No description available.

Broiler birds have drastically increased in size over the past few decades through improvements in genetic selection. Physical measurements such as surface area, volume and physical dimensions are needed information in the construction of animal facilities and developing management guidelines. The objective of this study was to develop a three-dimensional scanning methodology to characterize surface area and envelope volume of poultry, livestock, and equine using a commercially available 3D digitizer system. The digitizing system used phase shift moiré to capture images of three test cylinders, four fiberglass animal models, and a live broiler bird. These images were spliced into physical models using image processing software. System accuracy (< 2%) was verified with cylinder models. Recommended camera orientations and placements were established with each species by the use of fiberglass models. The methods will have to be fine-tuned for live animals as observed in the live broiler test.

Effective Field Theory (EFT) provides a systematic framework to study nuclear structure at low energy. The present work uses EFT to study the Coulomb effects in protonproton scattering. The Coulomb phase shift for elastic scattering is calculated. The calculation done here are on a lattice with periodic boundary conditions, and has a hard spherical wall imposed on it. The scattering inside the sphere produces spherical standing waves, the properties of which are exploited to calculate the phase shift and then compared with known theoretical calculations.

No description available.

Nanoparticles have become increasingly useful in the fields of drug delivery and biosensing. In these applications, nanoparticles are frequently exposed to biological fluids, where proteins will spontaneously adsorb to the nanoparticle surface when exposed to a mixture of proteins. This project aims to present a predictive host-guest system for quantifying each residue’s contribution to nanoparticle binding. Initial studies revealed that lysine at position 13 plays a crucial role in the adsorption of GB3 to 15 nm citrate-coated spherical gold nanoparticles (AuNPs). Therefore, we have constructed a library of K13X GB3 variants, and our initial findings confirm that basic residues (Arg) interact more favorably with AuNPs than other polar (Asn) and acidic side (Glu) chains, and a rank-ordering of side-chain affinity for AuNP surfaces could be inferred in further studies. A simple mechanism can be used to interpret these rankings with respect to thermodynamic vs. kinetic control in future studies.

Since the high $T_c$ superconductivity was discovered in iron pnictides in 2008, the interplay between the reduced dimensionality, magnetism and unconventional superconductivity has been attracting renewed interest. Recently, Bao et al. and Tang et al. discovered a series of quasi-one-dimensional (quasi 1D) superconductors: \K($T_c=6.1 K$), \Rb($T_c=4.8 K$), and \Cs($T_c=2.2 K$). In this thesis, we will discuss microscopic investigation of \Cs based on nuclear magnetic resonance techniques. The first chapter is a brief introduction to this series of superconductors. The second chapter is a summary of NMR techniques and theory. In the third part, I summarize $^{133}$Cs NMR and $^{75}$As Nuclear Quadrupole Resonance (NQR) measurements on a powder sample of \Cs ($T_c < 1.6$~K). From the $^{133}$Cs NMR Knight shift $^{133}K$ measured at the Cs1 site, we show that the uniform spin susceptibility $\chi_{spin}$ increases from 295~K to $\sim$ 60~K, followed by a mild suppression; $\chi_{spin}$ then levels off below $\sim$10~K. Low frequency Cr spin dynamics, reflected on $^{75}$As $1/T_1T$ (the nuclear spin-lattice relaxation rate $1/T_1$ divided by temperature $T$), shows an analogous trend as $\chi_{spin}$. Comparison with the results of $1/T_1T$ near $T_c$ with \K($T_c=6.1$~K) and \Rb($T_c=4.8$~K) establishes a systematic trend that substitution of K$^{+}$ ions with larger alkali ions progressively suppresses Cr spin fluctuations together with $T_c$. / Thesis / Master of Science (MSc)

