<p> DNA microarrays can measure the gene expression of thousands of genes at a time to identify differentially expressed genes. The Affymetrix GeneChip system is a platform for the high-density oligonucleotide microarray to measure gene expression using hundreds of thousands of 25-mer oligonucleotide probes.</p> <p> To deal with Affymetrix microarray data, there are three stages of preprocessing to produce gene expression measurements/values. These are background correction, normalization and summarization. At each stage, numerous methods have been developed.</p> <p> Our study is based on Affymetrix MG_U74Av2 chip with 12488 probe sets. Two strains of mice called NOR and NOR.NOD_Idd4/11 mouse are hybridized for the experiment. We apply a number of commonly used and state-of-art normalization methods to the data set, thus compute the expression measurements for different methods. The major methods we discuss include Robust Multi-chip Average (RMA), MAS 5.0, GCRMA, PLIER and dChip.</p> <p> Comparisons in terms of correlation coefficient, pairwise expression measures plot, fold change and Significance Analysis of Microarray (SAM) are conducted.</p> / Thesis / Master of Science (MSc)

Persistence is an important issue in higher education (Tinto, 1987a). Although social and economic benefits of a college education are well documented (Baum & Ma, 2007; Institute for Higher Education Policy, 2004), during the last 100 years the college graduation rate has remained at approximately 50% (Education Policy Institute, 2004). While prior academic achievement has proven to be a successful predictor of success in college (Camara & Echtnernacht, 2000; Sadler, Cohen, & Kockesen, 1997; Tinto, 1993), it does not account for all the variability in student retention. Research has shown that other factors, including social adaptation, physical fitness, and emotional stability can contribute to whether an individual continues to persist past the first year of college (Astin, 1993; McClanahan, 2004; Tinto, 1987b). The purpose of this study was to examine the relationship between self-reported wellness and academic success in first-year health science college students. In addition the study sought to determine if the relationship between wellness and academic success differs by gender, academic program, or ethnicity. Also examined was whether the factors of wellness could be used to predict academic success. Wellness was defined using the Myers and Sweeney (2005) conceptual framework, as measured in a series of constructs, including the coping self, creative self, essential self, physical self, social self and an overall wellness score. Academic success was defined as first semester grade point average. The study also controlled for high school grade point average (HSGPA) and scholastic aptitude test score (SAT) as factors of prior academic achievement that may affect academic success in college. The findings suggest that the impact of wellness differs by ethnicity and academic program. In addition, certain factors of wellness can be used to help predict academic success in the first semester of college. Finally, overall wellness had little if any bearing on academic success in first-time, first-year students. / Ph. D.

This work presents a modeling methodology that uses new types of models called low-frequency, large-signal models in a circuit simulator (Saber) to model a complex hybrid ac/dc power electronics system. The new achievement in this work is being able to model the different components as circuit-based models and to capture some of the large-signal phenomena, for example, real transient behavior of the system such as startup, inrush current and power flow directionality. In addition, models are capable of predicting most low frequency harmonics only seen in real switching detailed models. Therefore the new models system can be used to predict steady state performance, harmonics, stability and transients. This work discusses the modeling issues faced based on the author recent experiences both on component level and system level. In addition, it recommends proper solutions to these issues verified with simulations. This work also presents one of the new models in detail, a voltage source inverter (VSI), and explains how the model can be modified to capture low frequency harmonics that are usually phenomena modeled only with switching models. The process of implementing these different phenomena is discussed and the model is then validated by comparing the results of the proposed low frequency large signal (LFLS) model to a complete detailed switching model. In addition, experimental results are also obtained with a 2 kW voltage source inverter prototype to validate the proposed improved average model (LFLS model). In addition, a complete Verification, Validation, and Uncertainty Quantification (VV&UQ) procedures is applied to a two-level boost rectifier. The goal of this validation process is the improvement of the modeling procedure for power electronics systems, and the full assessment of the boost rectifier model predictive capabilities. Finally, the performance of the new models system is compared with the detailed switching models system. The LFLS models result in huge cut in simulation time (about 10 times difference) and also the ability to use large time step with the LFLS system and still capture all the information needed. Even though this low frequency large signal (LFLS) models system has wider capabilities than ideal average models system, it still can’t predict all switching phenomena. Therefore, another benefit of this modeling approach is the ability to mix different types of models (low frequency large signal (LFLS) and detailed switching) based on the application study they are used for. / Ph. D.

