Electrical and Computer Engineering Commons^™

Machine Learning Applications In Plant Identification, Wireless Channel Estimation, And Gain Estimation For Multi-User Software-Defined Radio, Viraj K. Gajjar

Doctoral Dissertations

"This work applies machine learning (ML) techniques to selected computer vision and digital communication problems. Machine learning algorithms can be trained to perform a specific task without explicit programming. This research applies ML to the problems of: plant identification from images of leaves, channel state information (CSI) estimation for wireless multiple-input-multiple-output (MIMO) systems, and gain estimation for a multi-user software-defined radio (SDR) application.

In the first task, two methods for plant species identification from leaf images are developed. One of the methods uses hand-crafted features extracted from leaf images to train a support vector machine classifier. The other method combines …

Signal Integrity Analysis And Electromagnetic Inteference Modeling For Autonomous Vehicles, Yuandong Guo

Doctoral Dissertations

"An autonomous vehicle (AV) may employ various sensors to detect the environment, and the hardware for self-driving should be able to process large volume of sensor data and make real-time decisions. For signal integrity (SI), both the bandwidths and data rates of the high-speed channels should meet the requirements. The electromagnetic interference (EMI) noise may become a major concern with the ever-increasing number of electric modules. This research focuses on both SI analysis and EMI modeling for AVs. The first topic is accurate and broadband three-phase motor modeling. The novel modeling methodology for a typical three-phase motor is presented, using …

Array Signal Processing For Source Localization And Enhancement, Sohel J. Patel

Doctoral Dissertations

“A common approach to the wide-band microphone array problem is to assume a certain array geometry and then design optimal weights (often in subbands) to meet a set of desired criteria. In addition to weights, we consider the geometry of the microphone arrangement to be part of the optimization problem. Our approach is to use particle swarm optimization (PSO) to search for the optimal geometry while using an optimal weight design to design the weights for each particle’s geometry. The resulting directivity indices (DI’s) and white noise SNR gains (WNG’s) form the basis of the PSO’s fitness function. Another important …

Modeling Strategy For Emi Filter And Flyback Transformer, Ruijie He

Doctoral Dissertations

“The switch-mode power supply is key to miniaturizing power adapters. However, the switching nature of the circuit introduces issues in conducted emissions. In a flyback converter, the transformer serves as the path for common mode current flowing from the primary side to the secondary side. Different winding technologies have been invented and implemented to reduce the capacitance between the primary side and the secondary side. But the repeatability of the winding is still poor due to the fluctuations of the winding machine. Thus, the resulting conducted emission has a fluctuation that can lead to failure in the compliance tests. EMI …

Characterizing And Modeling Methods For Power Converters, Anfeng Huang

Doctoral Dissertations

“Stable power delivery is becoming increasingly important in modern electronic devices, especially in applications with stringent requirements of its form factor. With the evolution of technology, the switching frequency in a power converter is pushed to a higher frequency range, e.g., several MHz or even higher, to decrease its size. However, the loss generated in the converter increases drastically due to the high switching frequency. In addition, a wide-band feedback controller is required to accommodate the high switching frequency in the converter. We focus on the characterization or modeling of the feedback control circuits and critical components in a switching …

Characterization And Modeling Of Esd Events, Risk And Protection, Jianchi Zhou

Doctoral Dissertations

“The ESD (Electrostatic discharge) failures have been raising critical reliability problems in electronic devices design. However, not all the ESD scenarios have been specified by the IEC standard and the characterizations of the ESD risk for different scenarios are essential to evaluate the ESD robustness of the devices in the real word.

The insulation of plastic enclosures provides protection against ESD to the electronic system inside. However, seams between plastic parts are often unavoidable. Different plastic arrangements are constructed to investigate the spark length and current derivatives and to understand the ESD spark behavior for geometries having spark lengths longer …

Novel Sensor Platforms Based On Fabry-Perot Resonators For Applications In Environmental Geophysics, Jing Guo

Doctoral Dissertations

"Fabry-Perot resonator sensors have been widely used for various physical and chemical measurements owing to their unique advantages over traditional sensors such as high measurement resolution, mechanically robust, and distributed sensing capabilities. This dissertation focuses on the development of robust fiber optic microwave sensors based on Fabry-Perot resonator mechanism for real-time applications in environmental geophysics.