Viruses may acquire mutations that result in a tropism shift. RNA viruses, such as Coronaviruses (CoVs), are susceptible to tropism shifts. A tropism shift occurs when a virus alters the tissue or cell type it infects, which can have important implications for disease pathogenesis, virulence, transmission, and treatment control. Tropism shifts can occur after cross-species jumps, as well as result from within-host evolution. Beyond the human host, CoVs can be highly pathogenic to a wide variety of wildlife and companion animals. A spillover event from animals to humans, resulting in a tropism shift, has occurred at some point in the evolutionary history of all three highly pathogenic human CoVs: severe acute respiratory syndrome coronavirus (SARS-CoV), middle eastern respiratory syndrome coronavirus (MERS), and severe acute respiratory syndrome 2 (SARS-CoV-2). Therefore, studying the evolution of CoVs in non-human animals may be of critical importance for pandemic prevention. This was the focus of my dissertation, to apply state-of-the-art codon models of evolution to a variety of CoV viral sequences to identify how natural selection may alter viral proteins priming them for tropism shifts. Statistical codon models can infer both which codon sites and genes have been subject to positive or negative selection, effectively differentiating signal between random mutations and those that may impact fitness. These models may also compare selection at homologous sites between different phenotypes (i.e., Spike protein sequences isolated from the gastrointestinal tract and those from macrophages) to identify where selection is acting differently between the phenotypes. In chapter 2 I examined a CoV sequence isolated from hospitalized humans in Malaysia that resembled a Canine Coronavirus (CCoV) to investigate how natural selection had shaped the Spike protein sequence in related animal CoV sequences priming it to jump into humans. In chapter 3 I compared the natural selection signals at specific codon positions in the Spike protein from sequences isolated from two separate feline tropisms (gastrointestinal and macrophage) to identity which adaptive mutations may be associated with the tropism shift and subsequent shift in virulence. This was performed on Feline Coronavirus (FCoVs), where almost 90% of all wild and domestic cats are gastrointestinally infected with FCoVs, and infection becomes highly pathogenic as a result of the shift in tropism to the macrophages. Since intra-host evolution can impact tropism shifts, in Chapter 4 I performed a detailed high-throughout analysis of intra-host evolution of RNAseq data of Equine Coronavirus (ECoV), as well as natural selection analyses of related embecoviruses that have colonized the human host. Taken together, I report on novel signals of natural selection across viral proteins, with an emphasis on Spike, on a diverse set of CoV clades that shed light on the complexities of coronavirus evolution as it relates to tropism shifts. / Biology

In the field of machine learning (ML) a very commonly encountered problem is the lack of generalizability of learnt classification functions when subjected to new samples that are not representative of the training distribution. The discrepancy between the training (a.k.a. source) and test (a.k.a.target) distributions are caused by several latent factors such as change in appearance, illumination, viewpoints and so on, which is also popularly known as domain-shift. In order to make a classifier cope with such domain-shifts, a sub-field in machine learning called domain adaptation (DA) has emerged that jointly uses the annotated data from the source domain together with the unlabelled data from the target domain of interest. For a classifier to be adapted to an unlabelled target data set is of tremendous practical significance because it has no associated labelling cost and allows for more accurate predictions in the environment of interest. A majority of the DA methods which address the single source and single target domain scenario are not easily extendable to many practical DA scenarios. As there has been as increasing focus to make ML models deployable, it calls for devising improved methods that can handle inherently complex practical DA scenarios in the real world. In this work we build towards this goal of addressing more practical DA settings and help realize novel methods for more real world applications: (i) We begin our work with analyzing and addressing the single source and single target setting by proposing whitening-based embedded normalization layers to align the marginal feature distributions between two domains. To better utilize the unlabelled target data we propose an unsupervised regularization loss that encourages both confident and consistent predictions. (ii) Next, we build on top of the proposed normalization layers and use them in a generative framework to address multi-source DA by posing it as an image translation problem. This proposed framework TriGAN allows a single generator to be learned by using all the source domain data into a single network, leading to better generation of target-like source data. (iii) We address multi-target DA by learning a single classifier for all of the target domains. Our proposed framework exploits feature aggregation with a graph convolutional network to align feature representations of similar samples across domains. Moreover, to counteract the noisy pseudo-labels we propose to use a co-teaching strategy with a dual classifier head. To enable smoother adaptation, we propose a domain curriculum learning ,when the domain labels are available, that adapts to one target domain at a time, with increasing domain gap. (iv) Finally, we address the challenging source-free DA where the only source of supervision is a source-trained model. We propose to use Laplace Approximation to build a probabilistic source model that can quantify the uncertainty in the source model predictions on the target data. The uncertainty is then used as importance weights during the target adaptation process, down-weighting target data that do not lie in the source manifold.

