Dissertation thesis is focused on the aircraft cooling system optimization. The knowledge of the pressure loss is necessity for design of an optimal system. In the beginning of the thesis, the different designs of piston engine cooling systems are described for both aircrafts and automobiles. Other sections outline the determination of the coolers characteristics and describe the different test sections applicable for the characteristics measurement. The test section built at the Institute of the aerospace engineering is described as well. Practical part of the thesis describes the calibration of the test section. The measurements of the specific coolers were compared with the CFD simulation of two models - the simplified cooler model and cooler element model. Additionally, the measurements were compared with results obtained at specialized coolers testing facility.

This study examines whether a trend following managed futures fund can improve its performance by changing its position sizing method. Trades for a simple trend following strategy was simulated on 47 futures contracts over the period 1990-2012, using varying methods for determining position size. Eleven different position sizing methods where investigated, among them Target Volatility, Omega Optimization and correlation ranking methods. Both methods previously detailed in academic papers as well as novel approaches was implemented, and compared to the baseline performance of the strategy. The results from this study show that the Target Volatility method, and to some degree Max Drawdown Minimize and Dynamic Stop Lock-In, improved the performance of strategy. The final recommendation for a trend following managed futures fund is to use Target Volatility as position sizing method, possibly in conjunction with Max Drawdown Minimize. / Denna studie undersöker huruvida en trendföljande managed futures-fond kan förbättra sina resultat genom att ändra positionsskalningsmetod. Handel med en enkel trendföljande strategi simulerades på 47 futureskontrakt åren 1990-2012, för olika metoder att för bestämma positionsstorlek. Elva positionsskalningmetoder undersöktes, exemplevis Target Volatility, Omega Optimization och metoder baserade i korrelationsrankning. Både tidigare beskrivna metoder och nya tillvägagångssätt testades, och jämfördes med den grundläggande strategin med avseende på risk och avkastning. Denna studies resultat visar att framförallt Target Volatility, och i viss uträckning Max Drawdown Minimize och Dynamic Stop Lock-In förbättrade nyckeltalen för den handlade strategin. Den slutgiltiga rekommendationen för en trendföljande managed futures-fond är att använda Target Volatility som positionsskalningsmetod, möjligtvis tillsammans med Max Drawdown Minimize.

Computed tomography (CT) is gaining popularity as a minimally invasive diagnostic modality in veterinary science. The use of contrast agents is well described and used with increasing frequency, having marked benefits over the invasive angiographic procedures used previously. Methods to perform CT angiographic (CTA) studies include bolus tracking, test bolus and empirical scan delay techniques. In human medicine, empirical scan delays have been extensively investigated but due to the marked patient variability encountered in veterinary medicine, this technique cannot, at this stage, be confidently used. This then poses the question that if both techniques can be used, is one significantly better than the other one when performing, in particular thoracic CTA? CTA studies were performed on 6 adult beagles, using the bolus tracking (BT) technique and the test bolus (TB) technique on two separate occasions, at least 2 weeks apart. Each beagle acted as its own control. The patients were placed under general anaesthesia. Two ml/kg of 300mgI/ml iodinated contrast agent was injected through a 20 gauge catheter place in the cephalic vein for the BT technique. Scans were initiated when the contrast in the aorta reached 150 Hounsfield units (HU). For the TB technique, the dogs received a test dose of 15% of 2ml/kg of 300mgI/ml iodinated contrast agent injected manually into the cephalic catheter, followed by a series of low dose sequential scans. Time attenuation curves were generated using dynamic evaluation software programs (DYNAEVA®). The full dose of 2ml/kg of the 300mgI/ml iodinated contrast agent was then administered and the scans were conducted at optimal times as identified from the time attenuation curves. The full dose of contrast administration was administered using a pressure injector operated at 3ml/sec and was followed by a manual saline flush for all studies. Mean attenuation in HU was taken at three consecutive levels in the aorta (Ao), caudal vena cava (CVC) and right caudal pulmonary artery (rCPA) by placing a region of interest (ROI) in the vessel of interest. These observations were done for both arterial and venous phases using the BT and TB studies in five of the six dogs. Additional observations included the visualisation of the smaller thoracic vessels, in particular the arteries, the study duration, milliampere (mA), computed tomography