Firstly, a novel Extrinsic Fabry-Perot Interferometer (EFPI)-based fiber-optic sensor for force measurement using a pre-buckled beam was demonstrated. The axial displacement is transferred and amplified to a horizontal deflection at the middle of the buckled beam, leading to a relatively significant change in the Fabry-Perot …

Distributed Fiber-Optic Temperature Sensors For Applications In The Steel Industry, Muhammad Roman

Doctoral Dissertations

"Steelmaking facilities require continuous temperature measurements throughout the manufacturing process to ensure consistent product quality and high productivity. Motivated by the limitations of conventional temperature sensors, distributed fiber-optic sensors (DFOS) were developed and deployed for various applications in the steel industry. Fiber-optic sensors offer various advantages over conventional sensors, such as the miniaturized size of the optical fiber, immunity to electromagnetic interferences, capability for multiplexing and distributed sensing, and the ability to withstand harsh environments.

Firstly, high-resolution Rayleigh backscattering based DFOS were demonstrated as potential solutions for temperature measurements in steelmaking processes by performing experimental simulations. Additionally, aluminum casting experiments …

Pi And Emi Analysis Of 3d Ic System With Tsvs And Co-Packaged Optics, Wei Zhang

Doctoral Dissertations

"With the demand for developing high-capacity, high-density, high-power optical links for data center communications, co-packaged optics is a promising solution. When the three-dimensional (3D) IC and optic engines are integrated into the same package, the design of such systems in terms of signal integrity, power integrity, and electromagnetic interference will be more challenging.

This first paper mainly focuses on modeling the bias-voltage-dependent, temperature-dependent, and light-intensity-dependent hysteresis behavior of a through-silicon-vias (TSV) structure in 3D IC and analyzing its effect on the power integrity of a system with 3D IC.

This second paper investigates the scaled tendency of the emissions in …

Accuracy Improvement Of Cable Harness Modeling & Analytical Modeling Of Multi-Reflections In High-Speed Signal Channels, Muqi Ouyang

Doctoral Dissertations

"Cable harnesses are widely used in modern vehicles to power electric systems locating at different locations throughout the vehicle and to establish signal channels for data communications between these systems. Generally, a cable harness consists of multiple wires and its geometry can be very complicated, making full-wave modeling and simulation approaches computational resource-intensive. To cut down the computational effort, the hybrid multiconductor transmission-line (MTL) method has been developed, and the generalized MTL (GMTL) method and the multiple scattering (MS) method have been developed to improve the accuracy of the hybrid MTL method. In this research, the convergence issue in the …

Multi-Stage Multi-Market Optimal Bidding Strategy In Electricity Markets, Hossein Mehdipourpicha

Doctoral Dissertations

"Restructuring of power industry created the competitive situations for different MPs through designing appropriate electricity markets. Typically, in pool-based electricity markets, MPs submit their offers, and ISO clears the market and notifies MPs of the cleared power amounts and market prices. In this situation, designing an effective bidding strategy for MPs who seek to maximize their profits has thus become one of the crucial yet difficult tasks that still need further exploration. The majority of recent studies overlooked this problem for the purely financial players. Therefore, designing an efficient strategy for virtual bidder in the pool-based electricity markets is studied …

Far-End Crosstalk Modeling And Prediction For High-Speed Pcb Design With Inhomogeneous Dielectric Layers (Idls), Yuanzhuo Liu

Doctoral Dissertations

"Far-end crosstalk (FEXT) noise is a critical factor that affects signal integrity performance in high-speed systems. The FEXT level is sensitive to the inhomogeneity of the dielectric layers in fabricated printed circuit boards (PCB). The stripline is laminated by multiple inhomogeneous dielectric layers (IDL). The dielectric layers of the stripline are laminated with epoxy resin and glass bundles. The dielectric permittivity of the epoxy resin and glass bundles are different, which causes the inhomogeneity of the dielectric layers while also increasing the FEXT magnitude. The dielectric of the microstrip in printed circuit boards (PCB) fabrication usually consists of two layers: …

