Master's completed
The following is a list of master's dissertations that were recently completed in this Department.
All these dissertations can be found at UPSpace, the University of Pretoria Institutional Repositotry.
The abstracts of older masters from the department are summarized in the following archive pages:
2018
NM van der Merwe, 2018, "Fully developed forced convection heat transfer and pressure drop in a smooth tube in the transitional flow regime"
Extensive work has been done on characterising convective heat transfer and pressure drop in smooth tubes in the laminar and turbulent flow regimes. However, little work was completed in the transitional flow regime. In all previous transitional studies, experiments that were conducted between the laminar and turbulent flow regimes were with mixed convection in the laminar flow regime and not in the forced convection flow regime. The secondary flow that occurs during mixed convection should most probably influence the characteristics in the transitional flow regime. It can therefore be expected that the transitional flow characteristics of forced convection and mixed convection will be different. However, the transitional characteristics of forced convection flow have not yet been determined. The purpose of this study was therefore to determine the heat transfer and pressure drop transitional characteristics specifically in the forced convection flow regime. Furthermore, to focus on determining these factors for a circular, horizontal smooth tube for fully developed flow. The characteristics were determined in an experimental setup through which flow occurred through a test section consisting of a horizontal and circular smooth tube. The testsection inside diameter was 4.04 mm, and the tube length was 8.4 m. Water was used as the test fluid and was circulated through the test section which was heated at a constant heat flux. A calming section with a square edge inlet was upstream of the test section. Temperatures at the tube inlet, outlet and outer surface of the test section were measured with a total of 58 thermocouples. Two pressure taps was also installed on the test section and was connected to a pressure transducer for pressure drop measurements. Experiments were conducted mainly on the last part of the test section where fully developed flow occurred. Experiments were conducted between Reynolds numbers of 1 000 to 10 000, Prandtl numbers of 3 to 8, and Rayleigh numbers of 330 and 11 000 (heat fluxes of 0.89 kW/m^{2 }to 3.26 kW/m^{2}). It was found that the heat transfer transitional range coincided with the friction factor transition range with a Reynolds number range of 2 484 to 2 849. Forced convection results in the laminar regime was achieved and compared well to literature. The results were mapped on published flow regime maps. This was inconclusive as the published flow regime maps have been specifically developed for fixed parameters that did not match the parameters of this study.
R.P. Gräbe, 2018, "Difference thresholds for a vehicle on a 4poster test rig"
To improve ride comfort a reduction in the acceleration experienced by occupants is required. Simulation software and test equipment are able to measure reductions in acceleration that are too small for humans to perceive. It is therefore important to know how large the reduction in vibration should be for occupants to perceive an improvement in comfort. This study determined difference thresholds (DTs) for ten automotive engineers seated in a vehicle on a 4poster test rig. Participants were exposed to all six axes of vibration. DTs were determined for two road profiles using vertical acceleration measured on the seat and seat rail. The two road profiles were obtained by scaling the magnitude of the vertical displacements of a test track used for ride comfort evaluations. The two roads had different magnitudes, but the same spectral shape, and were therefore used to investigate the validity of Weber's Law. The BS 6841 weighted r.m.s. magnitude of the vertical acceleration measured on the seat were 0.58 and 1.01 m/s for the two roads. An updowntransformedresponse (UDTR) test procedure was used with a threedownoneup rule to determine DTs. There was no statistically significant difference found in the medians of the relative difference thresholds (RDTs), calculated from the vertical seat acceleration, for the two roads. The median RDT for the two roads were 10.1 % and 8.6 % respectively. Results were consistent with Weber's law.
W.J.Reid, 2018, "Experimental Investigation of Circumferentially NonUniform Heat Flux on the Heat Transfer Coefficient in a Smooth Horizontal Tube with Buoyancy Driven Secondary Flow"
Most heat transfer tubes are designed for either fully uniform wall temperature or fully uniform wall heat flux boundary conditions under forced convection. Several applications, including but not limited to the solar collectors of renewable energy systems, do however operate with nonuniform boundary conditions. Limited research has been conducted on nonuniform wall heat flux heat transfer coefficients in circular tubes, especially for mixed convection conditions. Such works are normally numerical in nature and little experimental work is available. In this experimental investigation the effects of the circumferential heat flux distribution and heat flux intensity on the single phase (liquid) internal heat transfer coefficient were considered for a horizontal circular tube. Focus was placed on the laminar flow regime of water within a stainless steel tube with an inner diameter of 27.8 mm and a length to diameter ratio of 72. Different outer wall heat flux conditions, including fully uniform and partially uniform heat fluxes were studied for Reynolds numbers ranging from 650 to 2 600 and a Prandtl number range of 4 to 7. The heat flux conditions included 360˚ (uniform) heating, lower 180˚ heating, upper 180˚ heating, 180˚ left and right hemispherical heating, lower 90˚ heating, upper 90˚ heating and slanted 180˚ heating. Depending on the angle span of the heating, local heat fluxes of 6 631 W/m^{2}, 4 421 W/m^{2}, 3 316 W/m^{2}, 2 210 W/m^{2} and 1 658 W/m^{2} were applied. Results indicate that the local and average steady state Nusselt numbers are greatly influenced by the applied heat flux position and intensity. Highest average heat transfer coefficients were achieved for case where the applied heat flux was positioned on the lower half (in terms of gravity) of the tubes circumference, while the lowest heat transfer coefficients were achieved when the heating was applied to the upper half of the tube. Variations in the heat transfer coefficient were found to be due to the secondary buoyancy induced flow effect. The relative thermal performance of the different heating scenarios where characterised and described by means of newly developed heat transfer coefficient correlations for fully uniform heating, lower 180° heating, and upper 180° heating.
JL van Niekerk, 2018, "Degradation Estimation of High Energy Steam Piping Using Hybrid Recurrent Neural Networks"
This dissertation is a study on estimating degradation of high energy steam pipework using modern machine learning techniques. High energy piping systems are very complex to simulate due to the many variables that could influence the useful life of a component. In this research a hybrid recurrent neural network is created that consists of a combined recurrent neural network and a feed forward neural network. The machine learning model is trained on historical data that has been captured over a sixyear time period and is applied to a test dataset to see if any usable patterns exist within the training data.
In this research the following variables of the piping system components are used as input to the machine learning model: the operating temperature and pressure time sequence, the distance to the closest anchor point, the distances to neighbouring supports as well as their elevation survey readings and the last known creep damage of the component. The model is created in Python using the Tensorflow library. Two types of recurrent neural networks (RNN) are tested, gated recurrent unit (GRU) and long short term memory (LSTM). The standard gradient descent (GD) algorithm, as well as adaptive gradient descent (ADAGRAD) and adaptive movement estimation (ADAM) are tested. The model was able to predict the classification of a component with an accuracy of up to 91% on the training dataset and 56% on the test data set, which is considered to be high given the complexity of the problem.
The model is successful in recognising patterns within the data and offers an automated way to parse large data sets that consist of a temporal and static data mixture. This offers an approach to make an objective decision on similar complex data driven problems and its application is not constrained to this single problem. The methods applied in this research is expected to perform even better on problems where the frequency of data collection is higher than what is used in this research.
PJ van Niekerk, 2018, "Development of a wind turbine condition monitoring facility for drivetrain torsional dynamics investigations"
Maintenance can be performed according to one of two strategies, failure based or condition based. In most cases, where large and expensive assets such as wind turbines are operated on a continuous basis, condition based maintenance is preferred. However, condition based maintenance relies on the continuous and accurate gathering of conditioninformation of the particular machine and its various components.
This dissertation reports the experimental and numerical work performed as part of the development of an experimental facility that will allow the development of condition monitoring techniques for wind turbines. This work is focused on the torsional dynamics of a wind turbine setup. A physical setup, consisting of a 1.6 m diameter turbine, a 1:1.2 ̇ speedratio gearbox, and a 24 Volt direct current generator is built.
All of it is mounted within an openreturn wind tunnel, which is also designed and built as part of this work. The following two costeffective experimental techniques are used to measure the torsional natural frequencies: a shaft encoder tachometer from which instantaneous rotational frequency is obtained, and power signal analysis, where the generated voltage is recorded and analysed. It is shown how an algorithm developed by Diamond et al. (2016) is used for the shaft encoder geometry compensation. Frequency spectra based on Fourier transforms and short time Fourier transforms are used to identify harmonic frequencies. Both measurement techniques proves useful to identify not only natural frequencies of torsional vibration, but also various characteristic frequencies of the drivetrain such as shaft rotation, blade pass, gear mesh and generator armature. It is found that power signal analysis is more useful to identify the characteristic frequencies.
Torsional dynamics of the drivetrain and its components are also investigated with the following two numerical methods: an eightdegreeoffreedom torsional Lumped Mass Model (LMM), and a threedimensional Finite Element Model (FEM). Torsional mode shapes and frequencies are calculated with both methods and a good agreement is found in the lower four modes. Numerical results are then compared with the experimental results, where there is also good agreement in the lower four modes. Model updating is performed on the FEM and by changing the torsional stiffness of the flexible couplings, the difference between measured and calculated natural frequencies are reduced to less than 6 %. It is concluded that future models should address lateral vibration of the drivetrain and the support structure.
From this study the following is contributed to the wind turbine condition monitoring field: considerations for the design and a working example of an experimental facility for investigating torsional dynamics, illustration of two measurement techniques, and two types of validated numerical models.
JC van der Walt, 2018, A comparison between machine learning techniques to find leaks in pipe networks
In 2012, the National NonRevenue Water assessment revealed that South Africa has 37% of nonrevenue water. With the steadily growing demand for this scarce resource, the detection of leaks in pipe networks is becoming more important. Currently, in South Africa the primary method of detecting leaks is to install pressure management systems and monitoring minimum night time flows.
The pressureflow deviation method, can be used to formulate an inverse analysis model based leak detection problem. This problem can then be solved using Artificial Neural Networks, Support Vector Machines and other optimization methods.
With EPANET, different networks were tested to compare these methods to finding leaks, using an inverse analysis formulated problem. Four different numerical networks were modeled and tested, a simple single pipe network, a small agricultural site, a distribution network and the simulated model of the experimental network that was designed and commissioned during the study in our laboratory.
