Combined Cycle power plants have recently become a serious alternative for standard coal- and oil-fired power plants because of their high thermal efficiency, environmentally friendly operation, and short time to construct. The combined cycle plant is an integration of the gas turbine and the steam turbine, combining many of the advantages of both thermodynamic cycles using a single fuel. By recovering the heat energy in the gas turbine exhaust and using it to generate steam, the combined cycle leverages the conversion of the fuel energy at a very high efficiency. The heat recovery steam generator forms the backbone of combined cycle plants, providing the link between the gas turbine and the steam turbine. The design of HRSG has historically largely been completed using thermodynamic principles related to the steam path, without much regard to the gas-side of the system. An effort has been made using resources at both UK and Vogt Power International to use computational fluid dynamics (CFD) analysis of the gas-side flow path of the HRSG as an integral tool in the design process. This thesis focuses on how CFD analysis can be used to assess the impact of the gas-side flow on the HRSG performance and identify design modifications to improve the performance. An effort is also made to explore the software capabilities to make the simulation an efficient and accurate.

The flow field around a low pressure turbine (LPT) blade cascade model with and without flow control is examined using ejector nozzle (EN) and vortex generator jet (VGJ) geometries for separation control. The cascade model consists of 6 Pak-B Pratt andamp; Whitney low pressure turbine blades with Re = 30,000-50,000 at a free-stream turbulence intensity of 0.6%. The EN geometry consists of combined suction and blowing slots near the point of separation. The VGJs consist of a row of holes placed at an angle to the free-stream, and are tested at two locations of 69% and 10.5% of the suction surface length (SSL). Results are compared between flow control on and flow control off states, as well as between the EN, VGJs, and a baseline cascade with no flow control geometry for steady and pulsatile blowing. The EN geometry is shown to control separation with both steady and pulsatile blowing. The VGJs at 69% SSL are shown to be much more aggressive than the EN geometry, achieving the same level of separation control with lower energy input. Pulsed VGJs (PVGJ) have been shown to be just as effective as steady VGJs, and results show that a 10% duty cycle is almost as effective as a 50% duty cycle. The VGJs at 10.5% SSL are shown to be inefficient at controlling separation. No combination of duty cycle and pulsing frequency tested can eliminate the separation region, with only higher steady blowing rates achieving separation control. Thus, the VGJs at 69% SSL are shown to be the most effective in controlling separation.

An average-value model of a rotating rectifier circuit in a brushless excitation system is set forth, where a detailed simulation is required to extract the essential averaged-model parameters using numerical averaging. In the proposed approach, a synchronous machine model with saturation and cross saturation and an arbitrary rotor network representation that uses a voltage-behind-reactance representation for the field winding of the main machine is proposed. This allows the field winding to be represented as branches in a circuit solver, permitting straightforward simulation with connected circuitry. Also a brushless exciter model is introduced to be compatible with the averaged-model, where the exciter armature windings are represented using a voltage-behind-reactance formulation. The resulting average-value model is verified in time domain against detailed simulation, and its validity is demonstrated in all rectifier modes of operation.

Electricity generation from wind power and other renewable energy sources is increasing, and their variability introduces new challenges to the existing power system, which cannot cope effectively with highly variable and distributed energy resources. The emergence of smart grid technologies in recent year has seen a paradigm shift in redefining the electrical system of the future, in which controlled response of the demand side is used to balance fluctuations and intermittencies from the generation side. This thesis investigates the impact of smart grid technologies on the integration of wind power into the power system. A smart grid power system model is developed and validated by comparison with a real-life smart grid experiment: the Olympic Peninsula Demonstration Experiment. The smart grid system model is then expanded to include 1000 houses and a generic generation mix of nuclear, hydro, coal, gas and oil based generators. The effect of super-imposing varying levels of wind penetration are then investigated in conjunction with a market model whereby suppliers and demanders bid into a Real-Time Pricing (RTP) electricity market. The results demonstrate and quantify the effectiveness of DR in mitigating the variability of renewable generation. It is also found that the degree to which Greenhouse Gas (GHG) emissions can be mitigated is highly dependent on the generation mix. A displacement of natural gas based generation during peak demand can for instance lead to an increase in GHG emissions. Of practical significance to power system operators, the simulations also demonstrate that Demand Response (DR) can reduce generator cycling and improve generator efficiency, thus potentially lowering GHG emissions while also reducing wear and tear on generating equipment. / Graduate