Optimization Of Energy Storage Scheduling In Electricity Markets, Jian Liu

Doctoral Dissertations

"In the existing literature, merchants' trading actions are usually assumed not to affect market prices; however, a large-scale energy storage merchant’s actions can affect market prices. This work examined two electricity merchant scenarios: one with only energy storage and the other with both energy storage and renewable power plants. We approximated market impact via a linear function of the electricity traded by the merchant. This study began by applying dynamic programming to the optimal economic dispatch policy of electricity merchants and it considered the market impact, physical characteristics of storage systems, and the uncertainty of renewable energy sources. Then, this …

Theoretical Study Of Magnetic Particles In A Shear Flow Subjected To A Uniform Magnetic Field, Christopher A. Sobecki

Doctoral Dissertations

"Magnetic manipulation of non-spherical magnetic microparticles is important for applications in shape-based and magnetic-based separations such as waste management, disease diagnostics, drug delivery, and mining. Manipulations of magnetic microparticles also include chain formation to assemble compositions for electronics, drug loading designs, and magnetorheological fluids for smart armor, hydraulic brakes, and dampers. In microfluidic devices, separation-formation-effectiveness depends on the shape of the channel, the shear rate, and the magnetic field strength and direction.

Particle separation and chain formation involved highly complex and computational expense-demanding studies in microfluidic devices, magnetic fields, and particle- particle/wall interactions. This research took complex experimental studies and …

Investigating The Effect Of Operating Condition On Esd-Induced Soft-Failures, Omid Hoseini Izadi

Doctoral Dissertations

"Conflicting observations have been found in the literature regarding the effect of operating conditions on ESD (electrostatic discharge) susceptibility. While some studies have suggested a strong correlation between the two, others observed little to no correlation. In this work, a systematic study has been carried out suggesting the existence of a strong correlation between the ESD susceptibility and operating conditions. It is found that the root cause of this conflict is random ESD noise injection. A measurement approach is proposed to synchronize the noise injection with the system activity such as high/low CPU load. In this approach, the current consumption …

Integrated Solar And Ess Ac/Dc Systems For Various Grid Applications, Phani Ratna Vanamali Marthi

Doctoral Dissertations

"Modular multilevel converters (MMC) have gained prominence for high power applications in recent years due to their technical merits such as voltage scalability, harmonic performance, and high efficiency. Integration of renewable energy sources such as solar photovoltaic (PV) systems and energy storage systems (ESS) to the grid using power electronic (PE) resources is increasing due to the techno-economic successes achieved by such systems. Following these developments, the integration of MMCs with PV plants and ESS systems is one of the dynamic research avenues being pursued. Along these lines, an integrated development of PV plants, ESS, and alternating- and direct-current (ac/dc) …

Stability Analysis Of Microgrids Using Markov Jump Linear Systems, Gilles Mpembele

Doctoral Dissertations

"This research discusses stochastic models for a microgrid operating between standalone and grid-tied modes. The transitions between different modes are modeled as a continuous-time Markov chain (CTMC). In each operating mode, the system is modeled using conventional differential algebraic equations (DAEs), linearized around some equilibrium point.

In Topic-I, a model is developed using the Stochastic Hybrid Systems (SHSs) formulation. The microgrid is modeled as a Markov jump linear system (MJLS), which is a type of SHS in which the discrete events evolve according to a Continuous Time Markov Chain (CTMC). The model allows for the derivation of Ordinary Differential Equations …

Source Reconstruction In Near Field Scanning For Rfi Application, Hosseien Rezaei

Doctoral Dissertations

"This research is divided into three major topics. The first topic, mechanical magnetic field generator for communication in the ULF range, is discussed in the first paper. The second topic, source reconstruction in near field scanning for RFI application, is discussed over the course of two papers. The third topic, analysis of imbalanced 2 or 3 Wire VHF LISN, is discussed in the last two papers.

