The research presented in this doctoral dissertation introduces a developmental path to realize micro coaxial transmission lines for wafer-level microwave and millimeter wave circuits. The micro coax lines conceptualized in this research are three-dimensional, with a uniquely low vertical profile of 20 mu m. They are distinguishable from other transmission line topologies due to frequency limitless, cross-talk free performance in high density circuits resulting from the totally shielded construction. Design and optimization was done using equations and electromagnetic simulations. Extensive process development has been carried out to fabricate these devices on CMOS grade silicon wafers using processes amenable to the IC manufacturing industry. Micro coax lines of three different core types including air, semi-dielectric and all-dielectric cores were designed, fabricated and analyzed from 1 - 40 GHz with respect to line impedance and the volume of polyimide (dielectric) in the micro coax cavity. The insertion loss of the lines was found to vary from 2 to 9 dB/cm depending on the width the center conductor, total cross-sectional area and core type. The versatility of the micro coax technology developed in the earlier part of this research has been successfully integrated with RF MEMS to embed a MEM varactor inside the micro coax cavity capable of achieving a capacitance ratio around 7 with an actuation voltage of 40 - 60V. The vertical integration capability of the micro coax was demonstrated by locating a millimeter wave slot antenna on the top side shield of a rectangular micro coax line, also acting as the electromagnetically coupled feed for the antenna. The antenna was designed to operate in the full range of the unlicensed 57 - 64 GHz spectrum capable of Gbps data rates. A 15 dB bandwidth of 17% and a hemispherical radiation pattern was achieved. Lastly, a diversified application involving the application of an integrated micro coax probe-microfluidic chamber approach to detect bacteria through impedance measurements has been introduced. In summary, the development and implementation of micro coax technology on a large scale is bound to create a paradigm shift in the development of high density microwave and millimeter wave circuits for RF MEMS and integrated antennas.

The development of wireless communications increases the challenges on antenna performance to improve the capability of the whole system. New fabrication technologies are emerging that not only can improve the performance of components but also provide more options for materials and geometries. One of the advanced technologies, referred to as deep X-ray lithography (XRL), can improve the performance of RF components while providing interesting opportunities for fabrication. Since this fabrication technology enables the objects of high aspect ratio (tall) structure with high accuracy, it offers RF/microwave components some unique advantages, such as higher coupling energy and compacted size. The research presented in that thesis investigates the properties of deep XRL fabricated tall microstrip transmission line and describes some important features such as characteristic impedance, attenuation, and electromagnetic field distribution. Furthermore, since most of traditional feeding structure cannot supply enough coupling energy to excite the low permittivity DRA element (εr≤10), three novel feeding schemes composed by tall microstrip line on exciting dielectric resonator antennas (DRA) with low permittivity are proposed and analyzed in this research. Both simulation and experimental measured results exhibit excellent performance. Additionally, a new simulation approach to realize Dolph-Chebyshev linear series-fed DRA arrays by using the advantages of tall microstrip line feeding structure is proposed. By using a novel T shape feeding scheme, the array exhibits wide band operation due to the low permittivity (εr=5) DRA elements and good radiation pattern due to the novel feeding structure. The tall metal transmission line feed structure and the polymer-based DRA elements could be fabricated in a common process by the deep XRL technology. This thesis firstly illustrates properties and knowledge for both DRA element and the tall transmission line. Then the three novel feeding schemes by using the tall transmission line on exciting the low permittivity DRA are proposed and one of the feeding structures, side coupling feeding, is analyzed through the simulation and experiments. Finally, the T shape feeding structure is applied into low permittivity linear DRA array design work. A novel method on designing the Dolph-Chebyshev array is proposed making the design work more efficient.

Abstract Adding a dc circuit to an existing transmission line is one method of significantly increasing the power transfer capability of a transmission corridor. The resulting hybrid system has significant coupling between the ac and dc circuits, not only because of the proximity of the circuits, but also from the fact that they may share the same sending end or receiving end ac systems. The resultant interaction produces overvoltages on the dc system which can be somewhat higher than for a conventional dc scheme. This thesis investigates the overvoltages on a hybrid ac-dc transmission system and suggests some design considerations which could be taken into account to reduce stresses on certain critical components which result from such an arrangement. Blocking filters consisting of a parallel L-C combination in series with the dc converter were included to limit the flow of fundamental frequency current in the dc line. This thesis also investigates the proper blocking filter configuration to be used as an incorrectly chosen blocking filter can cause resonance overvoltages on the dc line at fundamental frequency. A method of eliminating dc components of the currents in the transformer windings of a dc converter is presented. The method uses the technique of firing angle modulation. It is shown that merely eliminating the fundamental frequency component on the dc side may not remove this dc component. The impact of such control action at one converter on the other converters in the dc transmission system is also presented. It is also shown that the undesirable side effects of such a scheme include increased generation of non-characteristic harmonies on both the ac and dc sides. The study is performed using an electromagnetic transients simulation program and theoretical calculations.

