Treasures @ UT Dallas
Not a member yet
7697 research outputs found
Sort by
Development of Transformations of Pyridotriazoles and Photoinduced Palladium Hydride-enabled Reactions
No full textThe thesis describes the development of denitrogenative transformations of pyridotriazoles and
photoinduced palladium hydride-enabled reactions. A general and efficient method for arylation,
X–H insertions, and cyclopropanation reactions of pyridotriazoles was developed. This protocol
proceeded via pyridyl carbene intermediates, under mild, transition metal-free, and light-induced
conditions. Furthermore, a Co-catalyzed transannulation reaction of pyridotriazoles with
isothiocyanates and xanthate esters was developed. This method features the conversion of
pyridotriazoles into two N-fused heterocycles, such as imino-thiazolopyridines and oxo-
thiazolopyridine derivatives, via one-step Co(II)-catalyzed transannulation reaction proceeding
though a radical mechanism. Next, three protocols enabled by visible light and palladium hydride
were developed. A variety of substrates, such as diazo compounds, N-tosylhydrazones, strained
molecules, acrylic acids, acrylic amides, and Baylis-Hillman adducts were found to produce hybrid
palladium alkyl radical intermediates under photoinduced and palladium hydride-catalyzed
conditions. These alkyl radical species subsequently participated in diverse transformations,
including Heck reaction, hydroalkenylation and cascade annulation reactions
Advances in Modular and Flying-capacitor Multilevel Converters
No full textIn the era of growing awareness and concern for environmental changes, power-electronic
engineers are leading the transition away from fossil fuels to renewable energy sources. In
general, renewable energy plants, such as wind farms and solar PV farms, are located far
from the consumption centers. Therefore, an efficient transfer of energy over long distances
or via submerged cables is needed at higher voltages. This was made possible by rapid
advancements in high-power converter technology. The voltage source modular multilevel
converter(VSC-MMC) is the workhorse of modern HVDC systems. A major portion of this
thesis is devoted to improvements in the existing MMC technology.
The submodule capacitor is a crucial component in an MMC system that determines its
size, weight, and cost. For offshore installations, the footprint of the converter is an important deciding factor. Therefore, the submodule capacitor sizing is pivotal, whose value is
determined by the peak-to-peak voltage ripple it handles during converter operation. In this
thesis, various techniques were explored to address the issue of voltage ripple in submodule
capacitors. A new submodule topology (SHBSM) was proposed to approach this problem
from a structural point of view. It has been shown with the help of simulation results that the
proposed SHBSM has 34% lower peak-peak ripple compared to the commonly used HBSMs
at low switching frequency operation. In addition, the proposed SHBSM has fewer device
losses compared to HBSM, making it a strong contender.
A control approach was explored to further reduce the submodule capacitor voltage ripple. The higher modulation index(Ma > 1) was investigated in addition to the injection
of second-harmonic circulating current. A wide range of modulation indexes(Ma ∈ [0.6, 2])
and SHCC indexes(a ∈ [−1, 2]) were considered to find the optimal combinations. It has
been identified that at Ma = 2/√3, a = 1, the ripple is at its minimum, which is one-fifth
of the standard operating condition(Ma = 1, a = 0) ripple. In addition to overmodulation
and SHCC injection, the third-harmonic common-mode voltage(THV) injection technique
was explored for further reduction in the voltage ripple. In the presence of THV injection,
Ma = 1.23, a = 0.81 was identified as the overall optimal point which results in the least
voltage ripple in the submodule capacitors. At this point, the ripple is a mere 13% of the
standard operating condition ripple voltage. Detailed mathematical derivations are provided
along with the simulation results to validate the ripple reduction at the identified optimal
operating points.
Submodule capacitor voltages need to be maintained close to their nominal voltages. In
the case of identical and fewer submodules in an MMC arm, a simple PS-PWM technique
ensures this voltage balancing. The required mathematical expressions are derived and
presented to adapt this technique for different types of submodules with different voltage
levels. Moreover, the required modifications to use for overmodulation MMC systems are
presented. The other method, which is suitable for energy balancing in a large number of
submodules with mismatched capacitance, is the sorting-selecting-based technique. In this
thesis, this method was extended to HBSM/FBSM-based hybrid MMCs and DCSM-based
MMCs.
To black-start an MMC from a completely de-energized state, the submodule capacitors
need to be precharged before converter operation. Although various precharging techniques
exist in the literature, they have not considered MMC operating in the over-modulation
range, which requires precharging of submodule capacitors to a higher voltage value. The
precharging technique presented in this thesis addresses this issue. Moreover, precharging
of different types of submodules was considered in this thesis, and detailed derivations and
implementation for each type were detailed. In addition, to eliminate the need for a large
precharging resistor and its corresponding dissipation losses, a modification using a smaller
auxiliary source was also presented. With the help of eight different case studies, the proposed
precharging technique was explained and validated using simulation and real-time hardware-
in-loop results.
