The principal objective of this work was to elucidate the effect of microstructural features on the intrinsic dislocation mechanisms in two FCC alloys. First alloy Al0.1CoCrFeNi was from a new class of material known as complex concentrated alloys, particularly high entropy alloys (HEA). The second was a conventional Al-Mg-Sc alloy in ultrafine-grained (UFG) condition. In the case of HEA, the lattice possess significant lattice strain due to the atomic size variation and cohesive energy differences. Moreover, both the lattice friction stress and the Peierls barrier height are significantly larger than the conventional FCC metals and alloys. The experimental evidences, so far, provide a distinctive identity to the nature and motion of dislocations in FCC HEA as compared to the conventional FCC metals and alloys. Hence, the thermally activated dislocation mechanisms and kinetics in HEA has been studied in detail. To achieve the aim of examining the dislocation kinetics, transient tests, both strain rate jump tests and stress relaxation tests, were conducted. Anomalous behavior in dislocation kinetics was observed. Surprisingly, a large rate sensitivity of the flow stress and low activation volume of dislocations were observed, which are unparalleled as compared to conventional CG FCC metals and alloys. The observed trend has been explained in terms of the lattice distortion and dislocation energy framework. As opposed to the constant dislocation line energy and Peierls potential energy (amplitude, ΔE) in conventional metals and alloys, both line energy and Peierls potential undergo continuous variation in the case of HEA. These energy fluctuations have greatly affected the dislocation mobility and can be distinctly noted from the activation volume of dislocations. The proposed hypothesis was tested by varying the grain size and also the test temperature. Activation volume of dislocations was a strong function of temperature and increased with temperature. And the reduction in grain size did not affect the dislocation mechanisms and kinetics. This further bolstered the hypothesis. The second part deals with deformation characteristics of Al-Mg-Sc alloy. The microstructure obtained from the severe plastic deformation (SPD) techniques differ in dislocation density, grain/cell size, and in the grain boundary character distribution. Therefore, it is vital to understand the deformation behavior of the UFG materials produced by various SPD techniques, as the microstructural features basically control the deformation mechanisms. In this study, a detailed analysis was made to understand the deformation mechanisms operative in various regimes of a stress-strain in UFG Al-Mg-Sc alloy produced via friction stir processing. The stress-strain curves exhibited serrations from the onset of yielding to the point of sample failure. The serration amplitude and frequency was higher in UFG material as compared to CG material. Furthermore, the microstructural features that result in the serrated flow were investigated along with the avalanche characteristics. The presence of both ultrafine grains and Al3Sc precipitates were the necessary conditions to reach the critical stress required to push the grain boundary into a critical state to set off an avalanche. The microstructural conditions that did not satisfy both the requirements did not exhibit deep serrations.

The strain rate dependent behaviour of fine grained soils is an important aspect of geotechnical engineering. During dynamic or rapid events such as earthquakes and rapid pile testing, a fine grained soil will display significantly different behaviour than may be observed over the long life span of a structure. There is currently little understanding of the factors which influence the behaviour of fine grained soils during dynamic events (extremely high strain rates), making their response difficult to predict. This research investigates the behaviour of fine grained soils subjected to a wide range of constant strain rates in monotonic triaxial compression testing. Each test is conducted under drained conditions to observe the behaviour of soils as they transition from a drained response at lower strain rates, through to an undrained or viscous response at higher strain rate tests. Where the response of soils is drained or partially drained, higher strain rate tests measure a decrease in strength. The point of transition from partially drained to undrained behaviour corresponds to the lowest strain rate dependent strength. Further tests at higher strain rates measure consistently greater strength. The strain rate dependence of three fine grained soils is investigated, enabling a comparison of strain rate effects with soil index properties. The influence of initial state on the strain rate dependence of these Kaolin based model soils is also evaluated. The drained to partially drained response of the soils to strain rate increase is controlled by the coefficient of consolidation. Tests at high strain rates show the undrained or viscous strain rate effect on strength is related to liquidity index. Local strain instrumentation allowed comparison of strain rate effects on small strain stiffness. At higher strain rate the soils display increasingly linear behaviour. At non-linear elastic strains, liquidity index appears to control the magnitude of the strain rate effects on stiffness.