The trend in isolated DC/DC bus converters is to increase the output power in the same brick form factors that have been used in the past. Traditional intermediate bus converters (IBCs) use silicon power metal oxide semiconductor field effect transistors (MOSFETs), which recently have reached the limit in terms of turn on resistance (RDSON) and switching frequency. In order to make the IBCs smaller, the switching frequency needs to be pushed higher, which will in turn shrink the magnetics, lowering the converter size, but increase the switching related losses, lowering the overall efficiency of the converter. Wide-bandgap semiconductor devices are becoming more popular in commercial products and gallium nitride (GaN) devices are able to push the switching frequency higher without sacrificing efficiency. GaN devices can shrink the size of the converter and provide better efficiency than its silicon counterpart provides. A survey of current IBCs was conducted in order to find a design point for efficiency and power density. A two-stage converter topology was explored, with a multiphase buck converter as the front end, followed by an LLC resonant converter. The multiphase buck converter provides regulation, while the LLC provides isolation. With the buck converter providing regulation, the switching frequency of the entire converter will be constant. A constant switching frequency allows for better electromagnetic interference (EMI) mitigation. This work includes the details to design and implement a hard-switched multiphase buck converter with planar magnetics using GaN devices. The efficiency includes both the buck efficiency and the overall efficiency of the two-stage converter including the LLC. The buck converter operates with 40V - 60V input, nominally 48V, and outputs 36V at 1 kW, which is the input to the LLC regulating 36V – 12V. Both open and closed loop was measured for the buck and the full converter. EMI performance was not measured or addressed in this work. / Master of Science / Traditional silicon devices are widely used in all power electronics applications today, however they have reached their limit in terms of size and performance. With the introduction of gallium nitride (GaN) field effect transistors (FETs), the limits of silicon can now be passed with GaN providing better performance. GaN devices can be switched at higher switching frequencies than silicon, which allows for the magnetics of power converters to be smaller. GaN devices can also achieve higher efficiency than silicon, so increasing the switching frequency will not hurt the overall efficiency of the power converter. GaN devices can handle higher switching frequencies and larger currents while maintaining the same or better efficiencies over their silicon counterparts. This work illustrates the design and implementation of GaN devices into a multiphase buck converter. This converter is the front end of a two-stage converter, where the buck will provide regulation and the second stage will provide isolation. With the use of higher switching frequencies, the magnetics can be decreased in size, meaning planar magnetics can be used in the power converter. Planar magnetics can be placed directly inside of the printing circuit board (PCB), which allows for higher power densities and easy manufacturing of the magnetics and overall converter. Finally, the open and closed loop were verified and compared to the current converters that are on the market in the 48V – 12V area of intermediate bus converters (IBCs).

We use molecular mechanics (MM) simulations with the tight-binding (TB) potential to study local and global instabilities in initially defect-free finite specimens of gold crystals deformed in shear, simple shear, tension/compression, simple tension/compression, and triaxial tension/compression. The criteria used to delineate local instabilities in a system include the following: (i) a second order spatial derivative of the displacement field having large values relative to its average value in the body, (ii) the minimum eigenvalue of the Hessian of the potential energy of an atom becoming nonpositive, (iii) and structural changes represented by a high value of the common neighborhood parameter. A specimen becomes globally unstable when its potential energy decreases significantly with a small increase in its deformations. It is found that the three criteria for local instability are satisfied essentially simultaneously at the same atomic position. Deformations of a specimen are quite different when it is deformed with some bounding surfaces free from external forces as opposed to essential boundary conditions prescribed on all bounding surfaces. It is found that the initial unloaded configuration (or the reference configuration) of the minimum potential energy has significant in-plane stresses on the bounding surfaces and nonzero normal stresses at interior points. In tensile/compressive deformations of a rectangular prismatic nanobar the yield stress defined as the average axial stress when the average axial stress vs. the average axial strain curve exhibits a sharp discontinuity depends upon the specimen size; a similar result holds for simulations of shear deformations. Specimens deformed with essential boundary conditions on all bounding surfaces experience instabilities at a higher value of the average strain than identical specimens deformed similarly but with one or more pairs of opposite bounding surfaces traction free. For the former set of deformations, the response of a specimen prior to the onset of instability is the same as that of a hyperelastic body with the strain energy derived from the TB potential and deformations obeying the Cauchy-Born rule. Specimens with some traction free bounding surfaces experience local instabilities prior to the onset of a global instability but the two instabilities occur simultaneously in specimens with essential boundary conditions prescribed on all bounding surfaces. It is believed that because of residual stresses in the reference configuration, the average axial stress at yield in compression is nearly one-half of that in tension. / Ph. D.