The EU has proposed an ambitious goal to achieve widespread E-mobility in both the electrical and commercial sectors. To accomplish this, a substantial number of DC fast-charging stations must be built. These power converters, installed in the DC fast-charging stations (DCFC), differ from traditional DCDC converters as they exhibit high power density, reaching tens of kilowatts. In contrast to traditional non-isolated power converters, isolated power converters offer ideal galvanic isolation, providing protection to both the local power grid and electric vehicles. Among the DC power converters designed for industrial applications, the LLC resonant converters and DAB converters (Dual Active Bridge) have gained significant popularity. When compared to LLC converters, DAB converters demonstrate a more flexible input and output power range, as well as a higher power density. Considering these advantages, a 10kW bidirectional DAB power converter has been designed for the purpose of fulfilling the requirements of this thesis project. The thesis is organized into four distinct parts. The first part focuses on conducting a comprehensive literature review to explore the challenges prevalent in the current electrical field. Various DC-DC topologies are compared based on different factors, including component analysis, controllability, safety considerations, and cost-effectiveness. By examining these aspects, potential solutions for Electric Vehicles (EVs) are identified. In the second part, a specific DC-DC converter with a power rating of 10kW is chosen, utilizing the DAB (Dual Active Bridge) topology. The selection is based on the analysis conducted in the literature review. The thesis delves into the issues and technical challenges associated with this choice, such as reactive power, peak current, zero-voltage switching (ZVS), and phase shift modulation. These topics are thoroughly explored and discussed within the literature study. The second part of the thesis involves the establishment of a DAB model, incorporating mathematical equations and physical derivations. This modeling and design section discusses the energy conversion process, starting from fundamental physical formulas and extending to the overall system setup. Utilizing the proposed model, a control method called SPS (Single Phase Shift) modulation is implemented in the circuit to achieve closed-loop control. Within this part, the relationship between current, voltage, and output power is derived and utilized for the design of a PI closed controller. To address challenges associated with SPS control, such as reactive power elimination and peak current suppression, an EPS (Enhanced Phase Shift) control scheme is introduced. The EPS control scheme not only fulfills the basic requirement of power transfer but also optimizes the system’s overall efficiency. In the third part of the thesis, a simulation is developed to validate the accuracy of the proposed DAB model and control methods. Simulations are implemented using Simulink, a widely used software for dynamic system modeling and simulation. Various aspects of the system are evaluated through the simulation, including the leakage inductor current, voltage waveforms on both the primary and secondary sides and output power. These parameters are plotted and analyzed to assess the performance of the DAB model and control methods. Additionally, loss and efficiency analyses are conducted using PLECS, a simulation platform that specializes in power electronics systems. By inputting the datasheet information of the switches and transformer, PLECS enables the evaluation of losses and efficiency within the system. This analysis provides valuable insights into the performance and energy efficiency of the proposed DAB-based converter. In the final part of the thesis, conclusions are drawn based on the theoretical findings and simulation results obtained throughout the study. These conclusions reflect the overall outcomes and implications of the research conducted. Furthermore, the future work section outlines the tasks that remain unfinished or areas that can be explored in subsequent studies. This section serves as a guide for future researchers, highlighting potential directions for further investigation and improvement in the field of DAB-based DC-DC converters for E-mobility applications. By presenting the conclusions and future work, the thesis provides a comprehensive summary of the research conducted, its contributions, and potential avenues for future research and development. / EU har föreslagit ett ambitiöst mål för att uppnå utbredd e-mobilitet inom både den elektriska och kommersiella sektorn. För att åstadkomma detta måste ett stort antal DC snabbladdningsstationer byggas. Dessa effektomvandlare, installerade i DC-snabbladdningsstationerna (DCFC), skiljer sig från traditionella DC-DC-omvandlare eftersom de uppvisar hög effekttäthet och når tiotals kilowatt. I motsats till traditionella icke-isolerade kraftomvandlare erbjuder isolerade kraftomvandlare idealisk galvanisk isolering, vilket ger skydd för både det lokala elnätet och elfordon. Bland likströmsomvandlarna som är designade för industriella applikationer har LLC-resonantomvandlarna och DABomvandlarna (Dual Active Bridge) vunnit betydande popularitet. Jämfört med LLC-omvandlare uppvisar DAB-omvandlare ett mer flexibelt in- och uteffektområde, såväl som en högre effekttäthet. Med tanke på dessa fördelar har en 10kW dubbelriktad DAB-effektomvandlare designats för att uppfylla kraven i detta examensarbete. Avhandlingen är organiserad i fyra distinkta delar. Den första delen fokuserar på att genomföra en omfattande litteraturgenomgång för att utforska de utmaningar som råder inom det nuvarande elektriska området. Olika DCDC-topologier jämförs baserat på olika faktorer, inklusive komponentanalys, kontrollerbarhet, säkerhetsöverväganden och kostnadseffektivitet. Genom att undersöka dessa aspekter identifieras potentiella lösningar för elektriska fordon (EV). I den andra delen väljs en specifik DC-DC-omvandlare med en märkeffekt på 10kW, som använder DAB-topologin (Dual Active Bridge). Urvalet baseras på den analys som gjorts i litteraturöversikten. Avhandlingen fördjupar sig i de problem och tekniska utmaningar som är förknippade med detta val, såsom reaktiv effekt, toppström, nollspänningsomkoppling (ZVS) och fasskiftsmodulering. Dessa ämnen utforskas och diskuteras grundligt inom litteraturstudien. Den andra delen av examensarbetet omfattar upprättandet av en DAB-modell, innefattande matematiska ekvationer och fysiska härledningar. Det här avsnittet om modellering och design diskuterar energiomvandlingsprocessen, med start från grundläggande fysiska formler och sträcker sig till den övergripande systemuppställningen. Genom att använda den föreslagna modellen implementeras en styrmetod som kallas SPS-modulering (Single Phase Shift) i kretsen för att uppnå sluten-loop-styrning. Inom denna del härleds förhållandet mellan ström, spänning och uteffekt och används för konstruktionen av en sluten PI-regulator. För att ta itu med utmaningar förknippade med SPS-kontroll, såsom eliminering av reaktiv effekt och undertryckning av toppström, introduceras ett EPS-kontrollschema (Enhanced Phase Shift). EPS-kontrollsystemet uppfyller inte bara det grundläggande kravet på kraftöverföring utan optimerar också systemets totala effektivitet. I den tredje delen av avhandlingen utvecklas en simulering för att validera noggrannheten hos den föreslagna DAB-modellen och styrmetoderna. Simuleringar implementeras med Simulink, en mycket använd programvara för dynamisk systemmodellering och simulering. Olika aspekter av systemet utvärderas genom simuleringen, inklusive läckans induktorström, spänningsvågformer på både primär- och sekundärsidan och uteffekt. Dessa parametrar plottas och analyseras för att bedöma prestandan hos DABmodellen och styrmetoderna. Dessutom genomförs förlust- och effektivitetsanalyser med hjälp av PLECS, en simuleringsplattform som är specialiserad på kraftelektroniksystem. Genom att mata in databladsinformationen för switcharna och transformatorn, möjliggör PLECS utvärdering av förluster och effektivitet i systemet. Denna analys ger värdefulla insikter om prestandan och energieffektiviteten hos den föreslagna DAB-baserade omvandlaren. I den sista delen av uppsatsen dras slutsatser baserat på de teoretiska fynden och simuleringsresultat som erhållits genom studien. Dessa slutsatser speglar de övergripande resultaten och konsekvenserna av den genomförda forskningen. Vidare skisserar det framtida arbetsavsnittet de uppgifter som förblir oavslutade eller områden som kan utforskas i efterföljande studier. Det här avsnittet fungerar som en guide för framtida forskare, och lyfter fram potentiella riktningar för ytterligare undersökningar och förbättringar inom området DAB-baserade DC-DC-omvandlare för e-mobilitetstillämpningar. Genom att presentera slutsatserna och det framtida arbetet ger avhandlingen en omfattande sammanfattning av den forskning som bedrivs, dess bidrag och potentiella vägar för framtida forskning och utveckling.