<p>Computation-intensive algorithms require a high level of parallelism and programmability, which </p><p>make them good candidate for hardware acceleration using fine-grained processor arrays. Using </p><p>Hardware Description Language (HDL), it is very difficult to design and manage fine-grained </p><p>processing units and therefore High-Level Language (HLL) is a preferred alternative. </p><p> </p><p>This thesis analyzes HLL programming of fine-grained architecture in terms of achieved </p><p>performance and resource consumption. In a case study, highly computation-intensive algorithms </p><p>(interpolation kernels) are implemented on fine-grained architecture (FPGA) using a high-level </p><p>language (Mitrion-C). Mitrion Virtual Processor (MVP) is extracted as an application-specific </p><p>fine-grain processor array, and the Mitrion development environment translates high-level design </p><p>to hardware description (HDL). </p><p> </p><p>Performance requirements, parallelism possibilities/limitations and resource requirement for </p><p>parallelism vary from algorithm to algorithm as well as by hardware platform. By considering </p><p>parallelism at different levels, we can adjust the parallelism according to available hardware </p><p>resources and can achieve better adjustment of different tradeoffs like gates-performance and </p><p>memory-performance tradeoffs. This thesis proposes different design approaches to adjust </p><p>parallelism at different design levels. For interpolation kernels, different parallelism levels and </p><p>design variants are proposed, which can be mixed to get a well-tuned application and resource </p><p>specific design.</p>

Object categorization is one of the fundamental topics in computer vision research. Most current work in object categorization aims to discriminate among generic object classes with gross differences. However, many applications require much finer distinctions. This thesis focuses on the design, evaluation and analysis of learning algorithms for fine- grained object classification. The contributions of the thesis are three-fold. First, we introduce two databases of high-resolution images of arthropod specimens we collected to promote the development of highly accurate fine-grained recognition methods. Second, we give a literature review on the development of Bag-of-words (BOW) approaches to image classification and present the stacked evidence tree approach we developed for the fine-grained classification task. We draw connections and analyze differences between those two genres of approaches, which leads to a better understanding about the design of image classification approaches. Third, benchmark results on our two datasets are pre- sented. We further analyze the influence of two important variables on the performance of fine-grained classification. The experiments corroborate our hypotheses that a) high resolution images and b) more aggressive information extraction, such as finer descriptor encoding with large dictionaries or classifiers based on raw descriptors, is required to achieve good fine-grained categorization accuracy. / Graduation date: 2013

Computation-intensive algorithms require a high level of parallelism and programmability, which make them good candidate for hardware acceleration using fine-grained processor arrays. Using Hardware Description Language (HDL), it is very difficult to design and manage fine-grained processing units and therefore High-Level Language (HLL) is a preferred alternative. This thesis analyzes HLL programming of fine-grained architecture in terms of achieved performance and resource consumption. In a case study, highly computation-intensive algorithms (interpolation kernels) are implemented on fine-grained architecture (FPGA) using a high-level language (Mitrion-C). Mitrion Virtual Processor (MVP) is extracted as an application-specific fine-grain processor array, and the Mitrion development environment translates high-level design to hardware description (HDL). Performance requirements, parallelism possibilities/limitations and resource requirement for parallelism vary from algorithm to algorithm as well as by hardware platform. By considering parallelism at different levels, we can adjust the parallelism according to available hardware resources and can achieve better adjustment of different tradeoffs like gates-performance and memory-performance tradeoffs. This thesis proposes different design approaches to adjust parallelism at different design levels. For interpolation kernels, different parallelism levels and design variants are proposed, which can be mixed to get a well-tuned application and resource specific design.