dose index volume (CTDI(vol)), dose length product (DLP) and the pressure and the duration of the contrast injection. These observations were done in all six dogs. Statistical analysis included the comparison of the attenuation achieved in the major vessels (Ao, CVC and rCPA) for the two techniques. The test bolus technique was deemed to be the gold standard, as this is adapted for each individual patient. In one beagle the study was considered non diagnostic with a later time to peak enhancement noted in the Ao than the CVC. No statistical analysis was done on the dynamic and attenuation data obtained in this dog. In all the other studies, the attenuation achieved in the vessels evaluated was deemed to be of diagnostic quality except for the attenuation achieved in the CVC during the arterial phase. The attenuation in the Ao during the arterial phase for the BT technique was not significantly different (660.52 HU ± 138.49) than the TB technique (469.82 HU ± 199.52)(p = 0.13). The attenuation in the Ao during the venous phase for the BT and TB techniques was also not significantly different (BT = 190.6 HU ± 28.29 and TB = 188.8 HU ± 21.9, p = 0.92). The attenuation in the CVC during the arterial phase and venous phases for the BT techniques were not significantly different (arterial phase BT = 37.84 HU ± 20.8, arterial phase TB = 91.48 HU ± 66.54, p = 0.069; venous phase BT = 171.3 HU ± 32.36, venous phase TB = 191.08 HU ± 19.59, p = 0.087). The attenuation in the rCPA during the arterial phase was not statistically different between the two techniques (BT= 606.34 HU ± 143.37 and TB = 413.72 HU ± 174.99, p = 0.28), nor was the attenuation in the rCPA during the venous phase (BT = 174.34 HU ± 27.25 and TB = 164.46 HU ± 18.51, p = 0.51). The mean mA for the BT technique (3538.5 ± 171.27) was significantly lower than the TB technique (3929.6 ± 312.3) with p = 0.024. The mean CTDI(vol) for the BT technique (24.42 ± 11.89) was significantly lower than the TB CTDI(vol) (45.32 ± 0.94) with a p-value of 0.013. The mean DLP did not differ significantly between the two techniques (BT = 139.1 ± 7.65, TB = 162.8 ± 33.1) (p = 0.12). The BT technique resulted in a significantly shorter procedural duration and utilised less contrast material than the TB technique. The injection duration and injection pressures did not differ significantly between the two techniques (p = 0.23 and p = 0.62 respectively). This study identifies that there is no preference for either technique when evaluating the Ao, CVC or rCPA, however, the BT technique is shown to be shorter in procedural duration, utilises less contrast material and results in less radiation dose to the patient when compared to the TB technique. Copyright / Dissertation (MMedVet)--University of Pretoria, 2011. / Companion Animal Clinical Studies / unrestricted

The demand for low-power A/D conversion techniques has motivated the exploration of charge-transfer amplifiers (CTAs) to construct efficient, precise voltage comparators. Despite notable advantages over classical, continuous-time architectures, little is understood about the dynamic behavior of CTAs or their utility in precision A/D converters. Accordingly, this dissertation presents several advancements related to the design and analysis of charge-transfer amplifiers for low-power data conversion. First, an analysis methodology is proposed which leads to a deterministic model of the voltage transfer function. The model is generalized to any timing scheme and can be extended to account for nonlinear threshold modulation. The model is compared with simulation results and test chip measurements, and shows good agreement over a broad range of circuit parameters. Three new charge-transfer amplifier architectures are proposed to address the limitations of existing designs: first, a truly differential CTA which improves upon the pseudo-differential configuration; second, a CTA which achieves more than 10x reduction in input capacitance with a moderate reduction in common mode range; third, a CTA which combines elements of the first two but also operates without a precharge voltage and achieves nearly rail to rail input range. Results from test chips fabricated in 0.6 um CMOS are described. Power dissipation in CTAs is considered and an idealized power consumption model is compared with measured test chip results. Four figures of merit (FOMs) are also proposed, incorporating power dissipation, active area, input charging energy and accuracy. The FOMs are used to compare the relative benefits and costs of particular charge-transfer amplifiers with respect to flash A/D converter applications. The first 10-bit CTA-based A/D converter is reported. It consumes low dynamic power of 600 uW/MSPS from a 2.1 V supply, 40% less than the current state of the art of 1 mW/MSPS. This subranging type converter incorporates capacitive interpolation to achieve a nearly ideal comparator count and power consumption. A distributed sample-and-hold (S/H) eliminates the need for a separate S/H amplifier. A test chip, fabricated in 0.6 um 2P/3M CMOS, occupies 2.7 mm2 and exhibits 8.2 effective bits at 2 MSPS.