Flying-capacitor multilevel converter(FCMC) is another topology that was focused on in this
thesis. Although this multilevel converter was first proposed almost four decades ago, with
the improvement in the high switching frequency capabilities of wide-band-pap devices, it is
grabbing the attention of researchers once again. At a high switching frequency, the power
density of FCMC is very high because of the smaller size requirement for flying capacitors.
However, this topology suffers from the precharging issue. All its flying capacitors need to
be charged to appropriate voltage levels before the converter operates. The conventional
PS-PWM technique, which is popular for its natural voltage balancing capability during
steady-state operation, fails to precharge all its flying capacitors in a single-phase odd-level
FCMC. This issue was resolved with the help of the proposed mPS-PWM. Firstly, the reason
for PS-PWM failure is investigated with the help of the state-space averaged model derived
for 5L-FCMC. Later, the rationale behind the successful precharging using the mPS-PWM is
explained mathematically and validated using extensive simulation analysis. With the help
of a 9L-FCMC hardware prototype designed and built in the lab, the proposed mPS-PWM
is validated for 5L, 7L, and 9L-FCMCs by demonstrating successful precharging. Later, the
proposed mPS-PWM technique was extended to three-phase odd-level FCMCs
Investigating the Role of Patterned Tissue Stiffness, Cell Proliferation and YAP Localization in Embryonic Kidney Development
No full textDuring renal development, the metanephric kidney arises when the ureteric bud forms along the
Wolffian duct and undergoes a series of branching events to build the ureteric tree. The tips of this
tree interact with renal vesicles in the metanephric mesenchyme to induce the formation of
developing nephrons and later fuse with them to form the filtration system within the kidney.
Kidney development, is an understandably a complex process, regulated in parts by GDNF/Ret
and Wnt signaling. A well-formed network of collecting ducts is essential for normal kidney
functioning as defects in branching morphogenesis are thought to be associated with chronic
kidney diseases. Proper renal branching morphogenesis depends crucially on cell proliferation,
and it is observed that proliferation is elevated specifically at the tips of branching ureteric tree.
Whether or not proliferation is developmentally patterned within the developing kidney and what
regulates this pattern of proliferation, is poorly understood. Although tissue mechanics has been
shown to influence the morphogenesis of other branched organs, such as the lung and mammary
gland, it is unclear how biophysical factors within the embryonic kidney, such as tissue stiffness
and cell proliferation, affect renal development and, how changes in the mechanical environment
in the embryonic kidney might interact with signaling cascades, such as the Hippo pathway, or
those downstream of GDNF, Wnt, and TGF-β, which are known regulate renal branching
morphogenesis. In this work, we quantified regional differences in tissue stiffness within the
embryonic kidney and investigated how these variations influence the patterns of proliferation that
sculpt the ureteric tree. We also investigated the effect on branching morphogenesis when patterns
of proliferation and Yap localization were disrupted. Taken together our data will help elucidate
the role and regulation of patterned biophysical factors within the embryonic kidney and its effect
on branching morphogenesis. These findings will further our understanding of branching-related
kidney diseases and can provide better insight towards tissue engineering efforts of rebuilding a
kidney
Data Subset Selection for Compute Efficient Deep Learning
No full textDeep learning has achieved tremendous success in a variety of machine learning applications,
such as computer vision, natural language processing, and speech recognition. However,
the pursuit of better performance has led to larger models and training datasets, increasing training times, energy usage, and carbon emissions. Real-world applications frequently
involve hyperparameter tuning, which necessitates multiple training cycles and further exacerbates computational expenses. Additionally, real-world datasets often exhibit challenges
like class imbalance, out-of-distribution data (OOD), and label noise, emphasizing the need
for a scalable and comprehensive framework for efficient and robust model training. Having
a framework for efficient model training allows for the democratization of these models to
those with limited resources and reduces carbon emissions, allowing us to take a step closer
to greener AI.
This dissertation addresses these challenges by developing theoretically sound and scalable
data subset selection techniques that can identify small, informative data subsets for training
machine learning models with minimal performance loss. We provide theoretical evidence
that our data subset selection approaches are effective by deriving the rate at which the
model converges to its optimal value when trained on the selected subsets using gradient
descent approaches. Additionally, we demonstrate the versatility of our data subset selection
approaches in supervised learning, semi-supervised learning, and hyperparameter tuning
across diverse real-world domains such as natural language processing, image classification,
and tabular data classification. We also empirically demonstrate that some of the proposed
subset selection formulations effectively handle data imperfections like class imbalance, OOD,
and label noise, providing a comprehensive solution for efficient deep learning.
