Master of Science

A theoretical model and associated computer program are developed to compute hydrodynamic interaction forces and moments on passing ships in a shallow asymmetric canal, by applying generalized Lagally's theorem. Steady lift force on the ships are estimated following standard slenderbody approach. Passing ships are assumed to travel at constant speeds, parallel to the canal walls. Ships are assumed to be slender, the fluid is ideal and wave making effects are ignored. Numerical calculations for several passing cases show good agreement with existing model test results. / M.S.

The lift and drag forces exerting on a sphere immersed in a shear flow above a flat channel bed are evaluated by solving the steady three-dimensional Navier-Stokes equations. The numerical technique which combines the Newton iteration method and the finite element method is used to solve the non-linear Navier-Stokes equations. The technique first linearizes the non-linear terms in the partial differential equations, then solves the linearized equations by the finite element method. The Newton iteration method is used to linearize the non-linear equations. Since the iteration method requires a good initial guess, the linear solution of the partial differential equations is used for the initial guess, where the linear solution is the obtained by solving the differential equations without non-linear terms. The computer model developed can evaluate the lift coefficients of a sphere stationed at various distance from the channel bed. The computational results agree very well with the experimental measurements cited in the literature. The lift coefficient of the sphere changes with the undisturbed approaching velocity profile as well as the gap ratio which is the ratio of the distance between the sphere and the channel bed and the diameter of sphere. For fixed gap ratios, higher Reynolds number gives smaller lift coefficient than that of the lower Reynolds number. On the other hand, the lift coefficient also changes with the diameter of sphere for each fixed gap ratio. For small gap ratios, the lift coefficient increases as the diameter of sphere increases. For large gap ratios, the lift coefficient increases in the negative (downward) direction as the diameter of sphere increases. / Ph. D.

<p>The objectives of this thesis are to characterise the performance of variably angled V-shaped opening ball valves in terms of pressure drop and cavitation. Both numerical and experimental techniques are utilised here to study the effects of apex angle of the V-opening on pressure drop and cavitation characteristics. Three different ball sizes are used to investigate scalability of the V-ball valves in terms of cavitation inception and pressure coefficient. The results of the pressure drop experiments show that as the size of the ball and the apex angle of the valve increases, the pressure coefficient tends towards a constant value. This means that larger V-opening ball valves can be scaled using pressure coefficient within 15%.</p> <p>It has been well established that cavitation causes high frequency noise. Therefore dynamic pressure transducers were used to detect acoustic cavitation noise by considering the high energy content in the 2 kHz to 45 kHz frequency band. To measure the energy levels, spectral analysis was performed and the power spectrum density was acquired using fast Fourier transform algorithm and the area under the curve was integrated in three different frequency intervals of 2 to 5 kHz, 5 to 10 kHz and 10 to 45 kHz to capture the frequency band at which cavitation onset occurs. This energy was compared to the reference energy levels in the same frequency intervals when no valve was installed. Two criteria were chosen to represent a cavitating flow, the first criteria was the start of a steep exponential increase in the energy from the reference energy in at least one of the frequency ranges defined and the second criteria is an increase in the coherence function in any of the three frequency ranges. This procedure was performed for all valves tested at 10% opening increments starting at fully open position and the inception cavitation number was recorded to define the onset of cavitation. It was observed that as the apex angle decreases, cavitation number also decreases where the size of the valve did not affect cavitation number. The conclusion was that as the opening decreases, cavitation inception occurs at higher pressure drops. However some deviations from this general trend were observed. These deviations are perhaps due to the turbulent structures such as flow separation and vortices, suggested by the pressure fluctuations in the static pressure.</p> <p>Finally numerical modelling of 1 inch 60V valves were performed using ANSYS Fluent computational fluid dynamics (CFD) software for three openings using two different turbulence models, standard k-ε and SST k-ω. The results were contrasted against the experimental data to evaluate which model performs the best for this application. From the results obtained, standard k-ε predicts the pressure drop within 15% of the experimental data. Also flow separation is a major cause of high local pressure drops and therefore cavitation and SST k-ω predicts cavitation better than standard k-ε. This is due to better performance of SST k-ω at predicting turbulence characteristics of flow separation region.</p> / Master of Applied Science (MASc)

No description available.

