At the core of the theoretical part of this thesis is a review of recent attempts at incorporating income distribution in economy-wide models. Most existing models are very detailed and complex but the mechanisms which dominate the results are simple and determined by conveniently extreme assumptions about prices and wages. Typically, prices are either all rigid or all flexible; the treatment of agriculture is unsatisfactory; and dualism in the labour market is ignored. A convincing case for the alleged insensitivity of the distribution of income to policy interventions can not be based on these models. The major part of the thesis describes two models, for Tanzania and Kenya, and their results. In the first model the emphasis is on the effects of migration and economic growth on the urban income distribution. It describes an economy in which neither prices nor wages respond to changes in market conditions. Distinctive features of the Kenya model include its submodel for smallholder agriculture, educational 'bumping' in the labour market and its emphasis on trade and pricing policies. The models are used to estimate the effects of policies (wages, trade, pricing, investment allocation, land redistribution) on growth and income distribution. The results contradict some of the conventional wisdom about the two countries. The models explicitly recognise a number of market imperfections, especially in the labour market, which are crucial in determining distributional changes but which are usually ignored in modelling. The results reveal a greater sensitivity of the distribution of income to changes in policies than has been found with some other models.

Although power system reliability analysis is a mature research area, there is a renewed interest in updating available network models and formulating improved reliability assessment procedures. The main driver of this interest is the current transition to a new flexible and actively controlled power supply system with a high penetration of distributed generation (DG) and energy storage (ES) technologies, wider implementation of demand-side management (DSM) and application of automated control, monitoring, protection and communication infrastructures. One of the aims of this new electricity supply network (’the smart grid’) is an improved reliability and power quality performance, realised through the delivery of an uninterrupted and high-quality supply of electrical energy. However, there is currently no integrated methodology to measure the effects of these changes on the overall system reliability performance. This PhD research aims to update the standard power system simulation engine with improved numerical software models offering new capabilities for the correct assessment of quality of supply in future electricity networks. The standard reliability analysis is extended to integrate some relevant power quality aspects, enabling the classification of short and long supply interruptions by the correct modelling of network protection and reconfiguration schemes. In addition, the work investigates the formulation and analysis of updated reliability indicators for a more accurate validation and benchmarking of both system and end-user performance. A detailed database with typical configurations and parameters of UK/European power systems is established, providing a set of generic models that can correctly represent actual distribution networks supplying a mix of residential, commercial and industrial demand for different load sectors. A general methodology for reducing system complexity by calculating both electrical and reliability equivalent models of LV and MV distribution networks is also presented. These equivalent models, based on the aggregation of individual component models, help to reduce calculation times while preserving the accuracy assessment of network’s reliability performance at bulk supply points. In addition, the aggregated counterparts (same and mixed-type) of different ’smart’ component models (DG, ES and DSM) are also included in the analysis, showing how their co-ordinated implementation and control could improve quality of supply. Conventional reliability assessment procedures are also extended in this thesis to include accurate reliability equivalent models, network contingency statistics, actual load profiles and empirical fault probability distributions, which are employed to assess the frequency and duration of interruptions in the supply system for different scenarios. Both analytical and probabilistic simulation techniques (Monte Carlo method) are developed to include up-to-date security of supply legislation, introducing a new methodology for calculating the standard set of indices reported annually to energy regulators.

Nontrivial domains can be difficult to understand and the task of encoding a model of such a domain can be difficult for a human expert, which is one of the fundamental problems of knowledge acquisition. Model learning provides a way to address this problem by allowing a predictive model of the domain's dynamics to be learnt algorithmically, without human supervision. Such models can provide insight about the domain to a human or aid in automated planning or reinforcement learning. This dissertation addresses the problem of how to learn a model of a continuous, dynamic domain, from sensory observations, through the discretisation of its continuous state space. The learning process is unsupervised in that there are no predefined goals, and it assumes no prior knowledge of the environment. Its outcome is a model consisting of a set of predictive cause-and-effect rules which describe changes in related variables over brief periods of time. We present a novel method for learning such a model, which is centred around the idea of discretising the state space by identifying clusters of uniform density in the probability density function of variables, which correspond to meaningful features of the state space. We show that using this method it is possible to learn models exhibiting predictive power. Secondly, we show that applying this discretisation process to two-dimensional vector variables in addition to scalar variables yields a better model than only applying it to scalar variables and we describe novel algorithms and data structures for discretising one- and two-dimensional spaces from observations. Finally, we demonstrate that this method can be useful for planning or decision making in some domains where the state space exhibits stable regions of high probability and transitional regions of lesser probability. We provide evidence for these claims by evaluating the model learning algorithm in two dynamic, continuous domains involving simulated physics: the OpenArena computer game and a two-dimensional simulation of a bouncing ball falling onto uneven terrain.