To help practitioners train their models more quickly and robustly, we release a modular,
user-friendly open-source repository called “CORDS,” which includes several data subset
selection implementations for efficient model training and tuning. We hope that our work
will make it easier for researchers and practitioners to develop and deploy deep learning
models efficiently
Projekt Melody: Avatar Sex Work, the Technological Prostitute Imagination, and Intimate Labor
No full textProjekt Melody is an anime avatar sex worker, streamer, and content creator operated by
anonymous individuals. Following their emergence in 2019, Melody has complicated the politics
of the online sex work industry with their novel use of technology and their bodiless
performance, amassing significant income and popularity in the process. An analysis of Projekt
Melody raises questions on how their sex work complicates existing scholarship on sex work
studies, media studies, and the economics and politics of gendered, intimate labor. As a digital
character that is coded as an Asian woman in a technologically mediated body work position,
Melody furthermore illuminates the relationship between technology, Orientalism, and erotics in
a technologically mediated body work position. In this paper, I intend to provide an
intersectional analysis of the ways in which Projekt Melody simultaneously subverts and
reinforces structures of power and provides a compelling glimpse into the future of sex work in
the age of adult content platforms
Study of Wall Turbulence Response to Large-scale Homogeneous and Heterogeneous Surfaces
No full textFully-rough wall-sheared turbulence consists of an inner and outer layer, each with its own distinct characteristics. The inner layer is composed of sinuous structures sustained by an autonomous cycle, while the outer layer boasts inclined parcels of relative momentum deficit and excess. The stair-case pattern of successive uniform momentum zones (UMZs) necessitates the existence of an interfacial shear layer of abrupt velocity change on the wall- normal direction. This phenomenon has been established in prior research and highlights the importance of understanding the structure and behavior of fully-rough wall-sheared turbulence.
A conditional sampling procedure has been leveraged in the LES statistics of fully rough channel flow to generalize the positions of UMZs, which prevail where sublayer interactions create large-scale and arbitrary momentum excesses or deficits. By observing the distribution of the interfacial shear layer during different conditions when fast or slow parcels of fluid pass the sampling location, gained are the insight into the flow structures and their interactions within the roughness sublayer. These observations are consistent with previous studies and provide a new perspective on the structure of fully-rough wall-sheared turbulence.
The relationship between wall roughness obliquity and the flow regime of the fully-rough wall-sheared turbulence has been investigated through parametric assessments. This research estimates between the flow pattern of the internal boundary layer (IBL) when there is large- scale orthogonal roughness heterogeneity, and the secondary flow of the second kind, i.e., the counter-rotating secondary cells, under the circumstance of the heterogeneity parallel to the streamwise direction. Of particular interest is the transition at the obliquity angle, 22π/56, where a critical obliquity is observed and the sheltering area abruptly changes. These findings provide valuable insights into the behavior of fully-rough wall-sheared turbulence and can be used to improve future turbulence models.
Additional research has found that the radial spacing between roughness elements in fully- rough wall-sheared turbulence is a crucial factor in determining the critical obliquity of the wall roughness heterogeneity. By adjusting the radial spacing between adjacent roughness blocks in a row, it is possible to shift the critical obliquity in a predictable way. This theory was discovered by studying turbulent channel flow cases of rows of roughness blocks with different oblique angles and radial spacing. The predictable influence of radial spacing on critical obliquity further highlights the interplay between successive element sheltering, the flow patterns of IBL, and secondary cells, which are all key factors that determine the structure and behavior of turbulence in fully-rough wall-sheared flows, and further to contribute to refining turbulence models and improving the understanding of fully-rough wall-sheared turbulence
Constructing Permutation Arrays of Known Distance
No full textA permutation array (PA) is a set of permutations on n symbols. A PA is said to have
distance d (under some metric) if every pair of distinct permutations in the array has distance
at least d. Commonly used distance metrics include Hamming distance and Chebyshev
distance. PAs of a known distance can be used to construct error-correcting codes and have
applications in communication over noisy channels.