Volcanic eruptions produce a range of hazards which can impact society. The most widespread of these hazards is volcanic ash fall which can impact a range of critical infrastructure. Power systems are particularly vulnerable to ash fall hazards and the resulting impacts may lead to power supply disruption. This can lead to cascading disruption of dependent systems, such as hospitals, water and wastewater treatment plants, telecommunications and emergency services. Typically, large emergency power generator sets are used to provide emergency power supply for essential services during electrical power outages. There has been little study of what impact ash fall exposure will have on generator performance. International experience suggests large generators can experience rapid performance reduction when exposed to high concentrations of suspended or falling ash due to obstruction of air filters and radiators, causing overheating of the engine and shut down of the generator system. However, it is not clear at what ash fall thresholds generators are likely to be disrupted. This research uses custom designed empirical laboratory experiments to investigate the performance of large generators subjected to a range of volcanic ash fall types and intensities, simulating both proximal and distal ash fall exposure from a range of eruptive styles. It also investigates the application of temporary external filters to minimise the ingestion of volcanic ash into generator housings. The results are used to inform recommendations on the likely impacts of ash to generators and the most effective type of mitigation, which maximises filtration whilst maintaining generator performance. Control tests recorded high particle concentrations (~0.006 mg/m3) which indicate substantial ash contamination is possible. Multiple factors were considered to determine the best mitigation measure including the lowest particle concentration, highest air speed and the ease with which the measure could be fitted. The study found material filtration to be the most effective measure; however as the quality of filtration increased, the air speed was reduced and thus so was the volume of air available to the generator engine. Therefore, the type of filtration required is dependent the ash fall intensity. The study also found that a deflection hood is an effective mitigation measure; maintaining airspeed while reducing particle concentrations within the generator. This research informs risk management strategies for critical infrastructure organisations to reduce the risk of generator disruption during volcanic ash falls.

The automatic Algorithm ARI developed in this paper can generate variates from a large class of unimodal discrete distributions. It is only necessary to know the mode of the distribution and to have a subprogram available that can evaluate the probabilities. In a set up step the algorithm constructs a table mountain shaped hat function. Then rejection inversion, a new variant of the rejection method for discrete distributions that needs only one uniform random number per iteration, is used to sample from the desired distribution. It is shown that the expeceted number of iterations is uniformly bounded for all T-concave discrete distributions. Utilizing a simple squeeze or an auxiliary table of moderate size, which is initialized during generation and not in the set up, Algorithm ARI is fast, at least as fast as the fastest known methods designed for the Poisson, binomial and hypergeometric distributions. The set up time of the algorithm is not affected by the size of the domain of the distribution and is about ten times longer than the generation of one variate. Compared with the very fast and well known alias and indexed search methods the set up of Algorithm ARI is much faster but the generation time is about two times slower. More important than the speed is the fact that Algorithm ARI is the first automatic algorithm that can generate samples from discrete distributions with heavy tails. (author's abstract) / Series: Preprint Series / Department of Applied Statistics and Data Processing