Organic materials are ubiquitous in the geologic environment, and can exert significant influence over the interfacial properties of minerals. However, due to the complexity in their structure and interaction with soil solids, their impact has remained relatively unquantified. This study investigated the engineering behaviors of organoclays, which were synthesized in the laboratory using naturally occurring clay minerals and quaternary ammonium compounds of controlled structure and density of loading. Organic cations were chosen to study the effects of functional group structure and size. The laboratory investigation showed that the presence of the organic cations on the mineral surfaces led to increased hydrophobicity of all clays tested. Conduction studies on the electrical, hydraulic, and thermal properties of the organoclay composites suggested that increasing the total organic carbon content resulted in decreased electrical and thermal conductivity, but increased hydraulic conductivity, due to the reduced swelling of the base clay mineral phase. Electrokinetic properties of the organoclays illustrated that compared with the clay's naturally occurring inorganic cations, exchanged quaternary ammonium cations were more likely bound within a particle's shear plane. Consequently, organoclays had less negative zeta potential than that of unmodified bentonite. Increasing the length of one carbon tail was more effective at binding organic cations within the shear plane than increasing the size of the cation, when compared on the basis of total organic carbon content. In terms of large strain strength, the modified organic clays exhibited increased shear strength, in part owing to the reduction in water content caused by the presence of the hydrophobic organic layering. Shear strength increased with single carbon tail length or with cation size, although the latter effect tended to reach a plateau as the length of the four short cation tails increased from 2 to 4. In terms of small strain behavior, the shear modulus was shown to be a function of the total organic carbon content. It is believed that number of particle contacts increased as the organic carbon content increased. Stiffness increased as either the size of the cation or the total organic carbon content was increased. Damping also increased as the organic loading was increased, with the organic phase acting as an energy dissipation mechanism.

Prediction as a tool in engineering has been used in taking right judgement in many of the professional activities. This being the fact, the role and significance of prediction in geotechnical practice needs no emphasis. Bulk of all man made structures are either made of soil or are resting on natural soil, involving large quantities of soil. Thus, it is often necessary for the geotechnical engineer to quickly characterize the soil and determine their engineering properties, so as to assess the suitability of the soil for any specific purpose. Obtaining these properties requires undisturbed samples, which involves time and money, and also elaborate laboratory procedures. Thus, it is desirable to find simpler and quicker methods of testing, using the data of which the engineering properties can be predicted satisfactorily especially so, for preliminary design purposes. Most often this can be achieved from simple tests known as inferential tests, and the engineering properties namely, compressibility, swell/collapse, hydraulic conductivity, strength and compaction characteristics can be obtained from empirical/semi-empirical correlations. The index tests namely the Atterberg limits form the most important inferential soil tests with very wide universal acceptance. These tests are relatively simple to perform and have provided a basis for explaining most engineering properties of soils in geotechnical practice. In this direction, this investigation has been carried out to correlate the engineering properties with the simple index properties and their indices, namely, the liquid limit, plastic limit, shrinkage limit, plasticity index and shrinkage index (liquid limit - shrinkage limit). Any good correlation in the prediction of engineering properties with the index properties will enhance the use of simple test for prediction purposes. This thesis is an attempt towards this direction. It is often necessary to identify the basic mechanisms controlling the engineering properties from a micro-mechanistic point of view and correlate with the index properties, thereby facilitating prediction of engineering properties better. Though attempts have been made in the past to predict the engineering properties of soils from the index properties/indic­es, they are not quite satisfactory. This thesis is an attempt to predict the engineering properties of fine-grained soils from the index properties taking into consideration the mechanisms controlling them. Since, the index properties are used for prediction of engineering properties, the existing methods of determining the same have been examined carefully and critically. It's satisfactory determination is found important because other indices namely plasticity index, Ip and shrinkage index, Is = (wL - ws), are determined based on it. Also the liquid limit is one of the important and widely used parameter in various existing correlations. In this direction, two new methods of determining the liquid limit have been developed, namely (i) absorption water content and liquid limit of soils and (ii) liquid limit from equilibrium water content under Ko-stress. In the absorption water content method, the water absorbed by an oven dried soil pat at equilibrium gives a good correlation with the liquid limit of soils. Here, the water holding capacity at equilibrium goes well with the mechanism of liquid limit, which is also the water holding capacity of a soil at a particular small but measurable shear strength. A good relationship