No description available.

The class of generalized feedback shift register (GFSR) random number generators was a promising method for random number generation in the 1980's, but was abandoned because of some flaws such as poor performance on certain tests for randomness. The poor performance may be due to the choice of primitive polynomials used in the generators, rather than inherent flaws in the method. The original GFSR generators were all based on primitive trinomials. This project examines several alternative choices of primitive polynomials with more than one "interior" term to address this problem and hopefully provide access to good random number generators.

Flow measurements were made on two highly loaded, low pressure turbine blade configurations in a low-speed, linear cascade facility. The L1M blade has a design Zweifel coefficient of 1.34 with a peak cp near 47% cx (mid-loaded) and the Pack B blade has a design Zweifel coefficient of 1.15 with a peak cp at 63% cx (aft-loaded). Flow velocity and surface pressure measurements were taken for Rec=20,000 and 3% inlet freestream turbulence. For these operating conditions, a large separation bubble forms on the blade suction surface, beginning at 59% cx and reattaching at 86% cx on the L1M blade and a non-reattaching bubble beginning at 68% cx on the Pack B. A spanwise row of discrete vortex-generating jets located at 59% cx on the Pack B and 50% cx on the L1M were used as a separation control device and were pulsed at a frequency of 5 Hz with a duty cycle of 25%. The Pack B with its open separation bubble proved to be a better candidate for VGJ control than the L1M with its closed separation bubble. Further studies were made on the Pack B blade comparing wake and VGJ effects. A wake generator was used to simulate the periodic passing of upstream wakes through the blade passage for the Pack B configuration. The wake passing frequency of 4.5Hz was set to match a typical engine flow coefficient for a low pressure turbine. Data were taken using PIV and a hot-film anemometer mounted on a blade following device. Velocity, turbulence, and intermittency measurements were made along the suction surface of the blade to characterize the bubble dynamics and transitional behaviors for both the presence of unsteady wakes and pulsing VGJs. The wakes caused early breakdown of the separated free shear layer resulting in a thinning of the separation region. The VGJs caused an upstream disturbance which convects downstream, temporarily pushing off the separation bubble. Overall, both wakes and VGJs suppress the size of the steady-state separation bubble, though through different mechanisms. Three-dimensional aspects of the jet disturbance are studied by investigating the effects of the VGJs at two spanwise locations.

In this paper we perform a statistical analysis of dark counts in superconducting nanowire single photon detectors (SNSPDs) with the end goal of creating a quantum random number generator (QRNG) using these dark counts. We confirm that dark counts are Poissonian for low bias currents and that no afterpulsing is present. However, we also show that an increase in bias current causes the dark counts to violate the independence assumption. For the non-Poissonian dark counts we identify three seemingly similar effects and confirm that: (i) a single event is at times regarded as two by the flat-threshold discriminator in the time-tagging device; (ii) a reflection in the readout circuit incites a second detection event shortly after the arrival of a first one, creating a conditionality between dark counts; (iii) a damped oscillation in the effective bias current immediately after a detection event shows itself in the inter-arrival time probability distribution. Finally, we present and evaluate a method for generating random numbers using the Poissonian dark counts as an entropy source with promising results.

The thesis discusses the development of closed loop system to control the DC voltage for 200 kW induction generator rated at a speed of 62000 RPM under different load conditions. The voltage regulation has been implemented using PI controller. A gain scheduling control algorithm has been developed to select the appropriate controller gains with respect to the generator load. Further, a relationship between the generator loads and the controller gains has been established. This relationship has been modeled using adaptive control technique to vary the gains automatically at any load condition. The adaptive control technique has been successfully generalized for real time DSP implementation to regulate the DC voltage for high speed induction generators rated from 5 kW to 200 kW.

This thesis combines Keysight’s SystemVue software with a Vector Signal Analyzer (VSA) and Vector Signal Generator (VSG) to test receiver equalization schemes over physical channels. The testing setup, “Equalization Verification,” is intended to be able to evaluate any equalization scheme over any physical channel, and a decision-directed feed-forward LMS equalizer is used as an example. The decision-directed feed-forward LMS equalizer is shown to decrease the BER from 10-2 to 10-3 (average of all trials) over a CAT7 and CAT6A cable, both simulated and physical, for 1GHz and 2GHz carrier, and 80MHz data rate. A wireless channel, 2.4GHz Dipole Antenna, is also tested to show that the addition of the equalization scheme decreases BER from 10-5 to less than 10-5. Then the simulation and equalization parameters (LMS step size, PRBS, etc.) are changed to further verify the equalization scheme. The simulated channel BER results do not always match the physical channel BER results, but the equalization scheme does decrease BER for both wired and wireless channels. Then transistor-based equalization model is created using both HDL SystemVue components and blocks easily implemented by transistors. The model is then verified using HDL, Spice, and SystemVue simulation. Overall this thesis accomplishes its goal of creating a testing setup, Equalization Verification, to show that adding a given simulated equalization scheme in SystemVue can improve the quality of the link, by decreasing BER by at least an order of magnitude, over a specific physical channel.