The steam generator plays a primary role in the safety and performance of a pressurized water reactor nuclear power plant. The cost to utilities is in the order of millions of Rands a year as a direct result of damage to steam generators. The damage results in lower efficiency or even plant shutdown. It is necessary for the utility and for academia to have models of nuclear components by which research and analysis may be performed. It must be possible to analyse steam generator performance for both day-to-day operational analysis as well as in the case of extreme accident scenarios. The homogeneous model for two-phase flow is simpler in its implementation than the two-fluid model, and therefore suffers in accuracy. Its advantage lies in its quick turnover time for development of models and subsequent analysis. It is often beneficial for a modeller to be able to quickly set up and analyse a model of a system, and a trade-off between accuracy and time-management is thus required. Searches through available literature failed to provide answers to how the homogeneous model compares with the two-fluid model for operational and safety analysis. It is expected to see variations between the models, from the analysis of the mathematics, but it remains to be shown what these differences are. The purpose of this study was to determine how the homogeneous model for two-phase flow compares with the two-fluid model when applied to a u-tube steam generator of a typical pressurized water reactor. The steam generator was modelled in both RELAP5 and in Flownex. A custom script was written for Flownex in order to implement the Chen correlation for boiling heat transfer. This was significantly less detailed than RELAP5’s solution of a matrix of flow regimes and heat transfer correlations. The geometry of the models were based on technical drawings from Koeberg Nuclear Power Plant, and were simplified to a one-dimensional model. Plant data obtained from Koeberg was used to validate the models at 100%, 80% and 60% power output. It was found that the overall heat transfer rate predicted with the RELAP5 two-fluid model was within 1.5% of the measured data from the Koeberg plant. The results generated by the homogeneous model for the overall heat transfer were within 4.5% of the measured values. However, the differences in the detailed temperature distributions and heat transfer coefficient values were quite significant at the inlet and outlet ends of the tube bundle, at the bottom tube sheet of the steam generator. In this area the water-level was not accurately modelled by the homogeneous model, and therefore there was an under-prediction in heat transfer in that region. Large differences arose between the Flownex and RELAP5 solutions due to difference in the heat transfer correlations used. The Flownex model exclusively implemented the Chen correlation, while RELAP5 implements a flow regime map correlated to a table of heat transfer correlations. It was concluded that the results from the homogeneous model for two-phase flow do not differ significantly when compared with the two-fluid model when applied to the u-tube steam generator at the normal operating conditions. Significant differences do, however, occur in lower regions of the boiler where the quality is lower. We conclude that the homogeneous model offers significant advantage in simplicity over the two-fluid model for normal operational analysis. This may not be the case for detailed accident analysis, which was beyond the scope of this study. / Thesis (MIng (Nuclear Engineering))--North-West University, Potchefstroom Campus, 2013.

The steam generator plays a primary role in the safety and performance of a pressurized water reactor nuclear power plant. The cost to utilities is in the order of millions of Rands a year as a direct result of damage to steam generators. The damage results in lower efficiency or even plant shutdown. It is necessary for the utility and for academia to have models of nuclear components by which research and analysis may be performed. It must be possible to analyse steam generator performance for both day-to-day operational analysis as well as in the case of extreme accident scenarios. The homogeneous model for two-phase flow is simpler in its implementation than the two-fluid model, and therefore suffers in accuracy. Its advantage lies in its quick turnover time for development of models and subsequent analysis. It is often beneficial for a modeller to be able to quickly set up and analyse a model of a system, and a trade-off between accuracy and time-management is thus required. Searches through available literature failed to provide answers to how the homogeneous model compares with the two-fluid model for operational and safety analysis. It is expected to see variations between the models, from the analysis of the mathematics, but it remains to be shown what these differences are. The purpose of this study was to determine how the homogeneous model for two-phase flow compares with the two-fluid model when applied to a u-tube steam generator of a typical pressurized water reactor. The steam generator was modelled in both RELAP5 and in Flownex. A custom script was written for Flownex in order to implement the Chen correlation for boiling heat transfer. This was significantly less detailed than RELAP5’s solution of a matrix of flow regimes and heat transfer correlations. The geometry of the models were based on technical drawings from Koeberg Nuclear Power Plant, and were simplified to a one-dimensional model. Plant data obtained from Koeberg was used to validate the models at 100%, 80% and 60% power output. It was found that the overall heat transfer rate predicted with the RELAP5 two-fluid model was within 1.5% of the measured data from the Koeberg plant. The results generated by the homogeneous model for the overall heat transfer were within 4.5% of the measured values. However, the differences in the detailed temperature distributions and heat transfer coefficient values were quite significant at the inlet and outlet ends of the tube bundle, at the bottom tube sheet of the steam generator. In this area the water-level was not accurately modelled by the homogeneous model, and therefore there was an under-prediction in heat transfer in that region. Large differences arose between the Flownex and RELAP5 solutions due to difference in the heat transfer correlations used. The Flownex model exclusively implemented the Chen correlation, while RELAP5 implements a flow regime map correlated to a table of heat transfer correlations. It was concluded that the results from the homogeneous model for two-phase flow do not differ significantly when compared with the two-fluid model when applied to the u-tube steam generator at the normal operating conditions. Significant differences do, however, occur in lower regions of the boiler where the quality is lower. We conclude that the homogeneous model offers significant advantage in simplicity over the two-fluid model for normal operational analysis. This may not be the case for detailed accident analysis, which was beyond the scope of this study. / Thesis (MIng (Nuclear Engineering))--North-West University, Potchefstroom Campus, 2013.