Electric power systems have been in existence for over a century. Electric power transmission line systems play an important role in carrying electrical power to customers everywhere. The number of transmission lines in power systems is increasing as global demand for power has increased. Parallel transmission lines are widely used in the modern transmission system for higher reliability. The parallel lines method has economic and environmental advantages over single circuit. A fault that occurs on a power transmission line will cause long outage time if the fault location is not located as quickly as possible. The faster the fault location is found, the sooner the system can be restored and outage time can be reduced. The main focus of this research is to develop a new accurate fault location algorithm for parallel transmission lines to identify the fault location for long double-circuit transmission lines, taking into consideration mutual coupling impedance, mutual coupling admittance, and shunt capacitance of the line. In this research, the equivalent PI circuit based on a distributed parameter line model for positive, negative, and zero sequence networks have been constructed for system analysis during the fault. The new method uses only the voltage and current from one end of parallel lines to calculate the fault distance. This research approaches the problem by derivation all equations from positive sequence, negative sequence, and zero sequence network by using KVL and KCL. Then, the fault location is obtained by solving these equations. EMTP has been utilized to generate fault cases under various fault conditions with different fault locations, fault types and fault resistances. Then the algorithm is evaluated using the simulated data. The results have shown that the developed algorithm can achieve highly accurate estimates and is promising for practical applications.

This thesis aims at detecting and classifying the power system transmission line faults. To deal with the problem of an extremely large data set with different fault situations, a three step optimized Neural Network approach has been proposed. The approach utilizes Discrete Wavelet Transform for detection and two different types of self-organized, unsupervised Adaptive Resonance Theory Neural Networks for classification. The fault scenarios are simulated using Alternate Transients Program and the performance of this highly improved scheme is compared with the existing techniques. The simulation results prove that the proposed technique handles large data more efficiently and time of operation is considerably less when compared to the existing methods.

The past decade has seen a transition of Graphics Processing Units (GPUs) from special purpose graphics processors, to general purpose computational accelerators. GPUs have been investigated to utilize their highly parallel architecture to accelerate the computation of the Transmission Line Matrix (TLM) methods in two and three dimensions. The design utilizes two GPU programming languages, Compute Unified Device Architecture (CUDA) and Open Computing Language (OpenCL), to code the TLM methods for NVIDIA GPUs. The GPU accelerated two-dimensional shunt node TLM method (2D-TLM) achieves 340 million nodes per second (MNodes/sec) of performance which is 25 times faster than a commercially available 2D-TLM solver. Initial attempts to adapt the three-dimensional Symmetrical Condensed Node (3D-SCN) TLM method resulted in a peak performance of 47 MNodes/sec or7 times in speed-up. Further efforts to improve the 3D-SCN TLM algorithm, as well as investigating advanced GPU optimization strategies resulted in performances accelerated to 530 MNodes/sec, or 120 times speed-up compared to a commercially available 3D-SCN TLM solver.

Abstract Adding a dc circuit to an existing transmission line is one method of significantly increasing the power transfer capability of a transmission corridor. The resulting hybrid system has significant coupling between the ac and dc circuits, not only because of the proximity of the circuits, but also from the fact that they may share the same sending end or receiving end ac systems. The resultant interaction produces overvoltages on the dc system which can be somewhat higher than for a conventional dc scheme. This thesis investigates the overvoltages on a hybrid ac-dc transmission system and suggests some design considerations which could be taken into account to reduce stresses on certain critical components which result from such an arrangement. Blocking filters consisting of a parallel L-C combination in series with the dc converter were included to limit the flow of fundamental frequency current in the dc line. This thesis also investigates the proper blocking filter configuration to be used as an incorrectly chosen blocking filter can cause resonance overvoltages on the dc line at fundamental frequency. A method of eliminating dc components of the currents in the transformer windings of a dc converter is presented. The method uses the technique of firing angle modulation. It is shown that merely eliminating the fundamental frequency component on the dc side may not remove this dc component. The impact of such control action at one converter on the other converters in the dc transmission system is also presented. It is also shown that the undesirable side effects of such a scheme include increased generation of non-characteristic harmonies on both the ac and dc sides. The study is performed using an electromagnetic transients simulation program and theoretical calculations.

