In this study, the effects of submergence on local scour in a uniform cohesionless sediment bed by a plane turbulent wall jet and the resulting flow field were investigated experimentally. Here, submergence is defined as the ratio of the tailwater depth to the thickness of the jet at its origin. The main focus was to determine scour dimensions at an asymptotic state, examine whether there was similarity in the velocity profiles for the flow in the scour hole, and to determine the growth of the length scales and decay of the maximum velocity of the jet. Also examined were the relationships between the scales for the velocity field in the scour hole and the scour hole size.<p>In the experiments, the range of submergence was varied from 3-17.5, whereas the range of densimetric Froude number and the ratio of the boundary roughness to the gate opening (relative boundary roughness) were varied from 4.4-6.9 and 0.085-0.137 respectively. The velocity field in the scour hole at asymptotic state was measured using a SonTek 16-MHz MicroADV. Time development of the characteristic dimensions of the scour hole was also measured.<p>The dimensions of the scour hole were found to increase with increasing submergence for all experiments with a bed-jet flow regime. In the bed-jet flow regime, the jet remains near the bed throughout the scouring process. Further, the time development of the scour hole dimensions were observed to increase approximately linearly with the logarithm of time up to a certain time before the beginning of asymptotic state for experiments with either the bed-jet or surface-jet flow regimes.<p> The flow field results showed that the velocity profiles in the region of forward flow and the recirculating region above the jet were similar in shape up to about the location of the maximum scour depth. Relationships describing this velocity profile, including its velocity and length scales, were formulated. The decay rate of the maximum velocity, the growth of the jet half-width, and the boundary layer thickness were also studied. The decay and the growth rate of the jet length scales were found to be influenced by the submergence ratio, densimetric Froude number, and the relative boundary roughness.<p>Two distinct stages in the decay of the maximum streamwise velocity, with distance along the direction of flow, were observed for the jet flows having a bed-jet flow regime. The first stage of velocity decay was characterized by a curvilinear decay of velocity, which followed that of a wall jet on a smooth, rigid bed for streamwise distance approximately equal to 2L. For the surface-jet flow regime, the decay of velocity was observed to be similar to that of a free-jump on a smooth, rigid bed for a streamwise distance approximately equal to L. Here, L is defined as the streamwise distance measured from the end of the rigid apron to where the maximum streamwise velocity in the jet is half the velocity of the jet at the end of apron. The streamwise maximum velocity of the jet was then seen to increase in what was called the recovery zone.<p>A relationship for the streamwise decay of the maximum velocity within the scour hole is proposed. Moreover, other scales representing the flow inside the scour hole such as the streamwise distance from the end of the apron to where the streamwise maximum velocity starts to deviate from curvilinear to linear decay and the streamwise distance to where maximum streamwise velocity starts to increase are suggested. Some new results on the velocity distribution for the reverse flow for a bed-jet flow regime are also presented. Finally, some dimensionless empirical equations describing the relationship between the jet scales for the jet flow in a scour hole and the scour hole size are given.

An efficient Light Emitting Diode (LED) lamp driver circuit is proposed for retrofitting the conventionally used incandescent lamps with existing TRIAC dimmer. The dimming feature in a wide range of firing angle from 30¢X to 130¢X can be accomplished by means of double pulse-width modulation (DPWM) and analog current regulation. The LED lamp driver adopts a flyback converter with an auxiliary active power MOSFET for synchronous switch and an associated inductor for zero voltage switching (ZVS), leading to lower switching loss and thus achieving a higher circuit efficiency. In the thesis, the mode operation of the driver circuit is analyzed and the design equations are derived accordingly. A laboratory circuit is designed for an 50 W LED lamp which is composed of 45 high-brightness white LEDs in series. Experiments are carried out to test the circuit performances with two dimming schemes. The experimental results indicate that the driver can achieve a circuit efficiency of 95 % at the rated output. When the LED lamp is dimmed, the circuit efficiency with DPWM is higher than that with the analog current regulation. On the other hand, the LED lamp dimmed by analog current regulation has a higher efficiency but a less color shift by DPWM.