A combination of climate change, due to anthropogenic CO2 emissions and concerns over energy security, due to fluctuating fuel prices, has led to the UK pushing to increase its renewable energy production. Harnessing tidal stream energy is an infant sector in the UK’s renewable energy portfolio, despite the UK’s large tidal stream resource, and is harnessed through deploying tidal stream energy converters. This thesis focuses on the design of a novel vertical axis tidal turbine (VATT) called CarBine which is driven by drag force and therefore has several unique advantages, including slow rotational speed, omni-directionality and a simple design. This research focused on the optimisation of the CarBine design and was split into two areas: physical testing and numerical modelling. A conventional Savonius was also analysed in both of these respects and this enabled a direct comparison of CarBine to a competing drag force driven VATT. Physical testing measured the power of the turbines and was conducted in hydraulic flumes at both Cardiff University (CU) and IFREMER in Northern France for a range of flow conditions. Testing at IFREMER enabled the quantification of the blockage effect in the CU experiments. Physical testing resulted in the optimised design of CarBine being one with a four arm configuration. From physical testing at CU, CarBine showed inferior efficiency performance to that of the conventional Savonius, a Cp of 0.117 compared to 0.225 at U∞=0.72 m/s. As a result a hybrid of both the Savonius and CarBine was tested, namely a Savonius with flaps. However, results from physical testing showed the Savonius with flaps to have inferior performance to both CarBine and the conventional Savonius, with a Cp of 0.103 at U∞=0.72 m/s at CU. Numerical modelling was conducted using the commercial CFD software package, Ansys CFX. Both transient and steady state simulations along with 2D and 3D models were used to model both CarBine and the Savonius. Both the k-ε and SST turbulence models were used for comparison. The two degrees of rotational freedom present in the CarBine design resulted in CarBine being difficult to model precisely. The numerical modelling results were validated against the physical testing results and where available, 3D results showed closer validation than the 2D results.

Based on the concepts of water exchange timescales, a three-dimensional model has been refined to predict the age of water (AW). The model has been applied to two estuaries with very different characteristics. One estuary is a partially stratified micro-tidal estuary, i.e., the Pearl River Estuary (PRE), China and the other one is a well-mixed macro-tidal estuary, i.e., the Severn Estuary and Bristol Channel, UK (SEBC). The focus of this study is to investigate the influence of density-induced circulation on the estuarine water exchange process. A comparison between the predictions made using the barotropic mode and the baroclinic mode has been undertaken in the above two estuaries. The results indicate that due to the partially stratified phenomenon, in the PRE lower AW values are observed at the water surface, with higher AW values occurring near the bed. In the wet season, a more obvious AW stratification can be observed. The density-induced circulation causes an increase in the water mixing rate and a decrease in the timescales by a ratio greater than 50%. The stratification of the AW distribution is also enhanced. However, in the homogenous phenomenon of the SEBC, the baroclinic forcing’s impact on the water exchange process is very small. A model investigation is then carried out to predict the spatio-temporal variation of AW distribution in response to other dynamic factors. A regression analysis reveals that the age of freshwater exponentially decreases with the growth of discharge. Scenario analyses show that the tidal influence becomes more pronounced when the discharge is relatively small, and the varying tidal range mainly impacts on the fluctuation range of the AW, instead of its depth-averaged value. Moreover, the iv vertical variation in the AW is enhanced during neap tides due to less mixing. The age of seawater in the estuary is lowest near the estuary mouth and it increases further upstream. In the SEBC, the impacts of proposed tidal stream turbines with two different array arrangements on the water exchange process are analysed to investigate changes in the water level, current speed and AW distributions. A physical model investigation into the water exchange processes in a small scale model of the SEBC is then undertaken to estimate the travel time between different locations, and the water and tracer fluxes through two selected cross sections.