Lubricants and liquid cooling are used in many industrial applications to ensure reliability and longevity of machinery. Oil cooling of both electrical and mechanical applications is of interest since oil is better suited for electrical applications than water and already available in the system as a lubricant. Brazed plate heat exchangers (BPHEs) have many advantages compared to other heat exchanger types commonly used in oil cooling applications. Flow maldistribution inside BPHEs can arise with highly viscous fluids like oil. Since flow is hard to measure when plate heat exchangers are brazed together, Computational Fluid Dynamics (CFD) can be used instead. This study investigates parameters that could affect flow distribution inside BPHEs with the CFD-tool COMSOL Multiphysics. The study is made on three different geometries at different detail levels. The purpose of the study is to expand the knowledge about fluid behavior in BPHEs and how it affects efficiency. It was proved from the Bernoulli equation that flow velocity, gravity and Reynolds number were some parameters that could affect flow distribution inside BPHEs. Two simplified models were built for evaluation of viscosity, gravity and Reynolds number. A more detailed model was provided by SWEP representing the fluid domain of a full-size distribution zone model. Model validation and mesh independence study were made with expressions due to the lack of experimental data. Investigations of viscosity, gravity and Reynolds number were made through isolation and alteration of the respective parameter. The validation and mesh independence study proved the models trustworthy and detailed enough to capture the physical behavior. Small deviations from expected validation results can be explained with the assumptions and simplifications made in the process. Results show that flow maldistribution increases with viscosity differences between channels. Viscosity maldistribution is greater for oil than for water. It is important to consider how the fluid viscosity changes with temperature under the respective working conditions. Gravity has no effect on flow distribution as long as it acts along or opposite the main flow direction. As plate heat exchangers are generally placed vertically, gravity will not affect flow distribution. Gravity has a significant effect on flow distribution if plate packages are places horizontally. High Reynolds numbers have a positive effect on flow distribution and reduce the difference between highest and lowest velocities across the outlet. Very low flow velocities should therefore be avoided since it increases flow maldistribution.

No description available.

International Telemetering Conference Proceedings / October 20-23, 2003 / Riviera Hotel and Convention Center, Las Vegas, Nevada / The Western Aeronautical Test Range (WATR) operates the ground stations for research vehicles operating at the NASA Dryden Flight Research Center (DFRC). Recently, the WATR implemented a new system for distributing Time, Space, and Position Information (TSPI) data. The previous system for processing this data was built on archaic hardware that is no longer supported, running legacy software with no upgrade path. The purpose of the Radar Information Processing System (RIPS) is to provide the ability to acquire TSPI data from a variety of sources and process the data for subsequent distribution to other destinations located at the various DFRC facilities. RIPS is built of commercial, off the shelf (COTS) hardware installed in Personal Computers (PC). Data is transported between these computers on a Gigabit Ethernet network. The software was developed using C++ with a modular, object-oriented design approach.