In this thesis, the conventional pulse-width modulated (PWM) and zero-voltage switching (ZVS) quasi-resonant buck converters are analyzed and a variable-frequency control technique is proposed to regulate the output voltage due to the immediate input line and load changes. The quasi-resonant technique provides favorable switching conditions for active switch to reduce switching losses and electromagnetic interference (EMI). The method is based on shaping the voltage across the active switch in quasisinusoidal fashion and the switching action occurs with nearly zero voltage across the active switch. This requires only two additional components to the conventional PWM buck converter. The proposed quasi-resonant converter is capable of operating in megahertz range with a significant improvement in performance and power density. Detailed analytic and small-signal models of the ZVS quasi-resonant buck converter are established and the switching behavior is investigated in order to provide nearly zerovoltage turn-on. The performance of the ZVS quasi-resonant technique is verified with the computer simulations. The results are compared with the experiments in the laboratory involving both the open-loop and closed-loop operations. The detailed experiment procedure is added to use this converter for educational purposes.

The lepton flavor violating interactions are interesting in the sense that they are sensitive the physics beyond the standard model and they ensure considerable information about the restrictions of the free parameters, with the help of the possible accurate measurements. In this work, we investigate the lepton flavor violating H+ ! W+l and the lepton flavor conserving H+ ! W+l decays in the general two Higgs doublet model and we estimate decay widths of these decays. After that, we analyze lepton flavor violating decay ! i in the same model and calculate its branching ratio. We observe that the experimental results of the processes under consideration can give comprehensive information about the physics beyond the standard model and the existing free parameters.

This research work is devoted to the design, development and implementation of a Current Source Converter (CSC) based Active Power Filter (APF) for Medium Voltage (MV) applications. A new approach has been proposed to the design of the CSC based APF for reducing the converter kVA rating considerably. This design approach is called the Selective Harmonic Amplification Method (SHAM), and is based on the amplification of some selected harmoniccurrent components of the CSC by the input filter, and the CSC control system, which is specifically designed for this purpose. The proposed SHAM has been implemented on the first industrial CSC based APF for the elimination of 11th and 13th current harmonics of 12-pulse rectifiers fed from Medium Voltage (MV) underground cables in order to comply with IEEE Std. 519-1992. 450 kVA rated APF with only 205 kVA CSC rating has been connected to the MV bus via a coupling transformer of 600kVA, 34.5/1.1 kV. The power stage of the CSC based APF is composed of water-cooled high voltage IGBT and diode modules. Reference currents to be generated by the CSC are obtained by the use of a selective harmonic extraction method, by mploying synchronously rotating reference frames for each selected harmonic component. An Active damping method is also used to suppress the oscillations around the natural frequency of the input filter, excluding the harmonic components to be eliminated by APF. Simulation and field test results have shown that SHAM can successfully be applied to a CSC based APF for reduction of converter kVA rating, thus making it a cost- competitive alternative to voltage source converter based APFs traditionally used in industry applications.

Most of the reinforced concrete systems are monolithic. During construction, concrete from the bottom of the deepest beam to the top of slab, is placed at once. Therefore the slab serves as the top flange of the beams. Such a beam is referred to as T-beam. In a floor system made of T-beams, the compressive stress is a maximum over the web, dropping between the webs. The distribution of compressive stress on the flange depends on the relative dimensions of the cross section, span length, support and loading conditions. For simplification, the varying distribution of compressive stress can be replaced by an equivalent uniform distribution. This gives us an effective flange width, which is smaller than the real flange width. In various codes there are recommendations for effective flange width formulas. But these formulas are expressed only in terms of span length or flange and web thicknesses and ignore the other important variables. In this thesis, three-dimensional finite element analysis has been carried out on continuous T-beams under different loading conditions to assess the effective flange width based on displacement criterion. The formulation is based on a combination of the elementary bending theory and the finite element method, accommodating partial interaction in between. The beam spacing, beam span length, total depth of the beam, the web and the flange thicknesses are considered as independent variables. Depending on the type of loading, the numerical value of the moment of inertia of the transformed beam crosssection and hence the effective flange width are calculated. The input data and the finite element displacement results are then used in a nonlinear regression analysis and two explicit design formulas for effective flange width have been derived. Comparisons are made between the proposed formulas and the ACI, Eurocode, TS-500 and BS-8110 code recommendations.