Background The Circle of Willis (CoW) is an arterial anastomotic circle located at the skull base thatmaintains collateral circulation in the brain. Variations in CoW anatomy are commonly presentin the general population. Absent or hypoplastic segments of the CoW have been linked to anincreased stroke risk. Aim To determine whether CoW completeness is related to future ischemic stroke in the elderlypopulation after adjustment for relevant clinical risk factors. Methods Consecutive patients that underwent computed tomography angiography (CTA) of the headbetween 2014-2015 (60 years or older) were included. CoW-integrity was determined on CTAimages. Patient journals were retrospectively examined for ischemic stroke events followingthe CTA. Cox proportional hazard regression analysis was performed to determine hazard forischemic stroke in patients with incomplete CoW. Results 147 patients were included. The median follow-up time was 6.4 years (interquartile range 3.0years). 17 ischemic stroke events occurred during the follow-up period. Age (hazard ratio, HR1.10, p=0.03) and gender (HR 0.25, p=0.04) were statistically significant risk factors forischemic stroke. Complete anterior and posterior CoW was associated with lower strokehazards, however, the association was not statistically significant (HR 0.36, p=0.34 and HR0.67, p=0.45, respectively). Conclusions No significant risk reduction could be observed for either complete posterior or anterior CoW.Sex and age were significant risk factors for ischemic stroke. Further research is necessary toinvestigate how CoW integrity influences stroke risk after adjustment for other risk factors.

Nowadays data networks occupy an important place in our world. Since the 70, data communications have evolved quickly, and data transmission, voice and video may be transmitted over different media (copper, fiber optic and wireless). For example we can mention video-conference, voice and IP telephony, electronic transactions, medical applications, industrial sensor networks, etc. These applications are carried out successfully through data communications protocols. Protocol is a set of rules and conventions that govern how network devices exchange information. The protocols analysis lets optimize network performance. The study and analysis of wireless personal area networks (WPAN) are developed in this thesis. The application fields include industrial monitoring and control, public safety, measured in vehicles, smart cards or boards, precision agriculture, among others. WPANs are suitable for use in small mobile devices such as sensors, cell phones, Personal Digital Assistants (PDAs), etc. The Institute of Electrical and Electronics Engineers (IEEE) specifies the rules for this type of networks under the IEEE 802.15.x. It have different subgroups:  802.15.1 specifications based on the Special Interest Group (SIG) Bluetooth.  802.15.2 developed a model of coexistence between networks Wireless LAN (Wireless Local Area Networks) and WPANs  803.15.3 transmission delivers high speed, low power consumption, low cost and quality of service.  802.15.4 researches and develops solutions that require low data transmission and thus the batteries lasting for months or even years. This thesis presents the development of proposals to improve the network performance over different real-time scheduling disciplines. First this thesis describes the standard 802.15.3, and then the real-time scheduling algorithms are explained. Then, several variants at MAC level are developed in order improve the real-time WPAN performance. / Las redes de datos hoy en día ocupan un lugar preponderante en mundo en que vivimos. Desde los años 70 los avances sobre las comunicaciones de datos han evolucionado a gran velocidad, pudiendo en la actualidad realizarse transmisiones de datos, voz y video sobre distintos medios (cobre, fibra óptica y sin cable). Como ejemplo podemos mencionar video-conferencias, voz y telefonía sobre IP, transacciones