This thesis investigates statistical techniques that can be used to improve estimates and methods in feasibility assessments of photovoltaic (PV) systems. The use of these techniques are illustrated for a case study of a 1MW PV system proposed for the Nelson Mandela Metropolitan University South Campus in Port Elizabeth, South Africa. The results from the study provide strong support for the use of multivariate profile analysis and interval estimate plots for the assessment of solar resource data. A unique view to manufacturing process control in the generation of energy from a PV system is identified. This link between PV energy generation and process control is lacking in the literature and exploited in this study. Variance component models are used to model power output and energy yield estimates of the proposed PV system. The variance components are simulated using Bayesian simulation techniques. Bayesian tolerance intervals are derived from the variance components and are used to determine what percentage of future power output and energy yield values fall within an interval with a certain probability. The results from the estimated tolerance intervals were informative and provided expected power outputs and energy yields for a given month and specific season. The methods improve on current techniques used to assess the energy output of a system.

Thesis (Ph. D.)--Florida State University, 2004. / Advisor: Dr. Anuj Srivastava, Florida State University, College of Arts and Sciences, Dept. of Statistics. Title and description from dissertation home page (viewed Jan. 12, 2005). Includes bibliographical references.

The majority of the world's offshore infrastructures are now showing the sign of aging and are approaching the end of their original design life. Their ability to withstand various operational and environmental changes have been the main concerns over the years. This is because the pipeline will still need to operate for a few more decades with increasing demand of oil and gas supply. To address the issues, an effective pipeline integrity management system is required to manage pipeline systems and to ensure the reliability and availability of the pipeline. The main goal is to identify, apply, and assess the applicability of the Bayesian network approach in evaluating the integrity of subsea pipelines that evolves with time. The study is aimed to specifically handle knowledge uncertainties and assist in the decision making of subsea pipeline integrity assessment. A static Bayesian network model was developed to compute the probability of pipeline condition and investigate the underlying factors that lead to pipeline damage. From the model, the most influential factors were identified and the sensitivity analysis demonstrated that the developed model was robust and accurate. The proposed model was then extended to develop a decision tool model using an Influence Diagram. The results from the proposed influence diagram were used to prioritize the maintenance scheme of the pipeline segments. Benefit to cost ratio was applied to determine the pipeline maintenance intervals. Dynamic Bayesian network was utilized to model timedependent deterioration of pipeline structural reliability. A good agreement with conventional structural reliability method is achieved. The present thesis has demonstrated the applicability and effectiveness of Bayesian network approach in the field of oil and gas. It is hoped that the proposed models can be applied by oil and gas pipeline practitioners to enhance the integrity and lifeltime of the oil and gas pipeline.

Thesis (M.Sc.)--Victoria University of Wellington, 2008. / Includes bibliographical references.

Thesis (Ph. D.)--Ohio State University, 2007. / Title from first page of PDF file. Includes bibliographical references (p. 108-118).