Thesis (MComm)—Stellenbosch University, 2015. / ENGLISH ABSTRACT: In recent decades, economists and sociologists have taken an increasing interest in the study of income attainment and income inequality. Many of these studies have used census data, but social surveys have also increasingly been utilised as sources for these analyses. In these surveys, respondents’ incomes are most often not measured in true amounts, but in categories of which the last category is open-ended. The reason is that income is seen as sensitive data and/or is sometimes difficult to reveal. Continuous data divided into categories is often more difficult to work with than ungrouped data. In this study, we compare different methods to convert grouped data to data where each observation has a specific value or point. For some methods, all the observations in an interval receive the same value; an example is the midpoint method, where all the observations in an interval are assigned the midpoint. Other methods include random methods, where each observation receives a random point between the lower and upper bound of the interval. For some methods, random and non-random, a distribution is fitted to the data and a value is calculated according to the distribution. The non-random methods that we use are the midpoint-, Pareto means- and lognormal means methods; the random methods are the random midpoint-, random Pareto- and random lognormal methods. Since our focus falls on income data, which usually follows a heavy-tailed distribution, we use the Pareto and lognormal distributions in our methods. The above-mentioned methods are applied to simulated and real datasets. The raw values of these datasets are known, and are categorised into intervals. These methods are then applied to the interval data to reconvert the interval data to point data. To test the effectiveness of these methods, we calculate some measures of inequality. The measures considered are the Gini coefficient, quintile share ratio (QSR), the Theil measure and the Atkinson measure. The estimated measures of inequality, calculated from each dataset obtained through these methods, are then compared to the true measures of inequality. / AFRIKAANSE OPSOMMING: Oor die afgelope dekades het ekonome en sosioloë ʼn toenemende belangstelling getoon in studies aangaande inkomsteverkryging en inkomste-ongelykheid. Baie van die studies maak gebruik van sensus data, maar die gebruik van sosiale opnames as bronne vir die ontledings het ook merkbaar toegeneem. In die opnames word die inkomste van ʼn persoon meestal in kategorieë aangedui waar die laaste interval oop is, in plaas van numeriese waardes. Die rede vir die kategorieë is dat inkomste data as sensitief beskou word en soms is dit ook moeilik om aan te dui. Kontinue data wat in kategorieë opgedeel is, is meeste van die tyd moeiliker om mee te werk as ongegroepeerde data. In dié studie word verskeie metodes vergelyk om gegroepeerde data om te skakel na data waar elke waarneming ʼn numeriese waarde het. Vir van die metodes word dieselfde waarde aan al die waarnemings in ʼn interval gegee, byvoorbeeld die ‘midpoint’ metode waar elke waarde die middelpunt van die interval verkry. Ander metodes is ewekansige metodes waar elke waarneming ʼn ewekansige waarde kry tussen die onder- en bogrens van die interval. Vir sommige van die metodes, ewekansig en nie-ewekansig, word ʼn verdeling oor die data gepas en ʼn waarde bereken volgens die verdeling. Die nie-ewekansige metodes wat gebruik word, is die ‘midpoint’, ‘Pareto means’ en ‘Lognormal means’ en die ewekansige metodes is die ‘random midpoint’, ‘random Pareto’ en ‘random lognormal’. Ons fokus is op inkomste data, wat gewoonlik ʼn swaar stertverdeling volg, en om hierdie rede maak ons gebruik van die Pareto en lognormaal verdelings in ons metodes. Al die metodes word toegepas op gesimuleerde en werklike datastelle. Die rou waardes van die datastelle is bekend en word in intervalle gekategoriseer. Die metodes word dan op die interval data toegepas om dit terug te skakel na data waar elke waarneming ʼn numeriese waardes het. Om die doeltreffendheid van die metodes te toets word ʼn paar maatstawwe van ongelykheid bereken. Die maatstawwe sluit in die Gini koeffisiënt, ‘quintile share ratio’ (QSR), die Theil en Atkinson maatstawwe. Die beraamde maatstawwe van ongelykheid, wat bereken is vanaf die datastelle verkry deur die metodes, word dan vergelyk met die ware maatstawwe van ongelykheid.

Thesis (MBA)--Stellenbosch University, 1989. / Some digitised pages may appear illegible due to the condition of the original hard copy. / ENGLISH ABSTRACT: see fulltext for abstract / AFRIKAANSE OPSOMMING: sien volteks vir opsomming