is found to exit between the absorption water content, wA and the liquid limit, wL, and it is given as : WA = 0.92 wL (i) In the second method, namely, the liquid limit from equilibrium water content under K0-stress, which is the equilibrium water content under a Ko stress of 0.9 kPa is found to be equal to the liquid limit obtained from the cone penetration method of determining the liquid limit It is found that this method of determining the liquid limit overcomes the limitations of the conventional methods of determining the liquid limit, also easy to determine with a simple apparatus and has good repeatability. Determination of plastic limit of the soils by the rolling thread method often poses a problem especially when the soil is less plastic. Hence, to overcome this problem, a new method has been proposed to predict the plasticity index in terms of the flow index. The relationship between the plasticity index and the flow index by the cone penetration cup method is found to be better than by the percussion cup method. Since, the cone penetration method of the liquid limit determination is more popular than the percussion cup method, the flow index from the cone method is recommended to determine the plasticity index from the correlation as given below: (/p)c = 0.74 Ifc (ii) Thus, the plastic limit can be determined with the plasticity index, thereby dispensing with the determination of plastic limit by the thread method. The determination of consolidation characteristics form an important aspect in the design of foundations and other earth retaining structures. The determination of consolidation characteristics namely the compression index, the coefficient of consolidation and the coefficient of secondary compression is time consuming. So, researchers have resorted to correlating the compressibility behaviour with simple index properties. While attempts have been made in the past to correlate the compressibility behaviour with various index properties individually, all the important properties affecting the compressibility behaviour has not been considered together in any single study to examine which of the index property/properties of the soils correlates better with the compressibility behaviour, especially with the same set of test results. Number of existing correlations with the liquid limit alone as a primary index property correlating with the compression index have limitations in that they do not consider the plasticity characteristics of the soils fully. The index parameter, shrinkage index, Is has a better correlation with the compression index, Cc and also the coefficient of volume change, mv than plasticity index. Coefficient of consolidation, Cv has also shown to correlate well with shrinkage index than the plasticity index. Even the coefficient of secondary compression, Cαε has shown to have a better correlation with shrinkage index than the plasticity index. However, liquid limit has a poor correlation with all the compressibility characteristics. The correlation of Cc and Cv with shrinkage index, Is is as given below: Cc = 0.007 (Is + 18) (iii) Cv = 3x10-2 (Is)-3.54 (in m2/sec) Further, to reduce the testing time of conventional consolidation test in order to obtain the compressibility characteristics, a new method known as rapid method of consolidation has been proposed, which is very effective in enormously reducing the time of consolidation without sacrificing the accuracy of the end results. The time required in the rapid method of consolidation testing could be as low as 4 to 5 hours to complete the whole test as compared to 1 to 2 weeks as the case may be by the conventional consolidation test. Using any curve fitting procedure the degree of consolidation, U for any pressure increment can be found out. Thus, the effective pressure at that stage can be calculated and further the pressure incremented without further delay. This procedure is repeated for every pressure increment with a load increment ratio of unity till the desired pressure level is reached. Even for a highly compressible soil like BC soil with a liquid limit of 73.5 %, the consolidation test could be completed within 5 hours by the rapid method, without any sacrifice of the accuracy of the results as compared to 7 days by the conventional method to reach a pressure of 800 kPa. Hydraulic conductivity is one of the basic engineering properties of soils. Of late hydraulic conductivity of fine-grained soils has assumed greater importance in waste disposal facilities. From the present investigation it is found that hydraulic conductivity with water for each pair of soils having nearly the same liquid limit but different plasticity properties is found to be vastly different, but found to correlate well with shrinkage index. A method to predict the hydraulic conductivity of fine -grained soils as a function of void ratio is proposed with the use of shrinkage index as given below: k = C [ ] (in m/sec) (v) 1 + e C = 2.5 x 10-4 (/s)-5.89 and n = 4 (vi) It has also been brought out that as the dielectric constant of the pore fluid decreases; there is a drastic increase in the intrinsic permeability of soil. These changes are attributed to the significant reduction in the thickness of diffuse double layer, which in turn is mainly dependent on the dielectric constant of the pore fluid. The quantification of the change in the hydraulic conductivity with the change in the pore fluids of extreme dielectric constant, i.e., from water to carbon tetrachloride could be expressed in terms of the volume of water held in the diffuse double layer and the same has a good correlation with shrinkage index. With the advancement in the knowledge of the engineering behaviour of fine-grained soils, there is an increasing trend toward larger involvement of fine-grained soils in earth structures and foundations. Though extensive work has been done in the past to understand the