We examine some of the methodologies implemented by the Centers for Disease Control and Prevention's (CDC) BioSense program. The program uses data from hospitals and public health departments to detect outbreaks using the Early Aberration Reporting System (EARS). The EARS method W2 allows one to monitor syndrome counts (W2count) from each source and the proportion of counts of a particular syndrome relative to the total number of visits (W2rate). We investigate the performance of the W2r method designed using an empiric recurrence interval (RI) in this dissertation research. An adaptive threshold monitoring method is introduced based on fitting sample data to the underlying distributions, then converting the current value to a Z-score through a p-value. We compare the upper thresholds on the Z-scores required to obtain given values of the recurrence interval for different sets of parameter values. We then simulate one-week outbreaks in our data and calculate the proportion of times these methods correctly signal an outbreak using Shewhart and exponentially weighted moving average (EWMA) charts. Our results indicate the adaptive threshold method gives more consistent statistical performance across different parameter sets and amounts of baseline historical data used for computing the statistics. For the power analysis, the EWMA chart is superior to its Shewhart counterpart in nearly all cases, and the adaptive threshold method tends to outperform the W2 rate method. Two modified W2r methods proposed in the dissertation also tend to outperform the W2r method in terms of the RI threshold functions and in the power analysis. / Ph. D.

The risk-adjusted Bernoulli cumulative sum (CUSUM) chart developed by Steiner et al. (2000) is an increasingly popular tool for monitoring clinical and surgical performance. In practice, however, use of a fixed control limit for the chart leads to quite variable in-control average run length (ARL) performance for patient populations with different risk score distributions. To overcome this problem, the simulation-based dynamic probability control limits (DPCLs) patient-by-patient for the risk-adjusted Bernoulli CUSUM charts is determined in this study. By maintaining the probability of a false alarm at a constant level conditional on no false alarm for previous observations, the risk-adjusted CUSUM charts with DPCLs have consistent in-control performance at the desired level with approximately geometrically distributed run lengths. Simulation results demonstrate that the proposed method does not rely on any information or assumptions about the patients' risk distributions. The use of DPCLs for risk-adjusted Bernoulli CUSUM charts allows each chart to be designed for the corresponding particular sequence of patients for a surgeon or hospital. The effect of estimation error on performance of risk-adjusted Bernoulli CUSUM chart with DPCLs is also examined. Our simulation results show that the in-control performance of risk-adjusted Bernoulli CUSUM chart with DPCLs is affected by the estimation error. The most influential factors are the specified desired in-control average run length, the Phase I sample size and the overall adverse event rate. However, the effect of estimation error is uniformly smaller for the risk-adjusted Bernoulli CUSUM chart with DPCLs than for the corresponding chart with a constant control limit under various realistic scenarios. In addition, there is a substantial reduction in the standard deviation of the in-control run length when DPCLs are used. Therefore, use of DPCLs has yet another advantage when designing a risk-adjusted Bernoulli CUSUM chart. These researches are results of joint work with Dr. William H. Woodall (Department of Statistics, Virginia Tech). Moreover, DPCLs are adapted to design the risk-adjusted CUSUM charts for multiresponses developed by Tang et al. (2015). It is shown that the in-control performance of the charts with DPCLs can be controlled for different patient populations because these limits are determined for each specific sequence of patients. Thus, the risk-adjusted CUSUM chart for multiresponses with DPCLs is more practical and should be applied to effectively monitor surgical performance by hospitals and healthcare practitioners. This research is a result of joint work with Dr. William H. Woodall (Department of Statistics, Virginia Tech) and Mr. Justin Loda (Department of Statistics, Virginia Tech). / Ph. D.