Pulsars, as rotating magnetised neutron stars got much attention during the last 40 years since their discovery. Observations revealed them to be gamma-ray emitters with energies continuing up to the sub 100 GeV region. Better observation of this upper energy cut-off region will serve to enhance our theoretical understanding of pulsars and neutron stars. The H-test has been used the most extensively in the latest periodicity searches, whereas other tests have limited applications and are unsuited for pulsar searches. If the probability distribution of a test statistic is not accurately known, it is possible that, after searching through many trials, a probability for uniformity can be given, which is much smaller than the real value, possibly leading to false detections. The problem with the H-test is that one must obtain the distribution by simulation and cannot do so analytically. For such simulations, random numbers are needed and are usually obtained by utilising so-called pseudo-random number generators, which are not truly random. This immediately renders such generators as useless for the simulation of the distribution of the H-test. Alternatively there exists hardware random number generators, but such devices, apart from always being slow, are also expensive, large and most still don't exhibit the true random nature required. This was the motivation behind the development of a hardware random number generator which provides truly random U(0,l) numbers at very high speed and at low cost The development of and results obtained by such a generator are discussed. The device delivered statistically truly random numbers and was already used in a small simulation of the H-test distribution. / Thesis (M.Sc. (Physics))--North-West University, Potchefstroom Campus, 2005.

Nerve cells are organized into complex networks that comprise the building blocks of our nervous system. Neurons communicate by transmitting messenger molecules released from synaptic vesicles. Alterations in neuronal circuitry and synaptic signaling contribute to a wide range of neurological conditions, often with consequences for movement. Intrinsic neuronal networks in the spinal cord serve to coordinate vital rhythmic motor functions. In spite of extensive efforts to address the organization of these neural circuits, much remains to be revealed regarding the identity and function of specific interneuron cell types and how neuromodulation tune network activity. In this thesis, two novel genes initially identified as markers for spinal neuronal populations were investigated: Slc10a4 and Dmrt3. The orphan transporter SLC10A4 was found to be expressed on synaptic vesicles of the cholinergic system, including motor neurons, as well as in the monoaminergic system, including dopaminergic, serotonergic and noradrenergic nuclei. Thus, it constitutes a novel molecular denominator shared by these classic neuromodulatory systems. SLC10A4 was found to influence vesicular transport of dopamine and affect neuronal release and reuptake efficiency in the striatum. Mice lacking Slc10a4 displayed impaired monoamine homeostasis and were hypersensitive to the drugs amphetamine and tranylcypromine. These findings demonstrate that SLC10A4 is capable of modulating the modulatory systems of the brain with potential clinical relevance for neurological and mental disorders. The transcription factor encoded by Dmrt3 was found to be expressed in a population of inhibitory commissural interneurons originating from the dorsal interneuron 6 (dI6) domain in the spinal cord. In parallel, a genome-wide association study revealed that a non-sense mutation in horse DMRT3 is permissive for the ability to perform pace among other alternate gaits. Further analysis of Dmrt3 null mutant mice showed that Dmrt3 has a central role for spinal neuronal network development with consequences for locomotor behavior. The dI6 class has been suggested to take part in motor circuits but remains one of the least studied classes due to lack of molecular markers. To further investigate the Dmrt3-derived neurons, and the dI6 population in general, a Dmrt3Cre mouse line was generated which allowed for characterization on the molecular, cellular and  behavioral level. It was found that Dmrt3 neurons synapse onto motor neurons, receive extensive synaptic inputs from various neuronal sources and are rhythmically active during fictive locomotion. Furthermore, silencing of Dmrt3 neurons in Dmrt3Cre;Viaatlx/lx mice led to impaired motor coordination and alterations in gait, together demonstrating the importance of this neuronal population in the control of movement.