The development and evaluation of the design of a Micro-Ram Air Turbine (µRAT), a device being developed to provide power for an autonomous boundary layer measurement system, has been undertaken. The design tools consist of a rotor model and a generator model. The primary focus was on developing and evaluating the generator model for the prediction of generator brake power and output electrical power with and without rectification as a function of shaft speed and electrical load, with only basic manufacturer specifications given as inputs. A series of motored generator evaluation test were conducted at speeds ranging from 9,000 to 25,000 rpm for loads varying between 1 and 3.02 Ohms with output power of up to 80 Watts. Results demonstrated that predicted generated power was at or below 3% error when compared to measured results with about 1% uncertainty. A rotor model was also developed using basic blade element theory. This model neglected induced flow effects and was therefore expected to over predict rotor torque and power. A second rotor model that includes induced flow effects, the open source program X-Rotor, was also used to predict rotor power and for comparison to the blade element rotor model results. Both rotor models were evaluated through wind tunnel validation tests conducted on a turbine generator with two different 3.25 in diameter rotors, rotor-1 (untwisted blades) and rotor-2 (twisted blades). Wind tunnel validation test airspeeds varied between 71-110 mph with electrical loads ranging from 1-20 ohms. Results indicated power predictions to be 50-75% higher for the blade element model and 20-30% for X-Rotor results. The blade element rotor model was modified by applying the Prandtl tip-loss factor to approximately account for the induced flow effects; this addition brought predictions much closer to X-Rotor results. Based on the motor-driven generator test results, it is believed that most of the discrepancy in baseline rotor/generator validation test between predicted and observed power generated is due to inaccuracy in the rotor performance modelling with likely contributors to error being induced flow effects, crude section lift/drag modelling, and aero-elastic deformation. It is concluded that the proposed generator model is sufficient although direct torque measurements may be desired and further development of the µRAT design tools should focus on an improved rotor performance model.

Generating random numbers can be accomplished through various methods, with the primary distinction lying between pseudo-random number generators (PRNGs), which are commonly used for applications that require a large amount of random data, and true random number generators (TRNGs), which are commonly used for applications that need security and unpredictability. This thesis explores the feasibility of harnessing frequency variations in the electrical grid as a source of entropy for a TRNG. By employing an iterative approach, the study has substantiated the likelihood that frequency fluctuations can serve as a reliable source of ran-domness for a TRNG. This assertion is supported by statistical testing using the comprehensive RNG testing suite known as DieHarder, where the final implementation of the TRNG yielded favourable outcomes. Nevertheless, it is worth noting that the artefact exhibited weaker resultson three specific tests within the suite, which can likely be attributed to a limited amount of generated data. Despite these limitations, the findings are undeniably promising, and futurere search endeavours should focus primarily on enhancing the generation speed of the TRNG. By doing so, it is anticipated that improved performance on the DieHarder suite and similar RNG testing suites can be achieved. / Generering av slumptal kan åstadkommas mev hjälp av flera olika metoder. De två stora grupperna är pseudo-slumptalsgeneratorer (PRNG:er), som vanligtvis används för applikationer som kräver en stor mängd slumpmässiga data, och sanna slumptalsgeneratorer (TRNG:er), som ofta används för applikationer som behöver säkerhet och oförutsägbarhet. Detta examensarbete undersöker möjligheten att utnyttja frekvensvariationer i det elektriska nätetverket som en källa till entropi för en TRNG. Genom att använda ett iterativt tillvägagångssätt har studien underbyggt sannolikheten att frekvensfluktuationer kan fungera som en pålitlig källa till slumpmässighet för en TRNG. Detta påstående stöds av statistiska tester med den omfattande RNG-testsviten känd som DieHarder, där den slutliga implementeringen av TRNG:n gav gynnsamma resultat. Det är värt att notera att artefacten visade svagare resultat på tre specifika tester inom sviten, vilket sannolikt kan tillskrivas en begränsad mängd genererad data. Trots dessa begränsningar är resultaten onekligen lovande, och framtida forskningsansträngningar bör främst fokusera på att öka generationshastigheten för generatorn. Genom att göra det förväntas det att bättre resultat från DieHarder och liknande RNG-testsviter kan uppnås.

The objectives of this study were to implement force sensors on the feet of a bipedal robot and analyze their response at different conditions. The data will be used to design a control strategy for the robot. The powered joints of the robot are driven by linear motors. A force sensor circuit was made and calibrated with different kinds of weight. A trajectory generator and inverse kinematics calculator for the robot were made to control the robot walking movement in an open-loop manner. The force data were taken at a certain period of time when the robot was in a standing position. Experiments with external disturbances were also performed on the robot. The ZMP position and mass of the robot were calculated by using the data of force sensors. The force sensor circuit was reliable in taking and handling the data from the sensor although the noise from the motors of the robot was present. / <p>Validerat; 20111115 (anonymous)</p>

Yes / Hybrid AC-DC microgrid (HMG) allows direct integration of both AC distributed generators (DGs) and DC DGs, AC and DC loads into the grid. The AC and DC sources, loads are separate out and are connected to respective subgrid mainly to reduce the power conversion, thus the overall efficiency of the system increases. This paper aims to introduce a novel hybrid AC-DC microgrid planning and design model within a microgrid market environment to maximize net social welfare (NSW). NSW is defined as present value of total demand payment minus present value of total planning cost including investment cost of distributed energy sources (DERs) and converters, operation cost of DERs and the cost of energy exchange with the utility grid subject to network constraints. Scenario Tree approach is used to model the uncertainties related to load demand, wind speed and solar irradiation. The effectiveness of the proposed model is validated through the simulation studies on a 28-bus real hybrid AC-DC microgrid.