swelling behaviour of expansive soils and the mechanisms involved therein, it is yet not satisfactory. From the literature it can be seen that lot of work has been done to correlate the swell potential with various physical properties. The simple means of identifying the swelling type of soils is by means of free swell tests with the ratio of free swell with carbon tetrachloride to the free swell of water. The same has found to correlate well with the percent swell/collapse of the ten soils used in the present investigation. However, it was found that shrinkage index has a better correlation with the swell/collapse behaviour of fine-grained soils, compared to the liquid limit or the plasticity index. In this study, it is also shown that neither the liquid limit nor the plasticity index can qualitatively describe the swell/collapse behaviour of fine-grained soils. This has been attributed primarily to two different mechanisms governing montmorillonitic and kaolinitic soils separately. Even swelling pressure has shown to have a good correlation with shrinkage index. It is found that the compression index of the samples consolidated from the swollen condition correlates well with the shrinkage index. Laboratory determination of the compaction characteristics are very much important for use in earth work constructions. It is found that only the plastic limit bears a good correlation with the compaction characteristics namely optimum moisture content and maximum dry unit weight. This conclusion is also supported by the data from the literature. The correlations are given as: OMC = 0.92 wp (in percent) (viii) and ydmax = 0.23 (93.3 - wp) (inkN/m3) (ix) Liquid limit, plasticity index and shrinkage index do not bear any correlation with the compaction characteristics. It is quite possible that, the plastic limit, which is the optimum water content of a saturated soil at which it behaves as a plastic material, and thus can be moulded to any shape, thereby the soil can be compacted or moulded to the densest possible state at that water content. Hence, possibly the good correlation. A simple method to predict the compaction curve is proposed based on the plastic limit of the soils. Of all the important engineering properties, both volume change (compressibility and swelling) and hydraulic conductivity have good correlation with the shrinkage index. However, the compaction characteristics correlate well with the plastic limit. Herein, an hypothesis is proposed to possibly explain why shrinkage index has shown to be a better parameter to correlate with most of the engineering properties with the exception of the compaction characteristics. The liquid limit is a parameter which takes part of the plasticity characteristics of a soil. Recently it has been well brought out that shrinkage limit is primarily a function of how the varying grain sizes are distributed in a soil. Thus, shrinkage limit takes care of the gradation of the soil fractions in it. Thus, by considering the shrinkage index, which is the difference of the liquid limit water content on one end and shrinkage limit water content on the other end, the primary physical properties of the soils namely the plasticity and the grain size distribution are considered. This possibly explains the good correlation of shrinkage index with the engineering properties of fine-grained soils. However, compaction being a moulding of the soils into a compact state, it has a good correlation with the plastic limit, which is the optimum water content of a saturated soil at which it behaves as a plastic material, and thus can be moulded to any shape, thereby the soil can be compacted or moulded to the densest possible state at that water content. Hence, the good correlation. As the present investigation gives the correlative equations to predict the engineering properties of fine-grained soils from the appropriate index properties, which are obtained from simple and quick laboratory tests, it is hoped that this will go a long way in being a handy tool for a practicing geotechnical engineer in the preliminary assessment of fine-grained soils and thereby take appropriate judgement in various aspects of geotechnical constructions with it.

During high flow rates, the acceleration of flow and turbulence around bridge foundations lead to scouring, defined as the removal of bed sediments. Due to the interparticle physico-chemical forces of clay particles, erodibility and transport mechanisms for fine sediments are different from those for coarse sediments, and the capability to predict the erosion resistance of fine sediments is still in question. In this study, silt-clay soil mixtures with different kaolin contents were prepared by mixing ground silica and Georgia kaolin with tap water. Geotechnical tests were carried out to obtain the physical properties of the specimens. The critical shear stress and yield stress of the soil mixtures were determined through hydraulic flume experiments and rheometer tests, respectively. Particle associations of the soil specimens were observed using the technique of scanning electron microscopy (SEM). From the laboratory work and data analysis, relationships among the critical shear stress, yield stress, and the soil physical properties were developed from multiple regression analysis. Specifically, values of the critical shear stress, yield stress, and their dimensionless form can be predicted by the soil properties including bulk density, clay content, and water content. Finally, a single relationship is obtained to predict the Shields parameter as a function of the corresponding dimensionless yield stress in this study. The results can be used to provide a methodology for engineering applications requiring the value of critical shear stress such as estimating fine sediment bed stability and assessing the erosion risk of river beds in proximity to bridge foundations and other flow obstructions.