Thesis (PhD)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: The geographic genetic population structure and evolutionary history of the African four-striped mouse, Rhabdomys pumilio, was investigated using mitochondrial (mtDNA) cytochrome b gene (1140 bp) and control region (994 bp) sequences and a combination of cytogenetic banding techniques (G- and C-banding), and fluorescence in situ hybridization. Two cytotypes (2n = 46 and 2n = 48) were identified by cytogenetic analysis. No evidence of diploid number variation within populations was found nor were there differences in gross chromosome morphology, or subtle interchromosomal rearrangements at levels detected by ZOO-FISH. The comparative painting data (using the complete suite, N = 20, of Mus musculus chromosome specific painting probes) show that 10 mouse chromosomes have been retained as chromosomal arms, or intact chromosome blocks within the R. pumilio genome, six produced double signals, while the remaining four hybridized to three or more R. pumilio chromosomes. In total, the 20 mouse chromosome paints detected 40 segments of conserved synteny. Their analysis revealed eight R. pumilio specific contiguous segment associations, a further two that were shared by R. pumilio and other rodents for which comparable data are available, the Black (Rattus rattus) and Norwegian (Rattus nONegicus) rats, but not by the Chinese hamster, Cricetulus grise us. The results suggest that mouse chromosomes 1, 10, and 17 have undergone extensive rearrangements during genome evolution in the murids and may be useful markers for enhancing our understanding of the mode and tempo of chromosome evolution in rodents. Following initial studies using control region sequences, the phylogeographic appraisal of R. pumilio was done using cytochrome b gene sequences. Analyses based on a variety of analytical procedures resulted in the detection of two major mtDNA lineages that correspond roughly to the xeric and mesic biotic zones of southern Africa. One clade comprises specimens with 2n = 48, and the other representatives of two cytotypes (2n = 48 and 2n = 46). The mean sequence divergence (12.0%, range 8.3% -15.6%) separating the two mtDNA clades is comparable to among-species variation within murid genera suggesting their recognition as distinct species, the prior names for which would be R. dilecfus and R. pumilio. Low sequence divergences and the diploid number dichotomy within the mesic lineage support the recognition of two subspecies corresponding to R. d. dilecfus (2n = 46) and R. d. chakae (2n = 48). The data do not support subspecific division within the nominate, R. pumilio. Molecular dating places cladogenesis of the two putative species at less than 5 million years, a period characterised by extensive climatic oscillations which are thought to have resulted in habitat fragmentation throughout much of the species' range. / AFRIKAANSE OPSOMMING: Die geografiesebevolkingsstruktuur en evolusionêre verwantskappe binne die Afrika streepmuis, Rhabdoys pumilio, is ondersoek deur middel van mitochondriale ONS volgordebepaling van die geenfragment sitochroom b (1140 basispare) en die reguleerstreek (994 bp) in kombinasie met sitogenetiese tegnieke (G- en Cbandkleuring en f1uoreseerende in situ hibridisasie). Twee sitotipes (2n = 46 en 2n = 48) is geidentifiseer deur sitogenetiese analasie. Geen bewys van variasie in die 2n chromosoomgetal binne bevolkings is gevind nie. Verder is daar ook geen verskil in die morfologies struktuur van chromosome aanwesig binne bevolkings nie. Vergelykende data (verkry met behulp van die N = 20 Mus musculus chromosoomspesifiekepeilers) dui daarop dat 10 muis chromosome behoud gebly het as chromosoomarms of chromosoomblokke binne die R. pumilio genoom. Ses peilers het dubbel seine gelewer terwyl die oorblywende vier peilers gehibridiseer het aan drie of meer R. pumilio chromosome. In totaal het die 20 muischromosoomverwe 40 konserwatiewe segmente geidentifiseer. Die analise dui agt R. pumilio spesifieke aaneenlopende segmentassosiasies aan, met 'n addisionele twee wat deur R. pumilio en ander muisagtiges vir wie vergelykende data beskikbaar is, byvoorbeeld die swart (Rattus rattus) en Noorweegse (R. norvegicus) rot maar nie die Chinese hamster, Cricetulus grise us, gedeel word. Die resultate stel voor dat muischromosoom 1, 10 en 17 ekstensiewe herrangskikkings ondergaan het gedurende die genoom evolusie binne die Muridae en dat hulle waarskynlik waardevolle merkers kan wees om beide die patroon en tempo van chromosome evolusie in muisagtiges verder te kan verstaan. Die filogeografiese verwantskappe binne R. pumilio is ondersoek deur middel van ONS volgordebepalings van die reguleerstreek asook sitochroom b. Die resultate van hierdie studie het twee divergente mitochondriale ONS eenhede ontdek wat gekorreleer kan word met xeriese en mesiese klimaatsones binne suidelike Afrika. Een groep bestaan uit diere met 2n = 48, terwyl die ander genetiese groep twee sitotipes (2n = 46 en 2n= 48) insluit. 'n Gemiddelde genetiese divergensie van 12.0% (varieer tussen 8.3% - 15.5%) verdeel die twee mtDNS-groepe en is vergelykbaar met tussenspesievariasie binne ander muisagtige genera, wat moontlik daarop dui dat twee verskillende spesies teenwoordig is; die voorgestelde name is R. di/ectus en R. pumilio. Lae genetiese divergensie binne die mesiese groep versterk die moontlike teenwoordigheid van twee subspesies, R. d. di/ectus (2n = 46) en R. d. chakae (2n = 48). Die data verleen egter nie steun aan die divisie binne R. pumilio nie. Molekulêre datering van die twee spesies dui daarop dat die divergensie ten minste 5 miljoen jaar gelede plaasgevind het. Die periode was gekarakteriseer deur ekstensiewe klimaatsossilasies, wat gely het tot habitat fragmentasie in die spesie se verspreidingsgebied.

published_or_final_version / abstract / Statistics and Actuarial Science / Master / Master of Philosophy