From the numerical investigation, it was found that the optimization methods struggled to find solutions for simple networks with infinite number of solutions for the problem. For more complex numerical networks, it was seen that the Support Vector machine and the Artificial Neural Networks trained to the averages of their respective data sets.
Errors to ensure an accurate solution found by these algorithms were calculated as 2.6% for the numerical experimental network. The experimental network consisted of six possible leaking pipes, each having a length of 3m and a diameter of 10mm. Three leak cases were tested with diameters of 3mm and 2mm. Overall, the Support Vector machine could locate the leaking pipe with the best accuracy, while the minimizing of nonregularized error could calculate the size and location of the leak the most accurately.
Multiple leak cases were measured with the experimental network. The Support Vector machine was tested on these measurements, where it was found that two of the three leak cases could be solved with relative accuracies. Sensor usage optimization was completed on the measurements for the experimental network, where it was found that the leaks could be classified correctly with probabilities higher than 98% if only two sensors were used in the training of the SVM instead of all twelve.
Overall this method of leak detection shows promise for certain applications in the future. With practical applications on water distribution, transportation, and agricultural networks.
AJJ Hayes, 2018, Characterisation of the core and winding vibrations of power transformers with regulator windings
This dissertation presents research and experimental work done to characterize the core and winding vibrations of power transformers with regulator windings by measuring the tank vibrations.
The experimental tests were performed in the manufacturing plant whilst the transformers under investigation were subjected to the standard factory acceptance tests, specifically the no load loss test (open circuit test) and the full load loss test (short circuit test). The vibration measuring sensors that were used included a laser Doppler vibrometer and a triaxial accelerometer and the vibrations were recorded with a CoCo80 data logger.
The test results show that the characteristics of the core and winding vibrations of transformers with and without regulator windings are very similar, but in the case of transformers with regulator windings, the winding vibrations have a few more dependencies.
Thus this research and experimental work provide key insights into how the core and winding vibrations of power transformers with regulator windings are influenced by the regulator windings, how the tank vibrations of transformers with regulator windings should be measured and the difference between the vibrations of transformers with and without regulator windings. The importance of this is that most of the research that has been done on transformer vibrations, have been done on transformers without regulator windings, but most practical transformers do have regulator windings. Thus there is a shortage of practical transformer vibration information, which this study aims to address.
JF Grobler, 2018, "Multistate hydropneumatic suspension system through the use of magnetorheological (MR) valves"
This study is focused on modifying an existing solenoid valve based semiactive hydropneumatic springdamper system using MagnetoRheological (MR) fluid. The MR fluid’s effective viscosity can be altered by application of a magnetic field. Therefore, using a magnetic/MR valve makes it possible to change the state of the system by simply changing the applied magnetic field.
A prototype MR valve was developed to determine whether a unit small enough for installation was possible. This prototype valve was designed from first principles and properties such as pressure drop over the valve (damping) and flow blocking (for switching between spring characteristics) were measured. The measured pressure drop over the valve was higher than what was design for which was due to an incorrect assumption for the viscosity of the thixotropic MR Fluid. The flow blocking ability of the valve was determined by constant force tests. Results showed that the valve could virtually block the flow of fluid for approximately a quarter of the vehicles weight.
With the second prototype, the valve design and magnetic circuit design were improved. Two valves were constructed and implemented on a prototype suspension system. The damping characteristics of the system were lower than expected, however they can be improved by changing the valve geometry. The base spring characteristics are acceptable, however the higher spring characteristics fail when a high force is exerted on the strut that exceeds the valves flow blocking capability. The response time of the valve is not yet sufficient to make the system viable for real world implementation, especially under extreme conditions that can change more rapidly than the current valves.
L.M. Cramer, 2018, "Enhancement of the thermal performance of solar heat exchangers with porous inserts"
For high thermal performance and effectiveness, the flat plate heat exchangers and cooling channels are designed based on the three basic criteria: (i) small heat transfer area or large surface area to volume ratio, (ii) high heat transfer rate, and (iii) small pumping power. Numerous amounts of research have been dedicated to the notion of decreasing the size of a heat exchanger by enhancing the convective heat transfer inside the channels of a heat exchanger. Recently, the internal porous fins and porous foams of high thermal conductivity have gained considerable attentions in the research and development for their light weight, reduced fluid pumping power requirements, and high heat transfer characteristics. The results from the investigations show the enhancement of heat transfer coefficients and friction factors with the wavy screens relative to those in a smooth channel. This experimental research project aims to investigate the effects of the geometrical properties such the amplitude, period, and porosity of wavy porous mesh screen insert may have on the thermal performance of a heat exchanger and quantify the thermal performance of the channel employing the wavy porous screens for a wide range of applications at low to high Reynolds numbers. The friction factors, and heat transfer were measured in a rectangular channel when sinusoidal screen inserts were employed as turbulence promoters. The screen was made from porous mesh of flat metal screen available commercially. Two mesh screens were employed; one with a 68% porosity and one with a 48% porosity. Both mesh screens had a square shape pore and is delivered as a spool of material. The period of the screen was bent into the wavy mesh screen using a jig with two jaws. The screen wave vector was placed normal to the mean flow of the channel and allowed the peaks of the wave to make only line contact with the two larger side walls of the rectangular channel. The inlet Reynolds number for the experiments covered all three flow regimes: laminar, transition and turbulent. The measurements include the static pressure drop and wall temperature distributions along the channel. For the heat transfer experiments, the parallel walls of the channel touching the screen peaks are heated with a constant heat flux to simulate the channels in a flat plate heat exchanger. Heat transfer experiments were also obtained with one heated wall with a constant heat flux to simulate the conditions of a single channel heat exchanger employed in solar heaters and electronic cooling. Baseline data in a smooth channel without the screen inserts were also measured for comparisons with the data obtained in the same channel with the screen insert. The results on friction factors and heat transfer coefficients were then presented as ratios of data from the screen channel to the smooth channel to provide the performance of the screen channel relative to the smooth channel. The data and ratios were also presented in such a manner that the effect of change in porosity, period and amplitude of the screen insert could be studied. The sinusoidal screen inserts in the channels of a flat plate heat exchanger can provide desirable effects on the heat transfer enhancements (Nu/Nu0 > 1.0) only for the range of Reynolds number tested. The wire diameter of the mesh screen can significantly influence the thermal performance and pressure penalty provided by the wavy screen based on the present investigations and Mahmood et al. The present results are thus beneficial to the design of porous inserts for the heat exchangers operating over a wide range of flow rates. The effects of screen porosity and wave period are strong only on the efficiency index. The present results thus indicate the viability of the wavy porous inserts for the heat exchanger.
RG du Toit, 2018, "A Stochastic Hybrid Blade Tip Timing Approach for the Identification and Classification of Turbomachine Blade Damage"
Industry is increasingly confronted by ageing turbomachines nearing their decommissioning dates. These turbomachines are especially prone to unexpected and catastrophic failure, which is often the consequence of rotating blade failures. Thus the failure of a single blade may result in immense safety and financial impacts. The aforementioned points raise further questions with regards to the optimal outage planning of turbomachines. The distinct need to monitor the conditions of turbomachine blades during operation was therefore identified. It was further identified that monitoring of the blade conditions should provide sufficient evidence as to when a blade damage threshold has been reached, therefore providing early warning of imminent blade failure.
Blade Tip Timing (BTT) has existed for many decades as an attractive vibration based condition monitoring technique for turbomachine blades. The technique is nonintrusive and online monitoring is possible. For these reasons, BTT may be regarded as a feasible technique to monitor the conditions of turbine blades. The processing of BTT data to find the associated vibration characteristics is, however, a nontrivial task. In addition, these vibration characteristics are difficult to validate, therefore resulting in questionable reliability of the various BTT techniques. The use of a hybrid blade condition monitoring approach is therefore proposed in this research project. This hybrid approach incorporates a stochastic Finite Element Model (FEM) modal analysis to supplement the BTT results, therefore creating a basis of comparison as the BTT results become available. The aim of this research is to test the ability of the proposed hybrid approach to perform the following processes: blade damage identification and damage classification.
The use of a new BTT technique based on Bayesian Linear Regression (BLR) was tested on an experimental setup, where the first bending mode of the blades in the rotor assembly were excited. BTT based on BLR assumes a SingleDegree of Freedom (SDOF) model to describe the blade tip displacements. The advantage of the BLR curvefitting is that it solves for the parameters in this SDOF model as multivariate probabilistic quantities. More so, this technique solves these parameters for each revolution during the measured blade resonance conditions. The use of the multivariate probabilistic quantities in a MonteCarlo Simulation (MCS) enables the amplitude and phase values of the blades to be derived (also as statistical quantities). The natural frequencies of the blades can then be determined by extracting features from the corresponding amplitude and phase results.
Incremental discrete damage was introduced to a particular blade to test the ability of the proposed technique to track the changes in the derived natural frequencies. Discrete damage was also introduced in the Finite Element Analysis (FEA). Slight variations in the material properties, operational conditions (centrifugal loads) and the geometry of the discrete damage were introduced in the FEA for each damage increment. This ensured that this analysis was stochastic rather than deterministic, thus enabling uncertainty to be modelled. The proposed damage identification requires that the change in natural frequencies of the BTT and FEA results as tracked as relative quantities. The changes in blade natural frequencies were inherently due to the increase in discrete blade damage. The effects of varying blade temperatures and how this would affect the performance of the proposed hybrid approach was also tested. Experimentally, this required the heating of the blades to the desired temperatures, before and during the BTT tests. The FEA incorporated temperature effects by modelling uncertainty in the material properties. The proposed BTT technique was able to detect the decrease in the natural frequencies of the blades due to the increase in temperature. More importantly, the hybrid approach demonstrated that it is general enough to still be applicable with regards to the relative natural frequency tracking (blade damage identification) with varying temperature effects. The relative changes in the natural frequencies from the undamaged, or reference states, of the relevant blades were computed to infer the degree discrete damage as part of a probabilistic damage identification procedure. A probabilistic damage threshold for the damage identification procedure was proposed based on the following question: What is the probability that the relative change in the natural frequency of the test blade is as large as what the FEM modal analysis {at a chosen discrete damage size) projected it to be? This probabilistic damage identification procedure was demonstrated for various scenarios, therefore demonstrating the ability of the hybrid approach to infer the degree of blade damage for various scenarios.