South Africa is at the southern fringe of sub-Saharan African countries which persist in experiencing malaria transmission. The purpose of the study is to analyse the geographical and temporal patterns of malaria transmission from 2000 to 2011 using Bayesian geostatistical modelling in Limpopo Province, South Africa. Hereafter, develop malaria case data-driven spatio-temporal models to assess malaria transmission in Limpopo Province. Malaria case data was acquired from the South African Medical Research Council (MRC). Population data was acquired from AfriPopo; and Normalised Difference Vegetation Index (NDVI), Land Surface Temperature (LST) and Land Cover data were acquired from MODerate-resolution Imaging Spectro-radiometer (MODIS). Rainfall, Altitude and distance to water bodies’ data were acquired from African Data Dissemination Service (ADDS), United States Geological Survey (USGS) and Environmental Systems Research Institute (ESRI), respectively. Bayesian spatio-temporal incidence models were formulated for Gibbs variable selection and models were fitted using the best set of environmental factors. Modelbased predictions were obtained over a regular grid of 1 x 1km. spatial resolution covering the entire province and expressed as rates of per 1 000 inhabitants for the year 2010. To assess the performance of the predicted malaria incidence risk maps, the predictions and field observations were compared. The best set of environmental factors selected by variable selection was Altitude and the night temperature of two months before the case was reported. The environmental factors were then used for model fitting and all of the covariates were important on malaria risk. Predictions were done using all the environmental factors. The predictions showed that Vhembe and Mopani district municipalities have high malaria transmission as compared to other district municipalities in Limpopo Province. Assessment of predictive performance showed scatter plots with the coefficient of determination ( R² ). The values representing the statistical correlation represented by the coefficient of determination ( R² ) were 0.9798 (January), 0.8736 (February), 0.8152 (March), 0.8861 (April), 0.9949 (May), 0.3838 (June), 0.7794 (July), 0.9235 (September), 0.8966 (October), 0.9834 (November) and 0.8958 (December). August had two values reported and predicted which resulted in R² of 1. The numbers of the The produced malaria incidence maps can possibly be considered as one of the baselines for future malaria control programmes. The results highlighted the risk factors of malaria in Limpopo Province which are the most important characteristics of malaria transmission. / M.Sc. University of KwaZulu-Natal, Durban, 2013.

M. Tech. Electrical Engineering. / Investigates a new approach of wheelchair control, based on the user behaviour recognition. This objective involves two steps in the resolution of the problem. The first step is to determine the action the user initiates. Therefore, the present study will mostly refer to literatures on car driver behaviour modelling, as several studies have been conducted in that domain. The proposed model of user's behaviour presented here is based on probabilistic graphical model, for instance, Bayesian network. The second step is the generation of an assistive control signal that will compensate the user input, depending on the driving task inferred by the Bayesian network.Experiments have been conducted on a virtual environment model developed in Matlab and several users participated to the experiments. The results show a great potential of Bayesian Network model to infer on human behaviour and also a satisfying output from the ANFIS model as it delivers signals following the user's behaviour.

Invasive alien plants have widespread ecological and socioeconomic impacts throughout many parts of the world, including Swaziland where the government declared them a national disaster. Control of these species requires knowledge on the invasion ecology of each species including how they interact with the invaded environment. Species distribution models are vital for providing solutions to such problems including the prediction of their niche and distribution. Various modelling approaches are used for species distribution modelling albeit with limitations resulting from statistical assumptions, implementation and interpretation of outputs. This study explores the usefulness of Bayesian networks (BNs) due their ability to model stochastic, nonlinear inter-causal relationships and uncertainty. Data-driven BNs were used to explore patterns and processes influencing the spatial distribution of 16 priority invasive alien plants in Swaziland. Various BN structure learning algorithms were applied within the Weka software to build models from a set of 170 variables incorporating climatic, anthropogenic, topo-edaphic and landscape factors. While all the BN models produced accurate predictions of alien plant invasion, the globally scored networks, particularly the hill climbing algorithms, performed relatively well. However, when considering the probabilistic outputs, the constraint-based Inferred Causation algorithm which attempts to generate a causal BN structure, performed relatively better. The learned BNs reveal that the main pathways of alien plants into new areas are ruderal areas such as road verges and riverbanks whilst humans and human activity are key driving factors and the main dispersal mechanism. However, the distribution of most of the species is constrained by climate particularly tolerance to very low temperatures and precipitation seasonality. Biotic interactions and/or associations among the species are also prevalent. The findings suggest that most of the species will proliferate by extending their range resulting in the whole country being at risk of further invasion. The ability of BNs to express uncertain, rather complex conditional and probabilistic dependencies and to combine multisource data makes them an attractive technique for species distribution modeling, especially as joint invasive species distribution models (JiSDM). Suggestions for further research are provided including the need for rigorous invasive species monitoring, data stewardship and testing more BN learning algorithms. / Environmental Sciences / D. Phil. (Environmental Science)