The progress in fuel cell technology has resulted in an increased interest towards hydrogen fuel. Consequently, water gas shift reaction has found a renewed significance. Even though iron- and copper-based catalysts have been used for water gas shift reaction for decades, the catalysts are not strong enough to bring carbon monoxide concentration to a level tolerable for a fuel cell working at low temperatures. This study is focused on hydrogen production from water gas shift reaction using a nickel catalyst. Literature review revealed that nickel is one of the promising catalysts for water gas shift reaction. A thermodynamic analysis proved that exothermic water gas shift reaction is thermodynamically favorable at low temperatures but kinetically limited, and vice versa at higher temperatures. Initial experiments using 12 catalysts supported over monolith alumina revealed that nickel supported on ceria-promoted monolith alumina (Ni/CeO2-Al2O3) performed best, especially at 500oC. At this temperature and steam flowrates of 0.1-0.5 ml/min, the nickel catalyst had an activity of 94-99%, H2 yield of 55-61 vol.%, and H2 selectivity of 77-99%. A second set of experiments examined nine nickel based catalysts using different supports (mostly in powder form) which also demonstrated that Ni/CeO2-Al2O3 is the most promising catalyst for high temperature (450oC) water gas shift reaction. When nickel loading was varied from 1 to 8% (w/w), it was apparent that the catalyst performance increased with the nickel loading. Powder alumina resulted in better catalysis than monolith alumina. In this experiment, it was evident that the presence of minor amounts (1% (w/w) of the nickel loading) of a dopant material that included cobalt, chromium, molybdenum, or ruthenium affected the catalytic activity of the primary catalyst. The addition of cobalt or chromium resulted in positive effect on the performance of Ni/CeO2-Al2O3 catalyst. There was no appreciable effect due to the addition of ruthenium, and there was negative effect owing to the presence of molybdenum. Undoped, cobalt-doped, or chromium-doped Ni/CeO2-Al2O3 catalyst performed much better for water gas shift reaction at 450oC than that of a commercial (control) catalyst. A kinetic study revealed that the activation energy of water gas shift reaction over Ni/CeO2-Al2O3 was to be 104.5 kJ/mol.