The blade damage classification process relies on the use of Kmeans clustering. The clustering implementation offers the advantage of a single BTT measurement being sufficient as an indication of blade specific conditions. The FEM natural frequency results were used to initialize cluster centroids and the individual BTT points were assigned to clusters with the closest centroid (based on the amplitude and natural frequencies). The classification of a point to a certain cluster thus provides an indication of the severity of the blade damage. This was done for the BTT tests with and without the effects of varying the blade temperatures. The accuracy of the damage classification implementation seems promising. The decision of whether a damage threshold has been reached for this implementation is purely based on the damage classification of the individual points.
The BTT methodology incorporating BLR proved to be reliable when used as part of a hybrid approach. Furthermore, the advantages of the stochastic nature of the hybrid approach are highlighted in terms of quantifying uncertainty. The proposed hybrid methodology demonstrates the ability to identify and classify blade damage. In doing so, it was possible to determine that a blade damage threshold had been reached. It was therefore shown that the proposed stochastic hybrid approach may offer many short and longterm benefits for practical implementation. The proposed method therefore offers a feasible turbomachine blade monitoring solution that provides early warning of imminent blade failure.
JJF Wiid, 2018, The experimental and numerical investigation of the effect of shaft rotation on leakage rate of noncontacting seals found in turbine applications
This project was initiated by ESKOM power generation. ESKOM loses up to 22% of their steam energy in the HP turbines due to leakage at the turbine labyrinth seals. Therefore the need was expressed to implement improved sealing configurations. The aim of this study is to investigate the effect that shaft rotation has on the leakage rate of labyrinth and brush seals. This is done by means of experimental and numerical methods.
For many decades it was assumed that the shaft rotation has no or little effect on seal performance and therefore it was neglected in seal design. It was decided to investigate this subject, in order to assist and improve in future seal design and operation.
Both labyrinth and brush seals were investigated experimentally on a test rig. A real life application of the labyrinth or brush seals can be found in the power generation industry where a turbine shaft has a diameter of 300 mm and rotates at 3 000 rpm. The test rig was designed to assist in this application. Therefor the test rig had a shaft diameter of 150 mm with shaft speeds ranging between 010 000 rpm and with five different upstream pressures ranging from 15 bar. The same seals were then simulated using the commercial Computational Fluid Dynamics (CFD) package STARCCM+ with the bristle pack of the brush seal modelled as a porous medium. The coefficients of resistance for the porous medium were experimentally obtained. The two investigation methods are compared for the labyrinth and brush seals. The labyrinth and brush seals are also compared against each other.
The results show that the experimental leakage rates have a good correlation with those predicted by CFD. The CFD simulation provided detailed leakage flow fields and pressure distributions of both seals. It was found that shaft rotation has an influence on the leakage rate of both seals. The leakage rate increased at higher shaft speeds, with the brush seal performing better than the labyrinth seal. An increase of up to 1.7% was found at 10 000 rpm for the labyrinth seal and 1.45% for the brush seal at 10 000 rpm.
Guidelines were created based on these results to assist with advanced seal design. It is recommended that these guidelines are used in future seal design and other research aspects of noncontacting seals in turbo machinery.
HJ Breedt, 2018 , Atmospheric Boundary Layer Stability and its Application to Computational Fluid Dynamics
In the wind resource and wind turbine suitability industry Computational Fluid Dynamics has gained widespread use to model the airflow at proposed wind farm locations. These models typically focus on the neutrally stratified surface layer and ignore physical process such as buoyancy and the Coriolis force. These physical processes are integral to the accurate description of the atmospheric boundary layer and reductions in uncertainties of turbine suitability and power production calculations can be achieved if these processes are included. The present work focuses on atmospheric flows in which atmospheric stability and the Coriolis force are included.
The study uses MoninObukhov Similarity Theory to analyse time series data output from a proposed wind farm location to determine the prevalence and impact of stability at the location. The output provides the necessary site data required for the CFD model as well as stabilitydependent wind profiles from measurements. The results show nonneutral stratification to be the dominant condition onsite with impactful windfield changes between stability conditions.
The wind flows considered in this work are classified as high Reynolds number flows and are based on numerical solutions of the ReynoldsAveraged NavierStokes equations. A twoequation closure method for turbulence based on the kε turbulence model is utilized. Modifications are introduced to standard CFD model equations to account for the impact of atmospheric stability and ground roughness effects. The modifications are introduced by User Defined Functions that describe the profiles, source terms and wall functions required for the ABL CFD model. Two MOST models and two wallfunction methods are investigated.
The modifications are successfully validated using the horizontal homogeneity test in which the modifications are proved to be in equilibrium by the model's ability to maintain inlet profiles of velocity and turbulence in an empty domain. The ABL model is applied to the complex terrain of the proposed wind farm location used in the data analysis study. The inputs required for the stability modifications are generated using the available measured data. Mesoscale data are used to describe the inlet boundary conditions. The model is successfully validated by cross prediction of the stabilitydependent wind velocity profiles between the two onsite masts.
The advantage of the developed model is the applicability into standard wind industry loading and power production calculations using outputs from typical onsite measurement campaigns. The model is tuningfree and the sitespecific modifications are input directly into the developed User Defined Functions. In summary, the results show that the implemented modifications and developed methods are applicable and reproduce the main wind flow characteristics in neutral and nonneutral flows over complex wind farm terrains. In additions, the developed method reduce modelling uncertainties compared against models and measurements that neglect nonneutral stratification.
V.H.Wehrmeyer, 2018, Untripped Manoeuvre Induced Rollover Prevention for Sport Utility Vehicles
Rollover accidents account for a high number of serious injuries and fatalities and thus it is greatly important to reduce the number of occurrences. Although a large number of rollovers result from factors external to the vehicle design such as environmental obstacles there is a significant portion of rollover accidents which are preventable. Onroad untripped rollovers are directly related to the vehicle design. It is possible to do work in this area to improve vehiclerelated safety factors.
During rollover, lateral acceleration acts on the centre of gravity of the vehicle over turning it about the outer wheels. Thus the method to reduce or prevent rollover of this study stemmed from decreasing the overturning moment by reducing the movement arm though which the lateral acceleration acts. This is achieved by lower the ride height of the test vehicle using slow active suspension control of the test vehicle (Land Rover Defender 110) fitted with a hydropneumatic suspension system.
An experimental validated mathematical model representing the test vehicle is created to develop a rollover prevention control system that reduces the vehicle’s ride height and reduces the propensity to rollover. The control system applies one of three discrete suspension settings depending on the severity of the manoeuvre as well as lowering the ride height. The model is used to simulate the Fishhook 1B and the ISO 3888 Double Lane Change manoeuvres to evaluate the roll prevention system.
The rollover prevention control system improved the two wheel lift off speed of the vehicle through a Fishhook 1 B manoeuvre by 64% and the body roll angle of the vehicle through the Double Lane Change manoeuvre by 13% and the body roll rate by 25.7%. The rollover prevention control system significantly improved the vehicle’s response with regard to smooth flat onroad untripped rollover. Further improvements could be possible with the use of the proposed rollover prevention control system in conjunction with a fully active suspension system allowing for faster corrective action.
T.S. Mokobodi, 2018,"Designing and developing a free fall absolute gravity measuring system, using pneumatic actuators"
A gravimeter is an instrument that measures gravitational force F_{g} (N) or acceleration g (m/s^{2}). Absolute gravity measurements are preferred in metrology, due to the shortest traceability links to the SI base units of length and time, realising acceleration. The investigation on the suitable method of gravity measurement was performed at the National Metrology Institute of South Africa (NMISA). The free fall gravity measuring system was adopted for development.
The metrological need on redefining the kilogramme standard using the watt balance, supported the decision to mandate this project. Free fall gravimeters were researched. The new concept of fully pneumatic controlled vacuum chamber was invented and manufactured. Pneumatic actuators were used on the vacuum chamber to align, launch, capture and reposition the falling test mass. Laser interferometer and highspeed digitiser with embedded accurate clock module, were used in realising displacement and time, through numerical computations. Using stabilised HeNe Laser red with wavelength =633 nm interferometer, freefalling test mass displacements were traceable to length standard.
Interference intensity signal produced from experimental free fall drops were converted to A digital voltage signal enabling processing. Post signal processing algorithms were applied to the signal to extract the displacement and time coordinates of the freefalling object, using a zerocrossing detection method in a LabView environment. The final prototype setup measured the value 9.786041 m/s^{2} with uncertainty of 0.0000705 m/s^{2 }at the vacuum pressure of 0.05 Pa. It was validated and compared with the Council of Geophysics’ measured value of the site of 9.7860985 m/s^{2}.
G.S.Heymans, 2018, "Design and development of a responsive magnetorheological (MR) equipped semiactive suspension system for offroad vehicles"
The aim of this study is to design, implement and investigate the use of a MagnetoRheological (MR) equipped HydroPneumatic suspension system to solve the ride versus handling compromise of offroad vehicles. This suspension technology makes use of MR fluid viscosity changes which are induced by a varying magnetic field which serves as the basis for changing the suspension system’s damping as well as the stiffness characteristics. The primary focus of the study aimed to improve the response time characteristics of an existing prototype suspension system developed at the University of Pretoria. This improvement would be achieved through comprehensive magnetic optimisation of the MR Valve which characterises and controls the system. Additionally, work was done to understand and model the complex physical conditions which create the output characteristics of the semiactive suspension system and ultimately to understand how they will influence a vehicle during dynamic situations. This modelling platform was developed to account for the interrelated nonlinear elements within the system as well as capture the MR valve characteristics observed experimentally. Further quartercar simulationbased experiments where used to determine the feasibility and future contribution of this technology platform to solve ride verses handling compromise of offroad vehicles.