Let M (n, d) represent the maximum size of a permutation array on n symbols with pairwise
Hamming distance d. Let P (n, d) represent the maximum size of a permutation array
on n symbols with pairwise Chebyshev distance d. Exact values of M (n, d) and P (n, d)
are unknown for most values n and d with the exception of some special cases. While
combinatorial upper and lower bounds exist for both M (n, d) and P (n, d), these can often be
improved through empirical techniques. We present several such techniques to construct PAs
under the Hamming and Chebyshev distance metrics, resulting in improved bounds for both
M (n, d) and P (n, d)
Privacy Compliance in U.S. Universities
No full textPrivacy law and compliance with those laws is a complex undertaking. This paper uses a mixed
methods approach to review the scope and breadth of compliance with privacy laws at four-year
universities in the United States. Starting with a Delphi method with privacy professionals
defining the triggers for privacy laws, the laws most important for U.S. universities, and then the
elements of a successful privacy program along with the risk factors for noncompliance, the
researcher then examines publicly available information on a sample population of universities
and lastly performs a legal review based on the Delphi findings and the Document Analysis.
Both scholars and practitioners should find the paper useful. The outcomes identify what data
subjects and activities trigger privacy laws at U.S. universities, what programmatic elements are
required for a privacy compliance program to be successful, and what risk factors universities
face in their privacy compliance efforts. All of this is reviewed through the Complexity Theory
lens, considering both universities and privacy laws as complex adaptive systems
A New Biomaterial for Vaccination: an Aqueous ZIF-8 Crystal Growth to Preserve Antigens and Enhance the Immune Activation
No full textBiomaterials encompass a broad range of applications for medical treatment and can be on the
macroscale, such as heart valves or hip implants, to the microscale and nanoscale, such as stitches,
dental fillings or particles for drug delivery. For the last thirty years, biomaterials with sustained
release properties have been investigated as a method to improve the delivery of vaccines.
Vaccines are considered one of the most significant inventions in the human history, as it has
prevented hundreds of millions of deaths since its invention and is the main cause for extending
the human life expectancy. However, a major issue with current methods of vaccination is their
low stability at room temperature and the requirement for multiple injections to produce an
immune response strong enough to develop long-term memory. For these reasons, the “cold chain”
infrastructure keeps them refrigerated from manufacturer to clinic, ensuring the epitopes in the
proteinaceous vaccines do not unfold. Any failures in this process can lead to the loss of billions
of dollars and, even with this system, the proteinaceous material will denature over time. On top
of that, the necessity of multiple injections to provide immunity it is clear how heavily we rely on
the cold chain. To solve these issues, biomaterials with sustained release properties has been
employed to “cage” proteins within a stabilizing polymer network that prevents conformational
changes and enables storage at room temperature. Herein, methods to encapsulate proteinaceous
material under aqueous conditions within zeolitic imidazole framework-8 (ZIF-8), a well-studied
metal organic framework, is investigated for it’s ability to activate the immune system in mice and
protect vaccines from denaturation
Hardware-assisted Malware Detection for Securing Embedded Systems
No full textIn the era of Internet of Things (IoT), Malware has been proliferating exponentially over the
past decade. Traditional Anti-Virus Software (AVS) is ineffective against modern complex
Malware. In order to address this challenge, researchers have proposed hardware-assisted
Malware detection using Hardware Performance Counters (HPCs). The HPCs are used to
train a set of Machine learning (ML) classifiers, which are deployed as Hardware-assisted
Malware Detectors (HMDs), and used to distinguish benign programs from Malware. Recently, adversarial attacks have been designed by introducing perturbations into HPC traces
to misclassify a program for specific HPCs. The attacks function by inducing sleep and
running dummy benign instructions to bolster the count of incurred HPCs. Furthermore,
HPC-based techniques can suffer from a high false positive rate due to the similar executed instructions in both benign and malicious applications. Lastly, HPC-based detection
can be infeasible in devices that do not possess HPCs or have limited profiling capabilities.
This dissertation extends and explores various improvements to current HPC-based detection schemes in a multi-part operation. First, various different traditional ML classifiers are
evaluated for HPC-based detection and this security is extended to automotive vehicles by
securing an engine control unit from malicious attacks. Second, a Moving Target Defense
(MTD) that dynamically changes the attack surface to jeopardize attackers’ endeavors, as
well as Non-Differential HMDs (ND-HMDS), which use gradient free classifiers, is developed. Third, tailor-made HPCs, which sample assembly instructions from an application’s
dynamic trace, are introduced as a solution for devices without HPCs in addition to providing better fine-grain precision for reducing false positives. Fourth, to further ameliorate the
aforementioned problems, a Sequential Time Series-based Detection (SEQ-TSD) framework
for identifying Malware is proposed that utilizes only a single HPC. Finally, an explainable
HPC-based Malware technique that furnishes the location of the most malicious instruction
is produced for providing human-readable results