This thesis examines the high peak-to-average-power ratio (PAPR) problem of OFDM and other spectrally-efficient multicarrier modulation schemes, specifically their stringent requirements for highly linear, power-inefficient amplification. The thesis then presents a most intriguing answer to the PAPR-problem in the form of a constant-envelope OFDM (CE-OFDM) waveform, a waveform which employs phase modulation to transform the high-PAPR OFDM signal into a constant envelope signal, like FSK or GMSK, which can be amplified with non-linear power amplifiers at near saturation levels of efficiency. A brief analytical description of CE-OFDM and its suboptimal receiver architecture is provided in order to define and analyze the key parameters of the waveform and their performance impacts. The primary contribution of this thesis is a highly tunable software-defined radio (SDR) implementation of the waveform which enables rapid-prototyping and testing of CE-OFDM systems. The digital baseband processing of the waveform is executed on a general purpose processor (GPP) in the Linux Ubuntu 14.04 operating system, and programmed using the GNU Radio SDR software framework with a mixture of Python and C++ routines. A detailed description of the software implementation is provided, and baseband simulations of the SDR CE-OFDM receiver in additive white Gaussian noise (AWGN) validate the performance of the implemented signal processing. A fully-functional CE-OFDM radio system is proposed in which GPPs executing the software defined transmitter and receiver routines are interfaced with Ettus Universal Software Radio Peripheral (USRP) transceiver front ends. A software testbench is created to enable rapid configuration and testing of the CE-OFDM waveform over all permutations of its parameters, over both simulated and physical RF channels, to draw deeper insights into the characteristics of the waveform and the necessary design considerations and improvements for further development and deployment of CE-OFDM systems. / Master of Science

Various types of current mode control are being used in different applications. Model for current mode control is indispensable for proper system design. Since 1980s, modeling of current mode control has been a hot topic in power electronics field. In current mode control, sub-harmonic oscillation is a common issue, especially for constant frequency current mode control: like peak current mode control, valley current mode control, or average current mode control. Recently V2 control is becoming more and more popular due to its simple implementation ad super fast transient response. V2 control can also run into sub-harmonic oscillation just as current mode control. Efforts have been devoted to modeling of V2 control. A common property of different types of current mode control and V2 control is that they are all multi-loop structures and the inner loops are all highly nonlinear. Due to the nonlinearity of the inner loops, modeling of these structures is extremely difficult. Up to now, there are two main problems which haven't been solved: 1. modeling of average current mode control; 2. modeling of V2 control with composite capacitors. This thesis tries to solve these two problems and starts with V2 control. For V2 control with single type of bulk capacitors, an accurate model has been proposed previously. In this thesis, an equivalent circuit model is proposed to get better physical understanding. This method makes use of previous current mode control modeling result and relates V2 control with current mode control. To model V2 control with composite capacitors, capacitor currents and output voltage time domain waveforms are analyzed. Based on describing function method, transfer function from control to output is derived. The modeling result shows that with more parallel ceramic capacitors, system has smaller stability margin. For average current mode control, the structure is compared with V2 control. Similarity between the structures of current compensator in average current mode and output capacitor network in V2 control is identified. V2 model is utilized for average current mode control. The modeling derivation process is simplified. For the current compensator in average current mode control, it is not desired to have a high frequency pole from stability point of view. As a conclusion, a circuit model for V2 control with bulk capacitors is proposed and another two problems are examined: modeling of V2 control with composite capacitors and modeling of average current mode control. It has been demonstrated that there is similarity between these two structures. The modeling results are verified through simulation and experiments. / Master of Science

Engineers of off-road equipment, on-road vehicles, pavement, and tires must assess the roughness of a terrain surface for the design of their products. The International Roughness Index (IRI), a standardized means of assessing longitudinal road roughness, quantifies roughness based on the average suspension travel for a particular vehicle at a prescribed speed. The Discrete Roughness Index (DRI) developed in this work address fundamental limitations of the IRI. Specifically, the DRI is calculated for each discretely measured location along a terrain surface and is applicable to vehicles traveling at varying speeds and using parameters other than the Golden Quarter-Car on which the IRI is based. The development of the DRI begins with a consistent discretization of the terrain surface, vehicle response, and the IRI. Next the Fractional Response Coefficient is developed, the properties of which are critical in the development of the DRI. The DRI is developed and its properties are discussed through theory and simulation of the ASTM E1926-08 profile. One important property of the average DRI is that it converges to the IRI as the distance between sampled points becomes smaller, for the particular case when the Golden Quarter-Car model is simulated at 80 kph. The DRI is not an alternative to the standard IRI, therefore, but a widely applicable roughness measure of which the standard IRI is a single specialized application. / Master of Science