R.E. Goliada, 2018, "An Investigation on Inefficiencies in Spare Part Management Processes in South African Power Plants"
Inefficiencies in spare parts management can, and do, have great repercussions for the execution of maintenance in Eskom power stations. Despite this, the subprocesses of spare parts management in power stations had not yet been subjected to rigorous analysis to identify inefficiencies. Consequently, the purpose of this research was to establish the inefficiencies that exist in the management of spare parts in South African power stations, and also to determine the causes of the inefficiencies. All this was done with the intent of recommending a solution to improve spare parts management. It was for this reason that this research was conducted in 13 Eskom coalfired power stations in South Africa. The first phase of the research was concerned with uncovering and documenting the spare parts management model presently used in Eskom. The second phase included the use of a modified process failure mode and effect analysis (PFMEA) and the Delphi method to identify inefficiencies, their sources, and their significance. The research found that the inefficiencies included unsuitable maintenance and inventory management strategies, as a result of inadequate analysis of plant history. Furthermore, the study established that inadequate analysis of history was a result of poor maintenance records, incompetence, and the lack of access to computerised maintenance management systems (CMMSs). With the inefficiencies and their causes identified, the third phase of the research was then devoted to developing a methodology for improving spare parts management. This led to the development of a framework for improving spare parts management practices in power stations. The framework was validated and verified through a Delphi process. This framework was then recommended for adoption in an Eskom standard procedure for improving spare parts management practices. The research was thus successful in recommending a solution to improve the operational effectiveness and efficiency of spare parts management in South African power stations.
Supervisor: Professor J.L. Coetzee
2017
K. Arnachellan, 2017, "Aerodynamic loss reduction in a vane cascade with leadingedge fillet and upstream endwall filmcooling."
Secondary flow structures account for nearly 50% of aerodynamic losses experienced in the turbine blade passages. The adverse effects of these vortex structures transport the hot mainstream fluid towards the endwall blade surfaces, which enhances thermal stresses and leads to blade failure. The effects of leadingedge fillets and filmcooling with flush slots located upstream near the leadingedge region were investigated experimentally in the study in a largescale linear vane cascade in which the aerodynamic flow field was considered. The introduction of slot film flow and fillet aimed to reduce the effects of the secondary flow structures from the leading edge through the passage towards the exit in an effort to decrease the pressure losses, improve filmcooling coverage and flow field uniformity for the next blade row. The twodimensional vane profile was obtained from the hubside airfoil of the GEE3 engine nozzle guide vane. The slots were configured for two experimental cases to evaluate the influence of coolant flow rate and momentum; first, the effects of slot film injection from all four slots were observed and then compared with the second case injecting coolant only through the two central slots. Further effects were investigated by combining slot filmcooling with the leadingedge fillets employed on the endwall blade junction. The flow field measurements were quantified with spatial distributions of axial vorticity, total pressure loss, endwall static pressure and flow angle deviations taken across the cascade passage. The measurements were obtained at a Reynolds number of 2.0E+05 based on the cascade inlet velocity and vane chord length. Filmcooling inlet blowing ratios between 1.1 and 2.3 were investigated with the supply of coolant provided by a secondary channel. Filmcooling results were compared with the baseline case without slot film flow and fillet. The results indicated substantial improvement in the passage and exit planes with high inlet blowing ratios. The introduction of high momentum coolant flow from the central slots was seen to create laterally reversed axial vorticity, thereby counteracting the crossflow tendency in the passage. The effects at the passage exit showed suppressed vortex structures with slot film injection from the two central slots only, with further improvements in the flow angle deviations. The leadingedge slots were seen to contribute positive axial vorticity, which enhanced the passage vortex that was pushed away from the endwall at the exit. When the fillet was introduced, it had favourable effects in reducing the pitchwise pressure gradients along the endwall. Filleted filmcooling then resulted in a faint passage vortex system (5080% size and 2050% strength reduction) with a restored endwall boundary layer at high film flow rates. The leadingedge fillet was highly effective at the inlet of the blade passage because it weakened the horseshoe vortex formation. Thus, upstream slot filmcooling has great potential to decrease the aerodynamic losses and is further compounded with the leadingedge fillet.
L.M.J. Pallent, 2017, "The influence of a multiple tube inlet condition on heat transfer in the transitional flow regime."
In the industrial design of heat exchangers, engineers have long followed the general rule of avoiding transitional flow, and have rather designed a system operating in the turbulent flow regime. Whilst the turbulent regime is better for heat transfer, the higher friction inside the tube results in a much higher pressure drop which inevitably results in the system requiring a more powerful pump than if the system were to operate in the laminar regime. Designing a heat exchanger that operates in the turbulent flow regime is often the safer option, since little published design data is available for flow in the transitional flow regime, giving rise to numerous unwanted uncertainties during the design phase. Recent research into the transitional flow regime has resulted in promising experimental data that shows the regime is not as unstable as previously suspected. The regime allows for higher heat transfer than flows in the laminar regime, yet lower pressure drops than flows in the turbulent regime. Numerous investigations have previously been performed on a single uniformly heated tube operating in the transitional flow regime, however, there exists no data on the influence of a multiple tube inlet condition, as typically found in shell and tube heat exchangers, on the heat transfer characteristics. The purpose of this study was thus to determine the influence of varying tube pitch ratios on the fully developed heat transfer characteristics of three smooth circular horizontal tubes. An experimental set up was designed and built to accommodate a single tube heat exchanger used for validation purposes, and a multiple tube heat exchanger comprising of three identical and equally spaced tubes. Using a DC power supply, the tubes were uniformly heated at 2, 3 and 4 kW/m2 along the length of the test section. The heat transfer characteristics were determined experimentally for outer diameter tube pitch ratios of 1.25 and 1.5 of three 4 mm inner diameter tubes, each 6 m in length for a range of Reynolds numbers of 1 000 to 7 000. Water was used as the test fluid. Using PT100 probes and thermocouples at the inlet, outlet and outer surface of the test section, it was found that the presence of multiple tubes at the inlet of the heat exchanger for a pitch ratio of 1.25 promoted the onset of transition for the centre tube, and sharpened the transition gradient of the outer tubes. This effect noticeably increased with increasing heat flux and was absent at the higher pitch ratio of 1.5.
K.R.S. WRight, 2017, "The Effects of Age and Wear on the Stiffness Properties of an SUV tyre"
With an increasing need for accurate full vehicle models, a sensitivity analysis of the modelling of tyres depending on their age and wear was conducted. This included a sensitivity analysis into the accuracy of acquiring the tyre stiffnesses on a static test setup.
An FTire model is developed with the aim to update this model with basic tests to give a more accurate representation of the aged or worn tyre. A wellresearched and documented method is used to artificially age the tyres. During the aging process the tyre was statically tested to monitor the potential changes in characteristics. Tyres were also worn on a dynamic test setup and periodically tested to monitor the property changes. These tests included both static and dynamic measurements.
The results indicate that the vertical and longitudinal stiffnesses of the tyre have convincing dependencies on the age and wear of the tyre. While the aging process was a trustworthy method, the wear process created irregular wear across and around the tyre subsequently skewing the results. Simple methods of updating the FTire tyre model without reparameterising the model completely, was found to be effective in accounting for age and wear.
M. Joubert, 2017, "The influence of a multiple tube inlet condition on fullydeveloped friction factors in the transitional flow regime"
Concentrated solar power systems, such as parabolic trough heat exchangers, are being used more frequently in the energy sector. Engineers require accurate design information to optimise the effectiveness of these heat exchangers. Recent studies have shown that the transitional flow regime could potentially be the optimum region of operation, as it provides the best possible compromise between high heat transfer and low pressure drop. However, designers often choose to avoid the transitional flow region entirely, to evade the various uncertainties associated with this flow regime. Avoiding the transitional flow regime is not always possible, as changes in operating conditions, design constraints, and various heat transfer augmentation methods often result in the heat exchanger operating in the transitional flow regime. Although some research has been conducted on the effect of different inlet configurations on the pressure drop characteristics in the transitional flow regime, most of the work focused on single tubes, whereas the bulk of the heat exchangers available in industry would typically have bundles of tubes. The aim of this investigation is to experimentally obtain fullydeveloped pressure drop characteristics in the transitional flow regime, for different tube pitch ratios, under isothermal and diabatic conditions. An experimental setup was designed and built to house three tubes, spaced apart at different pitch ratios. A single tube heat exchanger was used for validation purposes, since reliable pressure drop correlations are readily available in various published journals. The multitube test section consisted of smooth stainless steel tubes, each with an inner diameter and length of 4 mm and 6 m, respectively. The study was performed for Reynolds numbers ranging from 1 000 to 7 000, at three different heat fluxes. Transition effects were investigated for two tube pitch ratios, namely 1.25 and 1.5, based on the outer diameter of the tubes. It can be concluded that the presence of adjacent tubes had a significant effect under isothermal conditions, causing delayed and more abrupt transition from laminar to turbulent flow. This effect was more significant for the pitch spacing of 1.25 than for the pitch of 1.5. However, the disturbance caused at the inlet due to the presence of adjacent tubes seems to be damped out by buoyancy induced secondary flow effects with the addition of heat. This dampening effect becomes more significant for increasing heat fluxes.
E. Huisamen, 2017, "A thermohydraulic model that represents the current configuration of the SAFARI1 secondary cooling system"
This document focuses on the procedure and results of creating a thermohydraulic model of the secondary cooling system of the SAFARI1 research reactor at the Pelindaba facility of the South African Nuclear Energy Corporation (Necsa) to the west of Pretoria, South Africa.
The secondary cooling system is an open recirculating cooling system that comprises an array of parallelcoupled heat exchangers between the primary systems and the main heat sink system, which consists of multiple counterflowinduced draught cooling towers.
The original construction of the reactor was a turnkey installation, with no theoretical/technical support or verifiability. The design baseline is therefore not available and it is necessary to reverseengineer a system that could be modelled and characterised.
For the nuclear operator, it is essential to be able to make predictions and systematically implement modifications to improve system performance, such as to understand and modify the control system. Another objective is to identify the critical performance areas of the thermohydraulic system or to determine whether the cooling capacity of the secondary system meets the optimum original design characteristics.
The approach was to perform a comprehensive onedimensional modelling of all the available physical components, which was followed by using existing performance data to verify the accuracy and validity of the developed model. Where performance data is not available, separate analysis through computational fluid dynamics (CFD) modelling is performed to generate the required inputs.
The results yielded a model that is accurate within 10%. This is acceptable when compared to the variation within the supplied data, generated and assumed alternatives, and when considering the compounding effect of the large amount of interdependent components, each with their own characteristics and associated performance uncertainties.