In this study, a hydro-environmental numerical model is utilised to further demonstrate the applicability of computer models to predict tidal flow in coastal waters. In particular, high resolution model simulations are performed at two selected sites: the Ogeechee Estuary, USA to assess the hydro-kinetic energy potential near Rose Dhu Island, a small island in the estuary; and at Swansea Bay, UK to assess faecal coliform pollution levels in the bay. Model results from the Ogeechee Estuary simulations revealed that better representation of branching smaller creeks located inshore enhanced the magnitude of tidal currents by approximately 30% near Rose Dhu Island. Evaluation of spatial and temporal distribution of currents revealed that local hot-spots of hydro-kinetic energy exist within the estuary and a maximum annual power of 4.75MW is available from the tidal streams surrounding the island. Investigation of the sensitivity of model parameters related to intertidal storage and bottom friction showed that ebb tide dominance in the estuary is reduced by lowering wetland elevation and by increasing bottom friction in the channel. Increasing the marsh friction to represent the resistance offered by marsh vegetation decreased the influence of intertidal storage on tidal distortion as ebb-dominance is reduced. Model results from the Swansea Bay simulations showed that three distinct flow patterns exist in the bay including re-circulating eddy like patterns, due to the presence of a headland located towards to the south-west end of the bay. The model-predicted distribution of Faecal Indicator Organisms (FIO) helped identify major pollution sources that negatively influence the rating of the Swansea Bay bathing water site. Investigation of the spatial distribution of FIO concentrations at the Designated Sampling Point (DSP) revealed that that the samples collected at DSP for compliance monitoring would correctly represent the pollution levels in the surrounding areas, however, at locations further off-shore significant spatial variability by up to five times was observed. As expected, intermittent peaks in FIO concentrations were noticeable following rainfall events, however, a strong temporal variability within a day was also observed at the DSP with concentration values varying by up to ten times in magnitude.

Water distribution system utilities must be able to maintain a system’s assets (i.e., pumps, tanks, water mains, etc.) in good working condition in order to provide adequate water quantity and quality to its customers. Various asset management approaches are employed by utilities in order to make optimal decisions regarding the renewal of system components. Part of a good asset management approach is performing a comprehensive risk analysis which consists of considering all potential ways in which the system may fail, the likelihood failure of for each scenario, and the consequences of said failure. This study investigates a water distribution system’s risk of failure due to both acute transient events (e.g., pump trip) and standard pressure fluctuations due to daily system operations. Such an analysis may be useful in optimal decision making such as asset monitoring, scheduling of condition assessments or system renewal projects, policy implementation, and investment priorities in order to keep the utility’s total costs at a minimum. It may also be useful as a precautionary measure to help prevent catastrophic failures such as large main blowouts for which the utility would incur substantial costs, both direct and indirect. As part of this thesis, a database of water distribution system models is used to analyze the effects of an acute transient event for different system configurations. The database was created at the University of Kentucky and has been made available to the research community to test newly developed algorithms for various studies including optimal system operations and optimal system design.