electrónicas bancarias, aplicaciones en medicina, red de sensores industriales, etc. Para que todas estas aplicaciones se lleven a cabo con éxito son indispensables los protocolos que rigen los distintos tipos de comunicaciones. Tal como su definición lo dice un protocolo es un conjunto de normas y convenciones que rigen de qué forma los dispositivos de una red intercambian información. El estudio de éstos permite optimizar el funcionamiento de la red. En esta tesis se desarrolla el estudio y análisis de redes inalámbricas (wireless) de área personal (WPAN). Este tipo de redes se emplean en diversos campos como control y monitoreo industrial, seguridad pública, medición en automóviles, tarjetas o placas inteligentes, agricultura de precisión entre otros. En términos generales, las redes WPANs son adecuadas para el uso en dispositivos móviles pequeños, como sensores, teléfonos celulares, Asistente Digital Personal (PDAs), etc. El Instituto de Ingenieros Electricistas y Electrónicos (IEEE) especifica las normas para este tipo de redes bajo el estándar IEEE 802.15.x, dentro de éste tenemos distintos subgrupos los cuales son:  802.15.1 basado en las especificaciones del Grupo de Interés Especial (SIG) de Bluetooth.  802.15.2 desarrolló un modelo de coexistencia entre las redes Wireless LAN (Redes de Área Local Inalámbricas) y las WPANs  803.15.3 ofrece alta velocidad de transmisión, bajo consumo de energía, bajo costo y calidad de servicio.  802.15.4 investiga y desarrolla soluciones que requieren una baja transmisión de datos y con ello una duración en las baterías de meses e incluso de años. Sobre el estándar 802.15.3 se desarrollarán distintos análisis que nos permitirán realizar propuestas para mejorar el rendimiento de las redes WPANs bajo distintas disciplinas de diagramación sobre sistemas de tiempo real; este es el objetivo de la tesis. Para cumplir con este objetivo, se inicia explicando el funcionamiento del estándar 802.15.3, luego se enfoca sobre los sistemas de tiempo real analizando los distintos algoritmos de diagramación. A continuación se desarrollan diversas metodologías con el objetivo de mejorar el rendimiento de las redes WPANs evaluando su desempeño bajo distintos esquemas de diagramación en tiempo real.

Ces travaux traitent de la réalisation d’un prototype d’interrupteur monolithique bidirectionnel à base de transistor bipolaire. A terme, l’objectif est de développer un interrupteur intelligent à faible perte, complètement intégrable dans l’habitat. Nous nous intéressons d’abord aux composants bidirectionnels existants. Nous présentons ensuite deux transistors bipolaires bidirectionnels. Le premier à base large, de fabrication aisée. Le second, symétrique, latéral, sur substrat SOI, à base fine, verticale, autoprotégée, très novateur. Nous les comparons et optons pour le transistor latéral à base fine, puis discutons les différentes étapes de sa fabrication et montrons qu’elle constitue un véritable challenge. Ensuite, nous détaillons une méthodologie analytique 1D permettant de déterminer les éléments clefs de fabrication de la partie active du transistor. L’étude est validée par des simulations 2D numériques par éléments finis. Nous continuons par une réflexion sur la périphérie du composant et sa métallisation. Nous détaillons les variantes de réalisation envisagées et montrons que ce composant est robuste vis-à-vis des désalignements entre les masques lors de la fabrication. Finalement, nous caractérisons les transistors fabriqués. Initialement le dispositif est parasité par des effets de ségrégation des dopants aux interfaces SiO2 / Si. Nous expliquons qu’il est possible de contrecarrer ces effets, pour finalement valider le concept de transistor bipolaire symétrique latéral sur substrat SOI, à base fine, verticale, autoprotégée. / This study deals with the realization of a prototype of a low losses monolithic bidirectional switch. It is based on a SOI symmetrical and lateral bipolar transistor with a thin, vertical