is analysed showing that by choosing a particular LO frequency (2.2GHz in the UK), in the DSB frequency translation system, the original MIMO signals can always be translated into unoccupied frequency bands so that the same infrastructure can support multiple services. The idea of DSB frequency translation system can not only support MIMO radio over fibre but can also improve the SFDR of a general radio over fibre system. Because when the upper sideband and the lower sideband of the signal after translation are converted back to the original frequency band, the noise adds incoherently but the signals add-up coherently, this gives the system theoretically 2dB 3rd order SFDR improvement. If the idea of the DSB frequency translation is extended into a higher number of sidebands, the system SFDR can be further improved. Experimentally, the system 3rd order SFDR can be improved beyond the intrinsic optical link by 2.7dB by using quadruple sideband (QSB) frequency translation. It means the optical bandwidth in a general RoF system can be traded for the electrical SFDR. By integrating the analogue and the digital RoF systems, a hybrid DAS has been demonstrated, showing that the EVM dynamic range for the 4G LTE service (using digital RoF link) can be improved to be similar to the 3G UMTS service (using analogue RoF link), so that fewer number of RAUs for the LTE services are needed.

A method of predicting delamination in fibre-reinforced composite materials including several previously disregarded strength issues is presented. Thermal residual stresses, volume of stressed material, in-plane stresses and the hydrostatic stress in the polymer matrix are introduced and their influence on composite material strength discussed. These factors are then applied in a stress based method for predicting delamination which can deal with both unidirectional and general laminates. The results from a series of scaled unidirectional specimens designed to produce interlaminar tensile strength data are used to determine the strength parameters for the method. The method is shown to be effective in predicting failure in the fill-in region of two 'T'- piece specimen designs to within 14%. The failures were dominated by tension acting between fibres in large blocks of unidirectional material which had high thermal residual stresses and tensile hydrostatic stress due to constraint from the surrounding material. The method is also applied to a series of test pieces which used general laminates. The designs are based on sandwich panel sections and a tapered I-beam specimen. In the sandwich panel specimens, the edge closure sections were constructed using 0,90 and ±45° plies. Delamination occurred in a region of dropped plies and curvature making all the stress components important in producing accurate predictions, which are within 16% of the failure loads in testing. The tapered I-beam specimens were designed to delaminate in a doubly-curved laminate region of 90 and ±45° plies. The delamination predictions were within 13% of the first delamination loads found in testing. The method produced failure predictions which were all within 16% of the failure loads of the tested specimens. It is found that the local geometry of the delamination region is critical in determining the stress levels in the specimens and therefore their strength. Variations in the manufacture of such specimens and components is therefore clearly important in establishing the delamination loads of composite structures.

Ces travaux concernent la modélisation et la réalisation de fibres optiques micro-structurées, et plus particulièrement de fibres à bandes interdites photoniques actives et passives, à grande aire effective et destinées au transport ou à la génération de faisceaux lasers puissants.Une première partie du travail a porté sur l’étude d’une nouvelle géométrie de fibre micro-structurée - baptisée « fibre de Bragg pixélisée » - étudiée pour l’obtention d’un large cœur, monomode en pratique. Pour cette géométrie la fibre est rendue monomode en ajustant de façon optimale les distances entre les anneaux de haut indice de réfraction (condition dite demi-onde). Une première réalisation a permis de démontrer un diamètre de mode de 26μm à la longueur d'onde 1400nm dans une fibre passive. Un second aspect de ce travail a consisté en des études théoriques et expérimentales menées sur des fibres à bandes interdites photoniques présentant une gaine hétéro-structurée. Dans ces structures, la gaine comporte des résonateurs conçus pour éliminer les modes d’ordre supérieur par filtrage par les pertes. Des diamètres de mode allant de 19μm à 65μm ont ainsi été obtenus en régime monomode à 1050nm dans plusieurs fibres passives utilisées dans des bandes interdites photoniques différentes. Une fibre hétéro-structurée active a également été réalisée: le cœur, en silice pure dopée avec des ions ytterbium, a été obtenu via le procédé Sol-Gel. La fibre issue de cette réalisation a permis l’observation d’un effet laser avec une efficacité de 62.5%, pour un mode présentant un diamètre de 36μm. / These works concern the design and realization of micro-structured optical fibers, in particular, large mode area, active and passive, photonic bandgap fibers for high power laser beams generation and delivery. The first part of the work focused on the study of a new geometry of micro-structured fiber - so called "pixilated Bragg fiber" - in order to obtain a large, practically singlemode, core. For that geometry, the fiber is made singlemoded by optimizing the distances between the high index rings (Half wave stack condition). A first realization allowed to report a mode field diameter of 26μm measured at 1400nm wavelength in a passive fiber. The second aspect of this work included theoretical and experimental studies, of photonic bandgap fibers having a hetero-structured cladding. Specially designed resonators are added to the cladding of these fibers in order to eliminate higher order modes. Thus, 19μm to 65μm mode field diameters have been obtained in a singlemode regime at 1050nm wavelength for several passive fibers used in different bandgaps. An active fiber with hetero-structured cladding was also presented: the core was made of pure silica, ytterbium doped, synthesized using the Sol-Gel technique. The realized fiber allowed the observation of a laser emission with an efficiency of 62.5% and a mode field diameter of 36μm.