A thesis submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements of the degree of Doctor of Philosophy. Johannesburg, February 2017 / Rainwater harvesting is used as a way of improving crop yields in rain fed agriculture by capturing excess rainfall and storing it in-situ or in reservoirs for use during dry spells. Contour ridges are one of the many rainwater harvesting technologies that are used although little is known about their effectiveness. Contour ridges harvest runoff generated in the cropped field upstream of the ridges. The traditional contour ridge type in Zimbabwe was introduced by the government in the 1950s to control soil erosion through safely draining away runoff from cropped fields and is commonly referred to as graded contour (GC) ridges. In the 1990s the country experienced severe and more frequent droughts leading stakeholders to experiment on contour ridges that retain the runoff instead of draining it away which are known as dead level contour (DLC) ridges. There was therefore the need to find out if there are benefits derived from this change and assess conditions under which benefits would be experienced. Previous studies have shown that rainwater harvested by contour ridges can improve water availability in downstream fields. However these studies did not investigate the conditions under which such benefits are realised. In addition no attempt to model water harvesting by contour ridges have been made in Zimbabwe while the contour ridges are widely being used for soil and water conservation. This research investigated the effect of contour ridges by comparing soil moisture between plots with DLC and GC ridges using plots with no contours as a control. Experimental work was carried out in Zhulube, in Matebeleland South Province of Zimbabwe. Matebeleland South Province falls within the semi-arid area in which rainfall is characterised by mid-season dry spells leading to frequent crop failure. In addition, the area often receives high rainfall intensities leading to soil erosion and sedimentation of rivers. DLC and GC ridges were constructed in farmers’ fields where maize crops were planted. Soil moisture measurements were done using a micro gopher soil moisture profiler while runoff plots were used to measure runoff generation. A fuzzy model was developed using data from this experiment and a previous study in Masvingo Province of Zimbabwe to simulate runoff generation at field scale while a process based water balance model was also developed to simulate soil moisture changes within the root zone of the cropped area. The results from this study indicate that DLC are effective in clay and loamy soils where runoff generation is significant and not in sandy soils due to insignificant generation of runoff under the rainfall regimes of semi-arid areas. Fuzzy logic was found to be a useful method of incorporating uncertainty in modelling runoff at field scale. A mass water balance model developed on process based principles was able to model soil moisture in the root zone reasonably well (NSE =0.55 to 0.66 and PBIAS=-1.3% to 6.1%) and could help to predict the water dynamics in contour ridged areas as would be required in determining the suitable dimensions and spacing of contour ridges. Further research is required to improve the fuzzy component of the model for estimation of runoff when more data becomes available. In addition experiments to validate methods of estimating macro pore fluxes and lateral transfer of water from the contour ridge channel to the downslope field are also recommended. The model structure can be improved by adopting the representative elementary watershed approaches to include momentum and energy balances in addition to mass balance that was used in this study. / MT2017

This thesis considers the neural network learning control of a variable-geometry automotive active suspension system which combines most of the benefits of active suspension systems with low energy consumption. Firstly, neural networks are applied to the control of various simplified automotive active suspensions, in order to understand how a neural network controller can be integrated with a physical dynamic system model. In each case considered, the controlled system has a defined objective and the minimisation of a cost function. The neural network is set up in a learning structure, such that it systematically improves the system performance via repeated trials and modifications of parameters. The learning efficiency is demonstrated by the given system performance in agreement with prior results for both linear and non-linear systems. The above simulation results are generated by MATLAB and the Neural Network Toolbox. Secondly, a half-car model, having one axle and an actuator on each side, is developed via the computer language, AUTOSIM. Each actuator varies the ratio of the spring/damper unit length change to wheel displacement in order to control each wheel rate. The neural network controller is joined with the half-car model and learns to reduce the defined cost function containing a weighted sum of the squares of the body height change, body roll and actuator displacements. The performances of the neurocontrolled system are compared with those of passive and proportional-plusdifferential controlled systems under various conditions. These involve various levels of lateral force inputs and vehicle body weight changes. Finally, energy consumption of the variable-geometry system, with either the neurocontrol or proportional-plus-differential control, is analysed using an actuator model via the computer simulation package, SIMULINK. The simulation results are compared with those of other actively-controlled suspension systems taken from the literature.