The model pointed to potential problems within the current system, which comprised either an obstruction in a certain component or faulty measuring equipment. Furthermore, it was found that the current spray nozzles in the cooling towers are underutilised. It should be possible to use the current cooling tower arrangement to support a similar second reactor, although slight modifications would be required to ensure that the current system is not operated beyond its current limits. The interdependent nature of two parallel systems and the variability of the conditions that currently exist would require a similar analysis as the current model to determine the viability of using the existing cooling towers for an additional reactor.
C. Sanama, 2017, "Mathematical modelling of flow downstream of an orifice under flowaccelerated corrosion"
The main objective of this work is to establish an analytical model to evaluate the rate of corrosion in a horizontal pipe downstream of an orifice under flowaccelerated corrosion (FAC). FAC is a serious issue in nuclear and fossil power plants. In this work, an experimental setup was built to observe the effect of the flow on corrosion inside a tube. The experiments confirmed that the flow inside the tube caused more corrosion. However, accurate experimental data from literature has been selected and correlated by dimensional analysis, the modelling method of repeating variables and the Buckingham Pi theorem. It was found that the Sh number and the relative distance from the orifice are the main dimensionless parameters influencing FAC downstream of an orifice. The maximum value of the FAC rate could be wellpredicted for the OR of 0.25, while the location of the maximum FAC rate could be well predicted for the OR of 0.5. The maximum FAC rate occurs between 2D to 4D downstream of the orifice and increases with a decreasing OR. This work could be useful for professionals in industry and researchers in the field and could be the starting point for a new way of evaluating the FAC rate downstream of a flow’s singularity.
M.A.C Alfama, 2017, "Theoretical and Experimental Investigation of the Heat Transfer and Pressure Drop Optimisation on Textured Heat Transfer Surfaces"
Modern nuclear reactors still use Zirconium4 Alloy (Zircaloy®) as the cladding material for fuel elements. A substantial amount of research has been done to investigate the boiling heat transfer behind the cooling mechanism of the reactor. Boiling heat transfer is notoriously difficult to quantify in an acceptable manner and many empirical correlations have been derived in order to achieve some semblance of a mathematical model. It is well known that the surface conditions on the heat transfer surface plays a role in the formulation of the heat transfer coefficient but on the other hand it also has an effect on the pressure drop alongside the surface. It is therefore necessary to see whether there might be an optimum surface roughness that maximises heat transfer and still provides acceptably low pressure drop.
The purpose of this study was to experimentally measure pressure drop and heat transfer associated with vertical heated tubes surrounded by flowing water in order to produce flow boiling heat transfer. The boiling heat transfer data was used to ascertain what surface roughness range would be best for everyday functioning of nuclear reactors.
An experimental setup was designed and built, which included a removable panel that could be used to secure a variety of rods with different surface roughnesses. The pressure drop, surface temperature, flow rate and heat input measurements were taken and captured in order to analyse the heat transfer and friction factors.
Four rods were manufactured with different roughnesses along with a fifth rod, which remained standard. These rods were tested in the flow loop with water in the upward flow direction. Three different system mass flow rates were used: 0kg/s, 3.2kg/s and 6.4kg/s. Six repetitions were done on each rod for the tests, the first repetition was not used in the results since it served the purpose to deaerate the water in the flow loop. The full range of the power input was used for each repetition in the tests.
For the heat transfer coefficient at a system mass flow rate of 3.2kg/s, satisfactory comparisons were made between the test results and those found in literature with an average deviation of 14.53%. At 6.4kg/s system mass flow rate the comparisons deviated on average 55.45%. The velocity of the fluid in the test section was calculated from the pressure drop and was validated using separate tests. The plain rod, with no added roughness, was found to be the optimal surface roughness which is what is used in industry today.
The flow loop was in need of a couple of redesigns in order to produce more accurate results. Future work suggestions include adding more rods in the test section in order to investigate the nature of heat transfer in a rod bundle array as well as implementing all the suggested changes listed in the conclusion.
G.F. Knijnenburg, 2017, "Development of a vibration isolation system for a rotary wing unmanned aerial vehicle"
Antiresonance vibration isolation has long been a wellknown, studied and applied method for alleviating vibrations in stiff structures where small static deflection and a low transmissibility is needed, making it ideal for use in the rotorcraft industry. Most prior arts focus on passive single frequency antiresonance vibration isolation, while some, most notably liquid inertia vibration isolators, are adapted to actively isolate vibrations at more than one frequency. Very little literature is found on the adaptation of mechanical pendulum antiresonance vibration isolators for inflight tunable multiple frequency isolation, and although these systems predate the more modern liquid inertia type isolator, there is merit in their further development and use as low cost, robust and low maintenance isolators. A feasibility study on the performance of changing each fundamental design variable to achieve antiresonance tuning concludes, that for the antiresonance frequency shift range of interest in this dissertation, no specific design variable change quantifiably outperforms another with respect to tuning the antiresonance. Concept designs are created and investigated, finding the superior method of tuning the vibration isolator based on other criteria like overall weight, design simplicity, practicality, robustness and reliability. Shifting the tuning mass on the pendulum arm is deemed to be the superior concept, with respect to the helicopter being developed, and a tuneable multifrequency pendulum antiresonance vibration isolation system with a sliding concentrated mass is developed with ADAMS multibody dynamics software and SolidWorks. The isolation system along with a full scale dummy fuselage and transmissionrotor assembly is manufactured and experimentally tested. Initial experimental results show antiresonance frequencies 10Hz higher than the design targets, this phenomenon is later discovered to be related to friction in the pin joints of the pendulum hinges, increasing the system overall stiffness. Needle roller bearings are inserted to eliminate the friction, and experimental and ADAMS model results are again compared showing good correlation, with experimental results isolating close to the three target frequencies within 3% error. An astonishing level of vibration isolation is observed with the largest transmissibility obtained at the three frequencies being 0.5%. This dissertation proves the concept of a tuneable mechanical pendulum vibration isolator, and its design methodology, particularly with respect to shifting the position of the tuning mass. Suggestions for further work are: to implement this system with an actuation mechanism, further research on the effects of friction in isolators and the use of said phenomenon as a tuning method, development of isolators implementing the other concept of changing the design variables and a comparison between the effect of normal damping and friction damping on vibration isolation.
F.P.A. Prinsloo, 2017 "Investigation of turbulent heat transfer and pressure drop characteristics in the annuli of tubeintube heat exchangers (horizontal layout)"
Tubeintube heat exchangers are commonly used in many applications and are generally operated in a counterflow configuration. Unfortunately, existing correlations developed for heat transfer and pressure drop predictions for the outer annular flow passage have been found to sometimes produce large discrepancies between them.
In this experimental study research was performed to obtain experimental data with the lowest possible uncertainties associated with it in order to validate existing correlations and to identify the core aspects that influence the heat transfer and pressure drop characteristics in annular flow passages that have neither uniform wall temperatures nor uniform wall heat fluxes. Focus was placed on the turbulent flow regime and temperature and pressure drop measurements were taken at different fluid velocities, annular diameter ratios, and inlet temperature of water.
Four horizontal test sections with annular diameter ratios of 0.327, 0.386, 0.409 and 0.483 and hydraulic diameter of 17.00, 22.98, 20.20 and 26.18 respectively were constructed from hard drawn copper tubes. The test sections were equipped with industry standard inlet and outlet configurations and had pressure drop lengths of between 5.02 m and 5.03 m and heat transfer lengths of between 5.06 m and 5.10 m. This resulted in length to hydraulic diameter ratios of between 194 and 300. A wide range of annular flow rates were considered and Reynolds numbers ranges from 15 000 to 45 000 were covered for both heated and cooled annulus operating conditions. Specific attention was given to the influence of the inlet fluid temperature. For heated annulus cases an inlet temperature range of 10°C to 30°C was covered, while for cooled annulus cases an inlet temperature range of 30°C to 50°C was covered.
Since one of the main focuses of the study was to provide accurate temperature measurement, especially local wall temperature measurements of the inner tube, an insitu calibration technique of the wall thermocouples were used. This enabled continuous verification of the measurement accuracy and allowed reevaluation of readings.
Based on the processed experimental results, it was found that the direction of heat transfer did not affect the average heat transfer coefficient across the inner tube wall. Longitudinal local heat transfer coefficients were found to not be constant along the test section length, but continually decreased towards the annulus outlet, indicating undeveloped thermal flow. Heated annuli had a larger average heat transfer coefficients compared to cooled annuli at similar Reynolds numbers. This can be attributed to a dependency on fluid properties, which were less at higher bulk temperatures. Analysis showed although both had about the same local Nusselt numbers at the exit region, the heated annuli had much larger Nusselt numbers at the entrance region of the test section. The friction factor was mostly affected by the fluid velocity, but at low velocities higher friction factors were detected when inlet temperatures were lower.
For the data sets considered in this study, the average Nusselt number and the Colburn jfactor decreased somewhat with increase in annular diameter ratio. It seemed that the friction factor was also not influenced by the annular diameter ratio.
Supervisor:Prof. J. Dirker
Cosupervisor: Prof. J P. Meyer
JC Pieterse, 2017 "High Pressure Feedwater Heaters Replacement Optimisation "
Widespread uncertainty exists regarding the ideal replacement time of installed feedwater heaters in coal fired power plants. Eskom consequently identified the need for this research project to find the optimal age at which to replace high pressure (HP) feedwater heaters. Previous work has failed to quantify the unique financial risk of tube failures, which varies for individual heaters. Using life cycle cost (LCC) methodology, a framework is developed for the optimisation of the HP feedwater heater replacement age in Eskom coal fired power plants and integrated into existing software used in the organization. This entails identifying the most significant cost factors involved in the lifecycle of HP heaters and determining how they evolve over time by conducting a case study. Minimum life cycle cost for an actual HP heater is calculated in the case study based on failure data and cost information supplied by the power plant. This optimisation of replacement time can realise significant savings in annualised LCC compared to current practice.
Supervisor: Prof Jasper L. Coetzee
S Schmidt, 2017 "A costeffective diagnostic methodology using probabilistic approaches for gearboxes operating under nonstationary conditions"
Condition monitoring is very important for critical assets such as gearboxes used in the power and mining industries. Fluctuating operating conditions are inevitable for wind turbines and mining machines such as bucket wheel excavators and draglines due to the continuous fluctuating wind speeds and variations in ground properties, respectively. Many of the classical condition monitoring techniques have proven to be ineffective under fluctuating operating conditions and therefore more sophisticated techniques have to be developed. However, many of the signal processing tools that are appropriate for fluctuating operating conditions can be difficult to interpret, with the presence of incipient damage easily being overlooked.