Untersuchungen zur Dauerhaftigkeit von ARGlasbewehrung im Textilbeton haben gezeigt, dass durch die Alkalität des Betons in Verbindung mit Feuchtigkeit eine Glaskorrosion hervorgerufen wird, die im Laufe der Zeit zu Festigkeitsverlusten des Glases führt. Eine Möglichkeit, die durch die Glaskorrosion verursachten Festigkeitsverluste zu reduzieren, stellt die Polymermodifikation des Betons dar. Durch die Polymerzugabe wird die Wasseraufnahme der Feinbetonmatrix reduziert, dadurch sinkt der Gehalt an gelösten Alkalien im Bereich der Bewehrung. Um den Einfluss verschiedener Feinbetonmatrices auf die Dauerhaftigkeit von Textilbeton beurteilen zu können, sind u. a. zeit- und tiefenabhängige Informationen zur Feuchteverteilung erforderlich, die durch den Einsatz der NMR-Technik gewonnen werden. Der nachfolgende Artikel beschreibt den Feuchtetransport in einer speziell für den Textilbeton entwickelten Feinbetonmatrix sowie den Einfluss verschiedener Modifikationsstoffe auf das Wasseraufnahmeverhalten des Betons. / Durability tests of textile reinforced concrete revealed a loss of strength of the AR-glass reinforcement due to glass corrosion effected by the alkalinity and moisture content of the concrete. In order to reduce this strength loss of AR-glass in cementitious matrices, polymers can be used for concrete modification. The aim of the polymer addition is to reduce the amount of capillary water absorption of the matrix, which reduces the amount of free alkalies closed to the reinforcement. In order to evaluate the effect of the concrete matrix on the durability of TRC, it is necessary to determine the moisture content as functions of time and depth. This data can be obtained by the use of nuclear magnetic resonance (NMR) technique. This paper deals with the moisture transport in a finegrained concrete matrix especially developed for the use in TRC as well as the influence of polymer addition on the water absorption properties of the concrete matrix.

Die Einhaltung von Brandschutzanforderungen ist ein wichtiger Aspekt für sichere Baukonstruktionen. Beim innovativen Werkstoff textilbewehrter Beton, der einen Verbundwerkstoff aus einer Feinbetonmatrix und textiler Bewehrung darstellt, ist das Brandverhalten bisher nur unzureichend erforscht worden. Insbesondere das Tragverhalten der einzelnen Komponenten unter hohen Temperaturen stellt noch eine Wissenslücke in der heutigen Forschung dar. Dieser Artikel befasst sich mit den experimentellen Untersuchungen an einer Feinbetonmatrix, die ein Größtkorndurchmesser von 0,6 mm aufweist, sowie an AR-Glas- und Carbongarnen. Basierend auf instationären Versuchen werden das Spannungs- und Dehnungsverhalten unter hohen Temperaturen abgeleitet und Ansätze zur rechnerischen Beschreibung des Hochtemperaturverhaltens vorgeschlagen. / The design of structural members under fire attack is an important aspect for safe constructions. For the innovative material textile reinforced concrete (TRC), which is a composite material made of fine-grained concrete and textile reinforcement, the fire behavior has not been investigated insufficiently yet. Especially the load-bearing behavior under high-temperatures of the single components marks a gap in the state-of-the-art of science and technology today. This article deals with experimental investigations on a fine-grained concrete matrix, which has maximum grain size of only 0.6 mm, as well as yarns made of AR-glass and carbon. On the basis of transient tests the stress and strain behavior under high temperatures is derived. Finally, a calculative approach for the hightemperature behavior is presented.