In this study, leaf anatomies of five species of Limonium Mill. (L. echioides, L. globuliferum, L. tamaricoides, L. anatolicum and L. Sinuatum) are studied. Taca belonging to five different sections of the genus Limonium as / L. echioides of the section Schizyhymenium, L. globuliferum of the section Sphaerostachys, L. tamaricoides of the section Limonium, L. anatolicum of the section Sarcophyllum and L. sinuatum of the section Pteroclados are used in this study. Throughout the species studied, L. tamaricoides and L. anatolicum are endemic to Turkey. In order to examine anatomy of leaves, the paraffin sectioning and hand cross sectioning methods are used. Avarage stomata length, width and number per 210.68&micro / m x 263.27&micro / m area of the leaves are examined and the stomata types due to epidermal cells are defined. Additionally, the vascular bundles, upper and lower epiderms of leaves of each species are defined. It is determined that except the species L. globuliferum and L. tamaricoides, the stomata density per 210.68&micro / m x 263.27&micro / m area of the leaves are approximately same for upper and lower epidermis, but in these species, the stomata numbers at upper epidermis is higher than the stomata numbers in lower epidermis, which might be related with the altitude, microclimate and habitat of these two species. It is also found that except the shoot leaf of L. sinuatum, the phloem is closer to lower epiderm than xylem in vascular bundles. But in shoot leaf of L. sinuatum, the phloem circularly covers the xylem and makes a closed circle around.

Als Ergänzung für die Lehrbriefe des Instituts für Massivbau der TU Dresden zu den Grundlagen des Stahlbetons (Teil 1 und 3 sowie Übungen Teil 1 bis 3) werden hier Bilder von Versuchsreihen vorgestellt, die am Lehrstuhl für Stahlbeton, Spannbeton und Massivbrücken der TH/TU Dresden, aus dem das heutige Institut für Massivbau hervorging, innerhalb verschiedener Forschungsarbeiten von 1956 bis 1965 entstanden. Neben dem Einblick in die damaligen Arbeitsweisen und Möglichkeiten sind vor allem die zahlreichen Bruchbilder geeignet, sich in das Tragverhalten des Stahlbetons hineinzudenken. Diese Überlegungen gaben den Ausschlag dafür, das vorhandene Bildmaterial noch einmal zu ordnen und mit kurzen Erläuterungen zu versehen, um es so nochmals für Lehre und Forschung nutzbar zu machen.