and shielded base. The goal is to produce a switch which can be integrated to smart electronics functions. First, we compare the existing bidirectional solutions. We then introduce two bidirectional bipolar transistors: one with a wide base, easy to realize, and the other one, patented, symmetrical and lateral, using a SOI substrate, with a thin, vertical and shielded base. We compare these two devices and choose the novel and patented lateral bipolar transistor. We then discuss the challenge of its fabrication. We then detail a 1D analytical methodology allowing to define rapidly the key steps of the active area transistor realization. The study is then confirmed by finite element 2D numerical simulations (Sentaurus). Next, we discuss the periphery and metallization of the device. We detail the variant of process introduced. We finally show that this component is robust to masks misaligning during its fabrication. To the end, the transistors are realized and analyzed. We show that, initially, the segregation of dopants at SiO2 / Si interfaces implies a parasiting canal in parallel of the transistor. We then explain how to reduce these parasiting effects, to finally validate the concept of a symmetrical and lateral bipolar transistor on a SOI substrate, with a thin, vertical, shielded base.

Abstract Computed tomography (CT) and interventional procedures cause relatively high patient doses compared with other imaging modalities in radiology. The optimization of the imaging equipment and procedures is important and necessary due to the known risks caused by ionization radiation. The different irradiation geometries and dose units between imaging techniques complicate the comparison of patients’ radiation exposure. Absorbed doses (D) of organs predict the tissue reactions of the procedures, while effective dose (E) is a useful means of comparing the stochastic effects of the different imaging procedures. A and E can be estimated from dosimetric measurements in a phantom. In vivo measurements are used during a radiological examination. In the present thesis, various properties of radiophotoluminescence dosimeters (RPLD) were investigated to verify the applicability of RPL dosimetry in radiology. The absorbed dose was determined in vaginal fornix of seven pregnant women, in the area near the head of the fetus during prophylactic catheterization before uterine artery embolization (UAE). The evaluation of the radiation exposure of digital subtraction angiography (DSA) and computed tomography angiography (CTA) was performed using an anthropomorphic phantom. Absorbed doses of organs and effective dose were determined for a typical four-vessel angiography of the cerebral area, including intracranial vessels, and for the cervicocerebral area, including both cervical and intracranial vessels. RPLD shows excellent dose linearity and minimal fading. The low dose detection threshold was determined to be 20 µGy with a coefficient of variation (CV) of 12.2%. The methodology is independent of photon energy in the energy range used in radiology. For seven patients studied, the mean value of D in vaginal fornix was 11.2 mGy (range 2.2–28.7) being low dose study when pulsed fluoroscopy with an optimized protocol is used without angiography exposures. In the assessment of cerebral vessels, the effective dose for CTA was approximately one-fifth of the dose compared with DSA. The dose for cervicocerebral vessels CTA was approximately one third higher compared with DSA. Conversion factors from the DAP and the DLP to the effective dose were calculated for the specific angiographic protocols. RPL dosimetry shows sufficient reliability in measuring radiation doses in radiology. / Tiivistelmä Tietokonetomografiakuvauksissa ja toimenpideradiologisissa tutkimuksissa potilaan säteilyaltistus on suhteellisesti korkeampi verrattuna muihin radiologisiin kuvausmenetelmiin. Kuvauslaitteiden erilaiset kuvausgeometriat ja annosyksiköt hankaloittavat potilaan säteilyaltistuksen