This thesis examines the policy problem of economic unit definition from the perspective of the regional economist. The regional economist faces the challenge of disaggregating macroeconomic activity into subparts that accurately reflect the actual economic organization of a country or region. Such an exercise is important because the Governments of many developed countries rely on it to allocate scarce public resources. In the United States, the Bureau of Economic Analysis of the U.S. Department of Commerce is responsible for regional economic unit definition. To meet its mandate, the BEA has developed a complex assignment system based principally on commuting flows between regions of the Country. This assignment system works well for the centralized population centers that characterize the majority of the U.S. economy. However, the BEA system is less effective at reflecting the economic organization of rural areas, where there is little interregional commuting. To address this problem, the BEA has developed a practice of using newspaper circulation data as a proxy for economic organization. In this thesis I develop a partial set model of regional economic organization based on the mathematics of fuzzy logic and propose it as a superior alternative to the BEAs method.

With the continued development of economy, the energy demand increases continually . The industry is interested in evaluating the technology of heat pump, with applications mostly for water heater. This thesis analyzes and compares the heating efficiency of electric heater, gas heater and heat pump heater . A fuzzy theory is employed to the developed heating algorithm to raise the efficiency of the heat pump heater and suppress the unnecessary heating at the high temperature. According to the experiment, the heat pump heater could reach 60% energy saving while gas heater could attain 23% in comparison with the electric heater under the same temperature, The system could work all year around and can provide users with water at a proper temperature. In the winter, COP of heat pump heater could reache 2.1~2.5 while in the summer around 2.61~2.95. The energy saving is obvious.

Thesis (M.S.)--Worcester Polytechnic Institute. / Keywords: LabVIEW; fuzzy logic control; anesthesia. Includes bibliographical references (p. 71-74).

- / -

The problem of the evaluation of fuzzy situations arises quite often in real life. Images, symbols, signals and so on usually are fuzzily described and can be considered as fuzzy situations. So the parameters, which characterize the fuzzy situation, are not only numerical ones, but verbal too, for example, an object can be described as "heavy, little and moves fast". It is difficult to decide something formally about the object with such characterization. Leasing companies meet this problem in contracts classification process too. Accordingly most leasing companies have own contracts classification (recognition) systems, which mostly are statistically based and the similarities and differences between good and bad contracts are not computed. Therefore the demand arises to develop the intellectual recognition system using fuzzy logic. This system should be simply integrating in any leasing company. It also must compute and measure the similarities and differences between good and bad contracts from their evaluation history. Such intellectual system could be as an adviser for business expert. The module of fuzzily characterized situation's evaluation is implemented using: 1) fuzzy sets, 2) fuzzy clustering, 3) fuzzy recognition. The general idea was experimentally investigated using artificially generated data as well as using data from real contracts. The software developed during this research is under preparation to be integrated in companies "Baltic... [to full text]

Wireless sensor nodes are resource constrained and have limited amount of energy. Therefore, designing protocols that conserve energy is an important area of research. Researchers have investigated architectures and topologies that allow energy efficient operation of WSNs. One of the popular techniques in this regard is clustering. A typical clustering protocol contains two main steps: cluster head election and cluster formation. This thesis is aimed at investigation of the cluster formation process. We propose a Fuzzy Logic based approach that uses three descriptors namely: energy level, distance between cluster-head and base station, and distance between the cluster-head and the sensor’s node. We compare our proposed model, FLCFP (Fuzzy Logic Clustering Formation Protocol), with the most popular model, LEACH, which was proposed previously to prolong network lifetime. FLCFP approach is shown to prolong network lifetime. In addition, it is shown that sensor node energy is consumed in a more uniform fashion.

No description available.

No description available.