In this study, a costeffective diagnostic methodology is developed, using machine learning techniques, to diagnose the condition of the machine in the presence of fluctuating operating conditions when only an acceleration signal, generated from a gearbox during normal operation, is available. The measured vibration signal is order tracked to preserve the anglecyclostationary properties of the data. A robust tacholess order tracking methodology is proposed in this study using probabilistic approaches. The measured vibration signal is order tracked with the tacholess order tracking method (as opposed to computed order tracking), since this reduces the implementation and the running cost of the diagnostic methodology.
Machine condition features, which are sensitive to changes in machine condition, are extracted from the order tracked vibration signal and processed. The machine condition features can be sensitive to operating condition changes as well. This makes it difficult to ascertain whether the changes in the machine condition features are due to changes in machine condition (i.e. a developing fault) or changes in operating conditions. This necessitates incorporating operating condition information into the diagnostic methodology to ensure that the inferred condition of the machine is not adversely affected by the fluctuating operating conditions. The operating conditions are not measured and therefore representative features are extracted and modelled with a hidden Markov model. The operating condition machine learning model aims to infer the operating condition state that was present during data acquisition from the operating condition features at each angle increment. The operating condition state information is used to optimise robust machine condition machine learning models, in the form of hidden Markov models.
The information from the operating condition and machine condition models are combined using a probabilistic approach to generate a discrepancy signal. This discrepancy signal represents the deviation of the current features from the expected behaviour of the features of a gearbox in a healthy condition. A second synchronous averaging process, an automatic alarm threshold for fault detection, a gearpinion discrepancy distribution and a healthydamaged decomposition of the discrepancy signal are proposed to provide an intuitive and robust representation of the condition of the gearbox under fluctuating operating conditions. This allows fault detection, localisation as well as trending to be performed on a gearbox during fluctuating operation conditions.
The proposed tacholess order tracking method is validated on seven datasets and the fault diagnostic methodology is validated on experimental as well as numerical data. Very promising results are obtained by the proposed tacholess order tracking method and by the diagnostic methodology.
Supervisor: Prof. PS Heyns
Cosupervisor: Dr. JP de Villiers
M S Cowley, 2017 "Optimising pressure profiles in superplastic forming"
Some metals, such as Ti6Al4V, have a high elongation to failure when strained at certain rates and temperatures. Superplastic forming is the utilisation of this property, and it can be used to form thin, geometrically complex components. Localised thinning occurs if the specimen is strained too quickly, and components with locally thin wall thickness fail prematurely. The superplastic forming process is investigated with the finite element method. The finite element method requires a material model that describes the superplastic behaviour of the metal. Several material models are investigated in order to select a material model that can predict localised thinning at higher strain rates.
An optimisation algorithm is developed to minimise the forming time of some component by prescribing the pressure profile, subject to a lower limit on the minimum thickness. This algorithm involves fitting a metamodel to simulated data (using the finite element method), and using the metamodels to search for the optimum pressure profile. The final forming time of the superplastic forming of a rectangular box was successfully minimised while limiting the final minimum thickness. The metamodels predicted that allowing a 4% decrease in the minimum allowable thickness (1.0 mm to 0.96 mm) that the forming time is decreased by 28.84%. The finite element verification indicates that the final minimum thickness reduced by 3.8% and that the forming time reduced by 28.81%.
Supervisor:Prof S. Kok
R. Kombo, 2017 "Qualitative analysis of flow patterns: Twophase flow condensation at low mass fluxes and different inclination angles "
A great deal of work has been conducted on intube condensation in horizontal and vertical smooth tubes. The available literature points to mechanisms governing two‑phase condensation heat transfer coefficients and pressure drops, which are directly linked to the local flow pattern for both horizontal and inclined configurations. However, the work has been limited to flow pattern observations, heat transfer, pressure drops and void fractions for both horizontal and inclined tubes at high mass fluxes. No work has been conducted on the analysis of the observed flow patterns and the effect of temperature difference between the average wall temperature and average saturation temperature for different inclination angles at mass fluxes of 100 kg/m^{2}.s and below. The purpose of this study is to carry out a qualitative analysis of flow patterns, and show the effect of temperature difference on the heat transfer coefficient for inclination angles from +90° (upward flow) to ‑90° (downward flow) at mass fluxes below 100 kg/m^{2}.s. An experimental setup provided the measurements for the twophase condensation of R‑143a in a smooth tube with an inside diameter of 8.38 mm and a length of 1.5 m. The mass fluxes were 25 kg/m^{2}.s to 100 kg/m^{2}.s, the saturation temperature was 40 °C and the mean qualities were 0.1 to 0.9. A high‑speed camera was used to visually analyse and determine the flow patterns for both the inlet and the outlet of the test section. Through the results, eight flow patterns were observed: stratified‑wavy, stratified, wavy, wavy‑churn, intermittent, churn, annular and wavy‑annular. The maximum heat transfer was observed for downward flow between inclination angles of ‑15° and ‑30°. The ThomeHajal flow pattern map correctly predicted horizontal flow patterns, but failed to predict most of the inclined flow patterns. Various flow pattern transitions were identified and proposed for all the investigated inclination angles in this study. Finally, the heat transfer coefficient was found to be dependent on quality, mass flux, temperature difference and inclination angle.
Supervisor: Pro. J.P. Meyer
B. W. Kohlmeyer, 2017 "Development of an improved design correlation for local heat transfer coefficients at the inlet regions of annular flow passages"
Several applications, including those in the energy sector that require high thermal efficiency, such as those in the solar energy industry, require a careful thermal analysis of heat exchange components. In this regard, thermal resistance is a major cause of exergy destruction and must be minimised as much as possible, but also adequately designed.
In the past, a number of correlations have been developed to predict heat transfer coefficients in compact heat exchangers. The designers of such heat exchangers often exploit the development of thermal boundary layers to achieve higher overall efficiency due to increases in local heat transfer coefficients. However, most of the correlations that have been developed for heat exchangers neglect the specific effect of the thermal boundary layer development in the inlet region, and instead only offer effective average heat transfer coefficients, which most users assume to be constant throughout the heat exchanger. This is often an oversimplification and leads to overdesigned heat exchangers.
In this study, focus is placed on annular flow passages with uniform heating on the inner wall. This geometry has many applications. This study aims to collect experimental heat transfer data for water at various flow rates and inlet geometries, to process the data and determine local and overall heat transfer coefficients, and to develop an improved local heat transfer coefficient correlation.
Experimental tests were performed on a horizontal concentric tubeintube heat exchanger with a length of 1.05 m and a diameter ratio of 0.648. The surface of the inner tube was treated with thermochromic liquid crystals (TLCs), which allowed for highresolution temperature mapping of the heated surface when combined with an automated camera position system in order to determine local heat transfer coefficients. Conventional inline and outofline annular inlet configurations were evaluated for Reynolds numbers from 2000 to 7 500, as well as the transition from laminar to turbulent flow for a single inline inlet configuration.
It was found that the local heat transfer coefficients were significantly higher at the inlets, and decreased as the boundary layers developed. With the high resolution of the results, the local heat transfer coefficients were investigated in detail. Local maximum and minimum heat transfer coefficients were identified where the thermal boundary layers merged for high turbulent flow cases. The annular inlet geometries only influenced the heat transfer for Reynolds numbers larger than 4000, for which larger inlets are favoured. Outofline inlet geometries are not favoured for heat transfer. A new heat transfer correlation was developed from the experimental data, based on an existing heat transfer correlation for turbulent flow in an annular flow passage, considering the boundary layer development. The new correlation estimated the areaweighted heat transfer coefficients within 10% of the experimental data and closely followed trends for local heat transfer coefficients.
Supervisor: Prof. J. Dirker
Cosupervisor: Prof. J P. Meyer
M Kandindi, 2017 "Heat transfer and pressure drop investigation for prescribed heat fluxes on both the inner and outer wall of an annular duct"
Heat exchangers are used in industrial processes to recover heat between two processes fluids and are widely used. Although the equations for heat transfer and pressure drop characteristics in a doublepipe heat exchangers are available, there is still need to completely understand how these characteristics interact which geometrical factors like annular diameter ratio or some thermal boundaries conditions which have not yet drawn more attention from the research community.
The purpose of this study was to experimentally measure the heat transfer and pressure drop characteristics of a concentric annular duct of ratio 0.593 for different heat fluxes simultaneously on the inner and outer tube in the turbulent flow regime and to describe or discuss the impact or interaction of heat flux ratios on the flow and heat transfer behaviour.
An experimental setup was designed to achieve this goal. It consisted of an overall facility and a removable test section. The test section allowed for the measurement of the temperature along the length of the test section, the pressure drop, the heat flux inputs and the flow rate. These quantities were used to determine the heat transfer coefficients and friction factors of the system.
The concentric duct was an annulus formed of a single (15.88mmouter diameter and 14.46mminner diameter) copper tube inserted inside a 0.91mm thick copper tube of 26.76 mm of inner diameter. The overall length of the annular duct was 4.84 m. To transfer heat, a heating element made of constantan wire was wrapped around each heat transfer area.
Heat transfer and pressure drop data were obtained on heating the inner and the outer wall separately with four different heat flux densities and eight heat flux ratio were used for the case of simultaneously heating both walls. Reynolds numbers for unilateral heating range from 5 800 to 12 000 while bilateral heating were focus around two Reynolds numbers, 6 500 and 9 500.
Satisfactory results were found between the measurements of this experiment and currently available literature for the case of unilateral heating. An estimate of the accuracy of the experimental setup showed the maximum relative error was about 5 % in the determination of the Nusselt number and 1.8 % for the friction factor.
Diabatic friction factors have been presented using adiabatic friction factors with a correction term which considered the effect of temperature difference between the fluid and walls. Heat flux density ratio showed to have an impact on the heat transfer characteristics. The Nusselt number on the inner wall could be enhanced by 19% with increasing the heat flux ratio up to 2.3 times.