In the first topic, the possibility to use a mechanical system (a rotating magnet) as a source (generator or antenna combined) of the ULF magnetic field is investigated. Ultralow frequency (ULF) communication systems have …

Modeling And Analyzing Parasitic Parameters In High Frequency Converters, Mohamad Saleh Sanjari Nia

Doctoral Dissertations

"This research focuses on electromagnetic interference (EMI) / electromagnetic compatibility (EMC) design and analysis in power electronics systems. To limit the EMI under the standards, different methods and strategies are investigated. Parasitic parameters of high frequency (HF) transformer are analyzed using a novel analytical method, finite element method (FEM), and experimental measurements for different structures and windings arrangements. Also, the magnetic field, electric field, electric displacement, and electric potential distribution are simulated and analyzed. Moreover, a high voltage system is considered and analyzed to improve the EMC. The EMI propagation paths are analyzed. The EMI noise level of the system …

Pdn Modeling For High-Speed Multilayer Pcb Boards And Decap Optimization Using Machine Learning Techniques, Ling Zhang

Doctoral Dissertations

"In power distribution networks (PDN), there are two main challenges nowadays. One challenge is how to efficiently model and calculate the impedance for arbitrary-shape and multi-layer PDN systems. The second challenge is the optimization of decoupling capacitors in the pre-layout stage. This dissertation proposes novel solutions to these two challenges. To tackle the first challenge, a boundary element method (BEM) is utilized to calculate the quasi-static inductances between vertical vias for arbitrary-shape planes. Then a specialized circuit solver is developed to solve the equivalent circuit of inductances and capacitances for multi-layer PDN structures. Also, a contour integral method (CIM) is …

Fiber Optic Sensors For Industry And Military Applications, Yiyang Zhuang

Doctoral Dissertations

"Fiber optic sensors (FOSs) have been widely used for measuring various physical and chemical measurands owing to their unique advantages over traditional sensors such as small size, high resolution, distributed sensing capabilities, and immunity to electromagnetic interference. This dissertation focuses on the development of robust FOSs with ultrahigh sensitivity and their applications in industry and military areas.

Firstly, novel fiber-optic extrinsic Fabry-Perot interferometer (EFPI) inclinometers for one- and two-dimensional tilt measurements with 20 nrad resolution were demonstrated. Compared to in-line fiber optic inclinometers, an extrinsic sensing motif was used in our prototype inclinometer. The variations in tilt angle of the …

Protection Of Modern Distribution Systems, Fahd Amin Hariri

Doctoral Dissertations

“Motivated by the potential for improvements in the electric distribution system’s protection schemes, this work examined the challenges facing protection schemes due to the integration of Distributed Generators (DGs). Traditional protection schemes for radial distribution systems were designed based on the unidirectional power flow from the source down to the loads. Protective devices typically use are overcurrent relays, autoreclosers, fuses, and circuit breakers. However, these protective schemes may no longer be sufficient to ensure correct operation in the new era of distribution systems integrated by DGs. This research investigated the impact of DGs that might mislead the protection schemes in …

Emi Characterization For Power Supplies And Machine Learning Based Modeling In Emc/Si, Jiayi He

Doctoral Dissertations

"Signal integrity (SI) and electromagnetic interference (EMI) are essential for consumer electronics and high-speed server applications. It is necessary to do EMI and SI modeling at the design stage. In this research, several modeling approaches for EMI and SI problems are proposed. By using measurement-based modeling method, the mechanism of conducted emissions (CE) on ac to dc power supplies in an LED TV is analyzed. A system-level transient simulation model is built to predict the conducted emission (CE). To characterize the equivalent dipole moment sources in RFI and EMI problems, a dipole source reconstruction method based on machine learning techniques …

Modeling For Off-Chip High-Speed Designs, Siqi Bai

Doctoral Dissertations

”Signal integrity (SI) and power integrity (PI) are essential for datacenter application. A typical datacenter rack consists of a switch on the top of the rack, servers, and storage units. Three SI/PI related problems for datacenter application are analyzed in this research.