Thesis (PhD)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: Local scour is the erosion of a riverbed resulting from the flow of the river around an obstacle. It is a principal cause of failure of bridges and other hydraulic structures. Current design practice relies on the use of empirical formulae that are often extremely inaccurate, or on the use of physical models that are very expensive. Recent advances in the power of microcomputers have however made numerical simulation increasingly attractive. One obstacle to numerical simulation though is that there is no general agreement on the concept of incipient motion, that critical point at which motion - and hence scour - begins. In this dissertation, the unit stream power model developed by Rooseboom (1992) is extended to handle the complex three-dimensional flow conditions that pertain close to the riverbed in the vicinity of an obstacle. The relationship between unit stream power (the dissipation function) and the Movability Number (the ratio of the shear velocity to the terminal settling velocity of the critical sediment particles) is clearly indicated. Since incipient motion is probabilistic in nature, a relationship was established between the Movability Number and the intensity of motion with allowance for bed-slope and relative depth. An extension of this work resulted in a new bed-load transportation equation that could be used to determine the rate of scour development. Physical modelling in a laboratory flume aided the selection of suitable critical conditions for the onset of scour. The usefulness of the above-mentioned relationships was then demonstrated through the construction of a simple mathematical model of scour and deposition around a structure. This model was used in conjunction with commercially available computational fluid dynamics (CFD) software to predict the scour potential around typical engineering structures. Physical model data was obtained for four situations, and the measured scour was compared with that predicted by the numerical model. There was reasonable agreement between the different models and such differences as there were could be readily attributed to constraints on the numerical model, in particular the lack of a free-surface routine and the coarseness of the grid. This dissertation has opened up a new method for the prediction of local scour that could be readily extended to include all types of scour. With the advent of increasingly fast computers, it could become a useful engineering tool that would assist engineers in the design of safe and cost-effective foundations for hydraulic structures. / AFRIKAANSE OPSOMMING: Plaaslike uitskuring is die erosie van 'n rivierbed as gevolg van vloei verby 'n obstruksie. Dit is 'n belangrike oorsaak van die swigting van brfïe en ander hidrouliese strukture. Bestaande ontwerppraktyk berus op empiriese vergelykings wat dikwels hoogs onakkuraat is, of op fisiese modelle, wat baie duur is. Numeriese simulasie het die afgelope tyd 'n al hoe meer aantreklike opsie geword danksy die snelle toename in die kapasiteit van mikro-rekenaars. 'n Struikelblok met numeriese simulasies is die gebrek aan konsensus oor die konsep van begin-van-beweging, daardie kritieke toestand waarby beweging en derhalwe uitskuring begin. In hierdie proefskrif is die eenheidstroomdrywing model, ontwikkel deur Rooseboom (1992), uitgebrei om die komplekse drie-dimensionele vloeitoestande, wat teenaan die rivierbodem verby 'n obstruksie heers,te hanteer. Die verwantskap tussen Eenheid Stroomdrywing (Dissipasiefunksie) en die Beweeglikheidsgetal (verhouding tussen sleursnelheid en die ewewigvalsnelheid van die kritieke sedimentpartikels ) is duidelik uitgewys. Aangesien begin van beweging probabilisties van aard is, is die verwantskap bepaal tussen die Beweeglikheidsgetal en die Intensiteit van Beweging, met voorsiening vir bodernhelling en relatiewe diepte. Verdere uitbreiding het gelei tot 'n nuwe bedvrag vervoervergelyking wat gebruik kan word om die tempo van uitskuring te bepaal. Kritieke toestande, waarby uitskuring begin, is met fisiese modelle in die laboratorium gekwantifiseer. Die bruikbaarheid van bogenoemde verbande is gedemonstreer deur die ontwikkeling van 'n eenvoudige wiskundige model van uitskuring en afsetting rondom 'n struktuur. Hierdie model is saam met bestaande kommersiële sagteware vir vloeidinamika berekenings (CFD) ingespan om uitskuringspotensiaal rondom tipiese ingenieurstrukture te voorspel. Fisiese modelmetings van uitskuring vanaf vier uitlegte is vergelyk met die numeries voorspelde waardes. Bevredigende ooreenkoms is gevind en verskille kon geredelik gewyt word aan beperkings van die numeriese model, veral die gebrek aan' n vryvlakroetine en die growwe maas. Die proefskrif stel 'n nuwe metode vir die voorspelling van uitskuring daar wat geredelik uitgebrei kan word na ander vorms van uitskuring. Met die ontwikkeling van al vinniger rekenaars kan dit 'n nuttige hulpmiddel vir ingenieurs word om veilige en koste-doeltreffende fondamente in waterlope te ontwerp.