Supervisor: Prof. J. Dirker
Cosupervisor: Prof. J P. Meyer
K J Mujanayi, 2017 "Thermal Management and Optimization of Heat Transfer from Discrete Heat Sources"
These days, the cooling of new generation electronic servers is a challenge due to the immense heat generated by them. In order to avoid overheating caused by the important rise in temperature appropriate cooling procedures must be used in order to meet the thermal requirement. The current study aims at addressing the issue of overheating in this field, and focuses on the thermal management of electronic devices modelled as a discrete heat sources (mounted in a rectangular cavity) with uniform heat flux applied from the bottom. A review of the literature published regarding the convective heat transfer from heated sources as well as a thorough background on the theory of the cooling of discrete sources by forced convection in rectangular channel is provided in this study. It was showed that the heat transfer performance in channel is strongly influenced by the geometric configurations of heat sources. Therefore, the arrangement and geometric optimisation are the main considerations in the evaluation of thermal performance. Unlike experimental methods that were carried out widely in the past, which provided less costeffective and more timeconsuming means of achieving the same objective, in this study we first explore the possibilities and the advantages of using the CDadapco’s CFD package StarCCM+ to launch a three dimensional investigation of forced convection heat transfer performance in a channel mounted with equidistant heatgenerating blocks. Numerical results were validated with available experimental data, and showed that the thermal performance of the heat transfer increases with the strength of the flow. The second objective was to maximise the heat transfer density rate to the cooling fluid and to minimise both the average and the maximum temperature in the channel by using the numerical optimisation tool HEEDS/Optimate+. The optimal results showed that better thermal performance was not obtained when the heated sources followed the traditional equidistance arrangement, but was achieved with a specific optimal arrangement under the total length constraint for the first case. Subsequently, for the second case study, on the volume constraints of heat sources, the results proved that optimal configurations that maximise the heat transfer density rate were obtained with a maximum of either the heighttolength ratio or the heighttowidth ratio. It was concluded that the heat transfer rate to the cooling fluid increases significantly with the Reynolds number and the optimal results obtained numerically are found to be fairly reliable.
Keywords: Thermal management, discrete heat sources, CFD package, forced convection, numerical optimization, maximise heat transfer, optimal configuration, volume constraints
Supervisor: Prof. T BelloOchende
Cosupervisor: Prof. J P. Meyer
J. C. P. Brits, 2017 "An Experimental and Stochastic Approach to Estimate the Fatigue Crack Life of a Turbomachinery Blade using Finite Element Modelling"
Large rotating machines are expensive and not easily replaceable or repairable. If the useful life of turbomachinery blades has been reached, failure can occur and lead to unplanned downtime, repair, and maintenance costs. If a crack is found on a component during an inspection, the extent of the damage is not always certain. It is also not always possible to replace the cracked component or allow downtime while waiting for a replacement to be manufactured or shipped. Costly inspections will also be needed on the damaged blades, until they are replaced. Since a cracked component can still be in service, the fatigue crack life should be “known” before recommissioning to improve safety and make proper budgeting and planning for maintenance possible.
An approach to include modelling uncertainties and material variations in the input parameters when predicting the fatigue crack life of a turbomachinery blade during resonance conditions has been developed in the present study. As result, the reliability of the estimated lifetime is quantifiable. The determination of the fatigue crack life of a component is affected by various factors and these unknowns are generally taken into account by using conservative assumptions in deterministic models and rarely include a measure of uncertainty.
A FE model, built in MSC.Marc/Mentat 2016, was used to create a library of cracks with associated stress intensity factors of representative cracks within axial fan blades under cyclic loading. An experimental setup was designed to initiate and propagate a crack on multiple different blades to characterize the blade material using the RajuNewman formulation on a simplified geometry. To stimulate crack growth, a base excitation, at resonance, was applied to the test specimens and the crack growth was measured with digital image correlation. A Monte Carlo simulation was employed to assess the sensitivity of the lifetime estimation to material variations and modelling uncertainties.
Supervisor:Prof P. Stephan Heyns
Cosupervisor: Dr Helen M. Inglis
ATC Hall, 2017 "The Effect of Inlet Header Geometry on the Heat Transfer Performance of Smooth Horizontal Tubes in the Transitional Regime"
Heat exchangers are seen to bear significance in many different industries, especially in the generation of energy in its various forms. Accurate design information is therefore required in order to improve the efficiency of these systems. Heat exchangers often end up operating in the transitional flow regime, or close to the transitional flow regime. Previous studies in this flow regime concentrated on single tube test sections with a variety of inlet geometries. In some heat exchangers, such as in chillers with a large number of tubes, not every tube has its own inlet but an inlet header feeds the tubes. However, no work has been done to study the effect of such an inlet header geometry on the heat transfer in adjacent tubes. It was therefore the purpose of this study to experimentally investigate the effect of an inlet header on heat transfer in the transitional flow regime. An experimental test setup was constructed and commissioned for this purpose, that operated on water as working fluid and was validated against existing literature using results obtained from a single tube test section. A threetube inlet header was then used to obtain heat transfer measurements on three tubes in parallel across Reynolds numbers ranging from 950 to 6 200, Prandtl numbers of 3.6 to 5.7, at a heat flux of 3 kW/m². The tube inner diameter was 3.97 mm and the tube length was 6 m. Inlet and outlet temperatures were recorded, in addition to surface temperature measurements along the length of each tube, as well as flow rate in each individual tube. Comparisons were made of the heat transfer coefficients over the last 2 m of the tube where the flow was fully developed. An uncertainty analysis was done, revealing uncertainties to vary between 11 and 16% in the Nusselt numbers and between 4 and 6% in Colburn jfactors, while uncertainty in the Reynolds numbers remaining less than 3 and 5% throughout the testing range. It was found that the use of a three tube inlet header resulted in increased heat transfer performance in the centre tube of the test section. In addition, transition was seen to occur earlier in the centre tube, followed by a secondary transition that aligned with the transition observed in the outer tubes. It was noted that the heat generated by the outer tubes may have influenced the heat transfer performance of the centre tube.
Supervisor: Professor JP Meyer
T L Ottermann, 2017 "Experimental and Numerical investigation into the natural convection of TiO2Water nanofluid inside a cavity"
This Master of Engineering investigation focuses on the natural convection of nanofluids in rectangular cavities. The governing equations applied to analyse the heat transfer and fluid flow occurring within the cavity are given and discussed. Special attention is given to the models that were developed to predict the thermal conductivity and dynamic viscosity of such nanofluids.
A review concerning past investigations into the field of natural convection of nanofluids in cavities is made. The investigation is divided into experimental works and computational fluid dynamics (CFD) numerical investigations.
Through the literature review, it was discovered that many numerical models exist for the prediction of the thermophysical properties of nanofluids, specifically thermal conductivity and viscosity. Depending on the nanofluid and the application, different models can be used.
The literature study also revealed that most previous works were done in the CFD field. Very few experimental studies have been performed. Numerical CFD investigations, however, need experimental results for validation purposes, leading to the conclusion that more experimental work is needed.
The heat transfer capability and thermophysical properties of the nanofluid are investigated based on models found in the literature. The investigation includes measuring the heat transfer inside a cavity filled with a nanofluid and subjected to a temperature gradient. The experiment is performed for several volume fractions of particles. An optimum volume fraction of 0.005 is obtained. At this volume fraction, the heat transfer enhancement reaches a maximum for the present investigation.
The investigation is repeated as a numerical investigation using the commercially available CFD software ANSYSFLUENT. The same case as used in the experimental investigation is modelled as a twodimensional case and the results are compared. The same optimum volume fraction and maximum heat transfer is obtained with an insignificantly small difference between the two methods of investigation. This error can be attributed to the minor heat losses experienced from the experimental setup as in the CFD adiabatic walls considered.
It is concluded that, through the inclusion of TiO2 particles in the base fluid (deionised water), the thermophysical properties and the heat transfer capability of the fluid are altered. For a volume fraction of 0.005 and heat transfer at a temperature difference of 50 °C, the heat transferred through the fluid in the cavity is increased by more than 8%.
From the results, it is recommended that the investigation is repeated with TiO2 particles of a different size to determine the dependency of the heat transfer increase on the particle size. Various materials should also be tested to determine the effect that material type has on the heat transfer increase.
Supervisor: Prof. Mohsen Sharifpur
Cosupervisor: Prof. Josua P Meyer
E Grove, 2017 "Feasibility study on the implementation of a boiling condenser in a South African fossil fuel power plant"
The South African electricity mix is highly dependent on subcritical coalfired power stations. The average thermal efficiency of these power plants is low. Traditional methods to increase the thermal efficiency of the cycle have been widely studied and implemented. However, utilising the waste heat at the condenser, which accounts for the biggest heat loss in the cycle, presents a large potential to increase the thermal efficiency of the cycle. Several methods can be implemented for the recovery and utilisation of lowgrade waste heat.
This theoretical study focuses on replacing the traditional condenser in a fossil fuel power station with a boiling condenser (BC), which operates in a similar manner to the core of a boiling water reactor at a nuclear power plant (Sharifpur, 2007). The system was theoretically tested at the Komati Power Station, South Africa’s oldest power station. The power station presented an average lowgrade waste heat source. The BC cycle was theoretically tested with several working fluids and numerous different configurations. Several of the theoretical configurations indicated increased thermal efficiency of the cycle. The BC cycle configurations were also tested in two theoretical scenarios.
Thirty configurations and 103 working fluids were tested in these configurations. The configuration that indicated the highest increase in thermal efficiency was the BC cycle with regeneration (three regenerative heat exchangers) from the BC turbine. A 2.4% increase in thermal efficiency was obtained for the mentioned theoretical implementation of this configuration. The working fluid tested in this configuration was ethanol. This configuration also indicated a 7.6 MW generating capacity.
The increased thermal efficiency of the power station presents benefits not only in increasing the available capacity on South Africa’s strained grid, but also environmental benefits. The mentioned reduction of 7.6 MW in heat released into the atmosphere also indicated a direct environmental benefit. The increase in thermal efficiency could also reduce CO_{2} emissions released annually in tons per MW by 5.74%.
The highlevel economic analysis conducted, based on the theoretically implemented BC cycle with the highest increase in thermal efficiency, resulted in a possible saving of R46 million per annum. This translated to a saving of R19.2 million per annum for each percentage increase in thermal efficiency brought about by the BC cycle.