Mit der vorliegenden Arbeit soll ein Beitrag geleistet werden, die lokalen Tragmechanismen eines Brückenhohlkastens im Bereich der Feldumlenkstellen (Feldlisenen) zu untersuchen und zu optimieren. Beim Entwurf einer Brücke gewinnt die Frage der Dauerhaftigkeit u.a. in Deutschland und der Schweiz eine immer zentralere und grössere Bedeutung. Dadurch können die laufenden Kosten reduziert werden, während ausserhalb dieser beiden Länder im Allgemeinen mehr Gewicht auf die rationelle Fertigung und die Kostenersparnis in der Bauphase gelegt werden. Die Dauerhaftigkeit von extern vorgespannten Brücken mit ihren relativ geringen Wartungs- und Unterhalts- bzw. Instandstellungskosten sind das Hauptargument für diese Bauweise. Bei Vorspannung ohne Verbund sind für externe Spannglieder grössere Verankerungskonstruktionen erforderlich, weil die Spannglieder nicht im Steg selbst, sondern mit einem Mindestabstand zum Steg in Lisenen vorgespannt werden müssen. Über diese Verankerungsstellen werden in Hohlkastenbrücken grosse Kräfte konzentriert in die Stege, Boden- und Fahrbahnplatten eingeleitet. Im Sinne von Gewichtsreduktion des Brückenbauwerks sollen die Feldlisenen möglichst schlank und Platz sparend ausgeführt werden. Nur durch eine sorgfältige Bemessung und Konstruktion der Verankerungs- und Umlenkstellen kann die Voraussetzung für sichere und dauerhafte Brückenbauwerke gewährleistet werden. Die Bruchsicherheit ist dadurch nicht beeinträchtigt. Von grösstem Interesse ist jedoch die Gebrauchstauglichkeit, das heisst die Rissbreiten mit Rücksicht auf Korrosionsschäden in der Bewehrung. Bei den ersten in Deutschland extern vorgespannten Brücken schlug Prof. Eibl für den Stahl III vor, die Spannungen von 240 N/mm2 auf 220 N/mm2 zu begrenzen. In der Richtlinie für externe Vorspannung von 1998 wurden die Spannungen weiter auf 180 N/mm2 reduziert. Um eine unkontrollierte Rissbildung in den hoch bewehrten Verankerungskonstruktionen zu verhindern, sind genaue Kenntnisse der Kraftflüsse und für die Praxis taugliche Bemessungsmodelle notwendig. Die Geometrie und die statische Berechnung der gewählten Umlenkkonstruktion müssen entsprechend gewichtet werden. In dieser Arbeit werden an zwei Vergleichsmodellen die lokalen Tragmechanismen einer extern vorgespannten Hohlkastenbrücke im Bereich der Feldumlenkstelle (Feldlisene) untersucht. Mit linear-elastischen und nichtlinearen numerischen Berechnungen wird abgeklärt, ob unter dem Ansatz einer Spannungserhöhung im Bewehrungsstahl auf 285 N/mm2, die Gebrauchstauglichkeit noch gewährleistet werden kann. Den Berechnungen wird eine Krafteinleitung von 2 x 3 MN mit Umlenkkräften von 4 x 0,64 MN zu Grunde gelegt. Die Berechnungen zeigen, dass das Ziel, mit einer Bewehrung von As = 3,35 % m2/m unter Gebrauchslasten keine Rissbreiten über 0,2 mm auftreten, erreicht werden kann. Die Berechnungen werden zudem an einem Brückenmodell im Massstab 1:1 in einem Belastungsversuch an der Empa in Dübendorf auf ihre Aussagekraft und Richtigkeit überprüft und bestätigt. / This present work which is aimed at contributing to local carrying mechanisms of a bridging box girder in the field of belt reversals, should be investigated and optimized. In designing a bridge the main concern is that of durability; especially in Germany and Switzerland, this is invariably acquiring a more central meaning. In this way recurring expenses can be reduced, whilst beyond these two countries in general more weight is laid on the rational manufacture and cost saving in the construction phases. The durability of externally pre-stressed bridges with their relatively low maintenance and service as well as repair costs are the main arguments for this method of construction. Larger anchorage construction is necessary for pre-tensioning external tensions. This is because the tensions are themselves not placed on the ligament itself, but must be pre-stressed in pilaster strips with a minimum distance. Over these anchorage locations immense energy is concentrated in the large box girder bridges, and thus induced in the ligament, ground and track supporting layers. As far as weight reduction of the bridging structure is concerned, the field pilaster strips should be designed as thin as possible and also be able to save space. Only through careful measuring and construction of the anchor and turning points can the requirements for safe and lasting bridge construction works be guaranteed. The bridge safety is thus not affected. However, the greatest interest is that of userfriendliness, i.e. the width of the fissure with consideration of corrosion damages in the armouring. At the first external pre-stressed bridge in Germany, Prof. Eibl suggested that for the steel III, stressing of 240/Nmm2 should be limited to 220 N/mm2. In the guideline for external pre-stressing of 1998, the tensions were further reduced to 180 N/mm2. In order to prevent an uncontrolled crack formation in the highly armoured anchorage construction, concrete knowledge of the distribution of forces for the applicable measurement models is necessary. The geometry and the statistical calculation of the selected deflect construction must be weighed accordingly. In this project two comparative models were inspected, which tested the local load mechanism of an external pre-stressed box girder bridge in the field redirecting area. With linear elasticity and numerical calculations it is possible to clarify if the serviceability can be allowed under the accretion of a stress increase in the armouring steel to 285 N/mm2. The calculations are based on a load transmission of 2 x 3 MN with a deviation force of 4 x 0,64 MN. The calculations will have a force transmission of 2 x 3 MN with turning forces of 4 x 0.64 MN forming the basis. The calculations portray that the goal to avoid any fissures over 0,2 mm can be achieved, with an armoring of As = 3,35 % m2/m under the service load. Additionally, the calculations will be assessed of their validity and accuracy on a scale of 1:1 in a loading test at the EMPA (the Swiss Federal Laboratories for Materials Testing and Research) in Dübendorf and be therefore confirmed.

This work refers to ball-intersections bodies as well as covering, packing, and kissing problems related to balls and spheres in normed spaces. A quick introduction to these topics and an overview of our results is given in Section 1.1 of Chapter 1. The needed background knowledge is collected in Section 1.2, also in Chapter 1. In Chapter 2 we define ball-intersection bodies and investigate special classes of them: ball-hulls, ball-intersections, equilateral ball-polyhedra, complete bodies and bodies of constant width. Thus, relations between the ball-hull and the ball-intersection of a set are given. We extend a minimal property of a special class of equilateral ball-polyhedra, known as Theorem of Chakerian, to all normed planes. In order to investigate bodies of constant width, we develop a concept of affine orthogonality, which is new even for the Euclidean subcase. In Chapter 2 we solve kissing, covering, and packing problems. For a given family of circles and lines we find at least one, but for some families even all circles kissing all the members of this family. For that reason we prove that a strictly convex, smooth normed plane is a topological Möbius plane. We give an exact geometric description of the maximal radius of all homothets of the unit disc that can be covered by 3 or 4 translates of it. Also we investigate configurations related to such coverings, namely a regular 4-covering and a Miquelian configuration of circles. We find the concealment number for a packing of translates of the unit ball.