vertailua eri tutkimusmenetelmien välillä. Kudokseen absorboituneen annoksen (D) avulla arvioidaan ionisoivan säteilyn suoraa haittavaikutusta kudokselle, kun taas efektiivisen annoksen (E) määrittämisellä arvioidaan säteilyn satunnaista haittaa. D ja E voidaan mitata annosmittarilla käyttäen potilasvastinetta. In vivo mittauksella saadaan annostietoa tutkimuksen aikana suoraan potilaasta. Radiofotoluminesenssiannosmittarin (RPLD) soveltuvuus radiologisiin annosmittauksiin varmistettiin tutkimalla mittarin eri fysikaalisia ominaisuuksia. RPLD:a käytettiin intravaginaalisen annoksen mittauksessa seitsemällä raskaana olevalla naisella. Potilaille tehtiin massiivisen verenvuodon ehkäisemiseksi kohdun valtimosuonten katetrointi läpivalaisuohjattuna ennen kohtuvaltimoiden tukkimishoitoa (UAE). Mittaustulokset antavat tietoa kudokseen absorboituneesta annoksesta lähellä syntymätöntä lasta. Tietokonetomografia-angiografia (TTA) ja digitaalinen subtraktioangiografia (DSA) tutkimusten annosmittaukset tehtiin potilasvastineessa käyttäen RPLD:a. Elinannoksista laskettiin efektiivinen annos aivovaltimoiden ja aivokaulavaltimoiden angiografiatutkimuksille. RPL -annosmittausmenetelmän havaittiin olevan lineaarinen ja annoksen häviäminen mittarista on vähäistä. Matalan annoksen mittausraja oli 20 µGy 12.2% toistettavuudella. RPLD on riippumaton säteilyenergiasta radiologiassa käytettävällä energia-alueella. Seitsemän potilaan keskimääräinen absorboitunut annos oli 11.2 mGy (vaihteluväli 2.2–28.7 mGy) emättimen pohjukasta mitattuna. Annostaso osoittaa, että katetrointitoimenpide on matala-annostutkimus, kun toimenpide tehdään läpivalaisuohjauksessa, eikä angiografisia kuvia käytetä. Aivovaltimoiden TTA:n efektiivinen annos oli viidesosa DSA menetelmän annoksesta. Aivokaulavaltimoiden TTA:n efektiivinen annos oli arviolta 35% korkeampi kuin DSA menetelmän annos. Konversiokertoimet DAP ja DLP yksiköistä efektiivisen annoksen laskemiseksi määritettiin tutkimuksessa käytetyille kuvausohjelmille. RPL -menetelmä osoitti riittävää luotettavuutta radiologisten säteilyannosten mittauksessa.

In this master’s thesis there are summarized fundamental findings about air flow velocity measurements and afterwards introduced the measuring track for automobile car vents, measured object and track arrangement and all used components. Velocity field measurement method is how wire anemometry, and apart from 4 main cases measurements of free air jet, there are velocity profiles in piping measured as an input for CFD simulation. Air vent evaluation criteria are described and their assessment is done for specified cases. Further, there is a CFD simulation presented with comparison to experimental data. Uncertainty of flow rate measured and flow rate setting is determined. Improvement points and encountered issues are presented. Visualization and evaluation program coded in MATLAB environment enables easy data display of CTA results and will serve further vent quality assessment.

Title: Study of Cosmic-Ray Acceleration in Extragalactic Objects Author: Stanislav tefnik Department: Institute of Particle and Nuclear Physics Supervisor: RNDr. Dalibor Nosek, Dr. Abstract: This thesis deals with production processes of cosmic γ-rays in astro- physical objects and methods of their detection. Possible interactions leading to the emission of high energy γ-rays in the active galaxy Centaurus A are discussed in this context. Cherenkov Telescope Array is presented as a new experiment fo- cused on the detection of air showers initiated by cosmic γ-rays. Cherenkov light of air showers is studied in the simulations done by CORSIKA simulation tool. Method of data analysis within the framework of Cherenkov telescopes is descri- bed and performed on the dataset of the active galaxy PKS 2155-304. The results include statistical tests of γ-ray source presence and its time variability. Keywords: Cosmic rays, gamma rays, astroparticle physics, high energy astro- physics, acceleration, CTA experiment