The first topic is analysis on power via induced Quasi-quarter-wavelength resonance to reduce crosstalk. Using an analytical model and 3D full-wave simulation models, a physical explanation for the unexpected resonance in the differential signal pairs is proposed. Considering the difficulty in changing the pin map of the IC-package, several PCB layouts to eliminate the power via induced quasi-quarter-wavelength …

Fiber-Optic And Coaxial-Cable Extrinsic Fabry-Perot Interferometers For Sensing Applications, Chen Zhu

Doctoral Dissertations

”The fiber-optic extrinsic Fabry-Perot interferometer (EFPI) is one of the simplest sensing configurations and is widely used in various applications due to its prominent features, such as high sensitivity, immunity to electromagnetic interference, and remote operation capability. In this research, a novel one-dimensional wide-range displacement sensor and a three-dimensional displacement sensor based on fiber-optic EFPIs are demonstrated. These two robust and easy-to-manufacture sensors expand the application scope of the fiber-optic EFPI sensor devices, and have great potential in structural health monitoring, the construction industry, oil well monitoring, and geo-technology.

Furthermore, inspired by the fiber-optic EFPI, a novel and universal ultra-sensitive …

Adaptive Dynamic Programming Methods For Tracking Current Control Of Switched Reluctance Motor Drive, Hamad Alharkan

Doctoral Dissertations

“This research investigates the application of adaptive dynamic programming (ADP) methods for the current control of switched reluctance motors (SRM). Since the traditional current controllers have large current ripples that increase torque ripples and acoustic noise, optimal tracking controllers using reinforcement learning (RL) approaches are introduced as an alternative adaptive technique for SRM drives. The proposed optimal tracking controllers can minimize current ripples and cope with uncertainties within the inductance profile of the SRM. The work is composed of three papers. In the first paper, a novel Q-learning scheduling method for the current controller of a switched reluctance motor (SRM) …

Advanced Control Techniques For Modern Inertia Based Inverters, Sepehr Saadatman

Doctoral Dissertations

”In this research three artificial intelligent (AI)-based techniques are proposed to regulate the voltage and frequency of a grid-connected inverter. The increase in the penetration of renewable energy sources (RESs) into the power grid has led to the increase in the penetration of fast-responding inertia-less power converters. The increase in the penetration of these power electronics converters changes the nature of the conventional grid, in which the existing kinetic inertia in the rotating parts of the enormous generators plays a vital role. The concept of virtual inertia control scheme is proposed to make the behavior of grid connected inverters more …

Instrumentation, Modeling, And Sound Metamodeling Foundations For Complex Hybrid Systems, Natasha Amelia Jarus

Doctoral Dissertations

Many of our critical infrastructures, from power grids to water distribution networks, are complex hybrid systems that use software to control their non-trivial physical dynamics. These systems must be able to capably serve their purpose, while also being reliable, dependable, safe, secure, and efficient. Representation and analysis of these features requires the creation of several distinct models. These models may encode design goals or be derived from collected instrumentation data, reflecting both how a system ought to operate and how it does operate. It is essential to ensure that all of these models consistently and accurately describe the same system. …

Operation And Analysis Of High-Voltage-Gain Dc-Dc Converters With Voltage Multipliers, Mohammad Rashed M. Altimania

Doctoral Dissertations

"This research analyzes and investigates the steady-state operations of four different high-voltage-gain step-up dc- dc converters to achieve a high gain. These topologies consist of a two-phase interleaved boost stage with three various voltage multiplier (VM) stages. The first stage of the first three topologies is the same, which is a two-phase interleaved boost stage implemented by using inductors. The first family used in this research is based on a modified Dickson charge pump voltage multiplier stage, and this family has been studied in discontinuous conduction mode (DCM). The second family relies on a non-inverting or inverting diode-capacitors voltage multiplier …

Optimal Adaptive Control Of Time-Delay Dynamical Systems With Known And Uncertain Dynamics, Rohollah Moghadam

Doctoral Dissertations

"Delays are found in many industrial pneumatic and hydraulic systems, and as a result, the performance of the overall closed-loop system deteriorates unless they are explicitly accounted. It is also possible that the dynamics of such systems are uncertain. On the other hand, optimal control of time-delay systems in the presence of known and uncertain dynamics by using state and output feedback is of paramount importance. Therefore, in this research, a suite of novel optimal adaptive control (OAC) techniques are undertaken for linear and nonlinear continuous time-delay systems in the presence of uncertain system dynamics using state and/or output feedback. …