The past decade has seen a transition of Graphics Processing Units (GPUs) from special purpose graphics processors, to general purpose computational accelerators. GPUs have been investigated to utilize their highly parallel architecture to accelerate the computation of the Transmission Line Matrix (TLM) methods in two and three dimensions. The design utilizes two GPU programming languages, Compute Unified Device Architecture (CUDA) and Open Computing Language (OpenCL), to code the TLM methods for NVIDIA GPUs. The GPU accelerated two-dimensional shunt node TLM method (2D-TLM) achieves 340 million nodes per second (MNodes/sec) of performance which is 25 times faster than a commercially available 2D-TLM solver. Initial attempts to adapt the three-dimensional Symmetrical Condensed Node (3D-SCN) TLM method resulted in a peak performance of 47 MNodes/sec or7 times in speed-up. Further efforts to improve the 3D-SCN TLM algorithm, as well as investigating advanced GPU optimization strategies resulted in performances accelerated to 530 MNodes/sec, or 120 times speed-up compared to a commercially available 3D-SCN TLM solver.

This thesis describes the development of antennas suitable for mobile coinmunication systems based on a dielectric rod antenna fed from circular waveguide. Pattern control of the antenna is implemented using a combination of Frequency Selective Surface (FSS) elements and metallic endcaps placed on the antenna Both linear and circular polarised feeds have been made for these antennas to ensure they are suitable for a wide range of applications. The suitability of the dominant and next, higher order, waveguide mode were investigated and conclusions drawn as to their suitability for this type of antenna. The antennas were extensively modelled using a commercial TLM based solver and the results of these simulations were compared to the comprehensive set of antenna pattern measurements and S-parameter measurements obtained for the prototype antennas.

Orientador: Sérgio Kurokava / Banca: Luiz Fernando Bovolato / Banca: José Carlos da Costa Campos / Resumo: Sabe-se que uma linha de transmissão polifásica pode ser representada no domínio modal, por seus n modos de propagação que se comportam como sendo n linhas monofásicas independentes. Uma vez calculadas as correntes e tensões no domínio modal, as mesmas são convertidas para o domínio das fases por meio de uma matriz de transformação modal. A matriz de transformação modal é uma matriz cujos elementos são escritos em função dos parâmetros longitudinais e transversais da linha, variam em função da frequência e, geralmente, são obtidos por meio de métodos numéricos. Deste modo, diz-se que o modelo obtido é um modelo numérico de linha. Neste trabalho foi feita uma abordagem a respeito de um modelo analítico de linha de transmissão bifásica. O modelo proposto utiliza também a representação modal, mas a matriz de transformação será obtida analiticamente em função dos parâmetros da linha. Deste modo, foi possível obter, analiticamente, relações entre as correntes e tensões de fase da linha baseando-se unicamente nos parâmetros longitudinais e transversais da mesma / Abstract: It is know that polyphase transmission line can be represented in the modal domain its n propagation modes that behave as n independent single-phase lines. Once calculated the currents and voltages in the modal domain, they are converted into the realm of the phases by means of a modal transformation matrix. The modal transformation matrix is a matrix whose elements re written against the parameters of longitudinal and cross the and they are usually obtained by numerical methods. In this paper an approach was made on an analytical model of two-phase transmission line. The proposed model uses the modal representation, but the transmission matrix obtained analytical in terms of line parameters. The development of the analytical model will be based on the modal model. Thus, initially will be obtained analytically, a modal matrix decomposition that allows to calculate analytically the eigenvalues of the product [Z][Y] line. Once obtained the eigenvalues it possible to abtain the modes of propagation and characteristic impedance of the line modes. Then, using the solutions algebraic differential equation of a single-phase line, we abtain the equations of currents and voltages of each of modes of spread of the row. In a final step, the equations of modal currents and voltages are converted into the realm of the phases, resulting in algebraic equations that can calculate the currents and phase voltages of the line in the frequency domain / Mestre