The theoretical implementation of the BC, with regeneration (three regenerative heat exchangers) from the BC turbine and ethanol as a working fluid, not only indicated an increase in thermal efficiency, but also significant economic and environmental benefits.
Supervisor: Prof. Mohsen Sharifpur
Cosupervisor: Prof. Josua P Meyer
KG Katamba, 2017 "Investigation into waste heat to work in thermal systems in order to gain more efficiency and less environmental defect"
In most previous studies that have been conducted on converting waste heat energy from exhaust gases into useful energy, the engine waste heat recovery system has been placed along the exhaust flow pipe where the temperature differs from the temperature just behind the exhaust valves. This means that an important fraction of the energy from the exhaust gases is still lost to the environment. The present work investigates the potential thermodynamic analysis of an integrated exhaust waste heat recovery (EWHR) system based on a Rankine cycle on an engine’s exhaust manifold. The amount of lost energy contained in the exhaust gases at the exhaust manifold level, at average temperatures of 500 °C and 350 °C (for petrol and diesel), and the thermodynamic composition of these gases were determined. For heat to occur, a temperature difference (between the exhaust gas and the working fluid) at the pinch point of 20°C was considered. A thermodynamic analysis was performed on different configurations of EWHR thermal efficiencies and the selected suitable working fluids. The environmental and economic aspects of the integrated EWHR system just behind the exhaust valves of an internal combustion engine (ICE) were analysed. Among all working fluids that were used when the thermodynamic analysis was performed, water was selected as the best working fluid due to its higher thermal efficiency, availability, low cost and environmentally friendly characteristics. Using the typical engine data, results showed that almost 29.54% of exhaust waste heat can be converted. This results in better engine efficiency and fuel consumption on a global scale by gaining an average of 1 114.98 Mb and 1 126.63 Mb of petrol and diesel respectively from 2020 to 2040. It can combat global warming by recovering 56.78 1 011 MJ and 64.65 1 011 MJ of heat rejected from petrol and diesel engines, respectively. A case study of a Volkswagen Citi Golf 1.3i is considered, as it is a popular vehicle in South Africa. This idea can be applied to newdesign hybrid vehicles that can use the waste heat to charge the batteries when the engine operates on fossil fuel.
Supervisor: Prof. Mohsen Sharifpur
Cosupervisor: Prof. Josua P Meyer
J Joubert, 2017 "Influence of a magnetic field on magnetic nanofluids for the purpose of enhancing natural convection heat transfer"
Natural convection as a heat transfer mechanism plays a major role in the functioning of many heat transfer devices, such as heat exchangers, energy storage, thermal management and solar collectors. All of these have a large impact on the generation of solar power. Considering how common these devices are – not only in power generation cycles, but in a majority of other thermal uses – it is clear that increased performance for natural convection heat transfer will have consequences of a high impact. As such, the purpose of this study is to experimentally study the natural convection heat transfer behaviour of a relatively new class of fluids where nanosized particles are mixed into a base fluid, also known as nanofluids. Nanofluids have attracted widespread interest as a new heat transfer fluid due to the fact that the addition of nanoparticles considerably increases the thermophysical properties of the nanofluids when compared to those of the base fluid. Furthermore, if these nanoparticles show magnetic behaviour, huge increases to the thermal conductivity and viscosity of the nanofluid can be obtained if the fluid is exposed to a proper magnetic field. With this in mind, the study aimed to experimentally show the behaviour of these socalled magnetic nanofluids in natural convection heat transfer applications.
In this study, the natural convection heat transfer of a magnetic nanofluid in a differentially heated cavity is investigated with and without an applied external magnetic field. The effects of volume concentration and magnetic field configuration are investigated. Spherical Fe_2 O_3 nanoparticles with a diameter of 20 nm are used with a volume concentration ranging between 0.05% and 0.3%, tested for the case with no magnetic field, while only a volume concentration of 0.1% was used in the magnetic cases. The experiments were conducted for a range of Rayleigh numbers in 1.7 ×〖 10〗^8<Ra<4.2 ×〖 10〗^8. The viscosity of the nanofluid was determined experimentally, while an empirical model from the literature was used to predict the thermal conductivity of the nanofluids. An empirical correlation for the viscosity was determined, and the stability of various nanofluids was investigated.
Using heat transfer data obtained from the cavity, the average heat transfer coefficient, as well as the average Nusselt number for the nanofluids, is determined. It was found that a volume concentration of 0.05% showed an increase of 3.75% in heat transfer performance. For the magnetic field study, it was found that the bestperforming magnetic field enhanced the heat transfer performance by 1.58% compared to the 0.1% volume concentration of the nanofluid with no magnetic field.
Supervisor: Prof. Mohsen Sharifpur
Cosupervisor: Prof. Josua P Meyer
S Roberts, 2017 "Characterising the Behaviour of an Electromagnetic Levitation Cell using Numerical Modelling"
Experimental investigations of high temperature industrial processes, for example the melting and smelting processes taking place inside furnaces, are complicated by the high temperatures and the chemically reactive environment in which they take place. Fortunately, mathematical models can be used in conjunction with the limited experimental results that are available to gain insight into these high temperature processes. However, mathematical models of high temperature processes require high temperature material properties, which are difficult to measure experimentally since container materials are often unable to withstand high enough temperatures, and sample contamination often occurs. These difficulties can be overcome by employing containerless processing techniques such as electromagnetic levitation to allow for characterisation of high temperature material properties.
Efficient design of electromagnetic levitation cells is challenging since the effects of changes in coil design, sample size and sample material on levitation force and sample temperature are not yet well understood. In this work a numerical model of the electromagnetic levitation cell is implemented and used to investigate the sensitivity of levitation cell operation to variations in coil design, sample material and sample size.
Various levitation cell modelling methods in literature are reviewed and a suitable model is chosen, adapted for the current application, and implemented in Python. The finite volume electromagnetic component of the model is derived from Maxwell's equations, while heat transfer is modelled using a lumped parameter energy balance based on the first law of thermodynamics. The implemented model is verified for a simple case with a known analytical solution, and validated against published experimental results. It is found that a calibrated model can predict the lifting force inside the levitation cell, as well as the sample temperature at low coil currents.
The validated model is used to characterise the operation of a levitation cell for a number of different sample materials and sample sizes, and for varying coil geometry and coil current. The model can be used in this way to investigate a variety of cases and hence to support experimental levitation cell design. Based on model results, operating procedure recommendations are also made.
Supervisor: Prof. S. Kok
Cosupervisors: Dr J. H. Zietsman, Dr H. M. Inglis
M.J.R. Schoeman, 2017 "Development and comparison of strategies for the reconstruction of full and partial skull geometries"
The development and comparison of strategies for the reconstruction of full and partial surface mesh based skull geometries is presented. The intended application is to aid the South African Police Service Victim Identification Centre (SAPS VIC) with forensics, specifically prediction of a mandible when only the cranium is available.
Various methods for the registration of surface meshes are outlined. A new nonrigid iterative closest point (NRICP) algorithm based on an adaptively refined least square Radial Basis Function (RBF) approximation of the forward and backward nearest neighbour correspondence is developed. The newly developed nonrigid registration strategy is demonstrated and characterised for various parameters using an artificial mandible dataset constructed through MonteCarlo (MC) sampling of a quadratic displacement field. Various suitable parameters are shown to result in imperceptible visual registration differences, with the correspondence error mainly distributed insurface.
Multivariate regression techniques suited to the application of geometry prediction are considered, specifically for cases where the data is expected to be multicollinear and the number of variables are far greater than the number of observations. Two regression approaches based on spatial information are considered. The first is the classical use of Procrustes Analysis where the Cartesian coordinates are used directly for regression. The second is a new Euclidean distance based approach utilizing pairwise distances to consistent reference points. The proposed regression methods' timespace complexity is investigated to limit system sizes that result in time tractable crossvalidation and model comparison. Pre and postprocessing required for tractability considerations are also developed for both approaches.
Proof of concept of the registration based prediction strategies are demonstrated. This is accomplished through the use of an artificial dataset with embedded covariance and the use of registration targets without pointwise correspondence. The registration based prediction strategy is shown to be capable of accurate predictions for data with strong underlying structure/covariance.
The proposed registration based prediction strategy is demonstrated on a real cranium and mandible dataset, where the mandible geometry is predicted from the cranium geometry. Marginal improvement over the geometric mean is obtained. Observation scaling suggests that model accuracy is improved for increased observations, which merits expanding the dataset.
The proposed registration strategy has the limitation that it is not capable of registration of significant partial/incomplete geometries. A new regressionregistration hybrid strategy is developed for use with partial geometries, when a full dataset of the given geometry is available. The regressionregistration hybrid strategy is demonstrated on a real mandible dataset and mandible fossil.
Supervisor: Prof. S. Kok
Cosupervisor: Dr. D.N. WIlke
MD Marais, 2017 "Computational fluid dynamics investigation of wind loads on heliostat structures"
Heliostats, mirrors tracking the sun and reflecting onto a target, make up the solar collector field of a central receiver solar plant. Since a large amount of these structures are needed to achieve high temperatures at the receiver, they make up a significant portion of the initial capital investment. The optimum design of heliostats is, therefore, an important research field as it has been identified as a key cost saving area in the bid to make concentrating solar power a viable alternative to current fossil fuel technologies.
Due to topology considerations, central receiver plants are generally constructed in flat, open country environments where the heliostat fields are subjected to atmospheric winds. The structural design of the individual heliostats has to take into account the windinduced forces and moments, being able to resist structural failure during storm loads with the mirrors in stow position. Heliostats should also not suffer static deflections or vibrations which causes reflected solar radiation to miss the target during operation, a time when the mirror is at varying angles to the attacking wind. Overdesign should be avoided in order to limit expenditure.
The prediction of mean and peak wind loads on heliostat structures has, therefore, been a key research area for more than three decades. Experimental wind tunnel studies, in which the atmospheric wind profiles for velocity and turbulence are replicated, have been the favoured method to obtain wind loading data. Computational fluid dynamics (CFD) have been less popular, but have the potential to serve as a cost effective analysis and optimization tool, provided that such models are properly validated using experimental data.
Supervisor: Prof KJ Craig
Cosupervisor: Prof JP Meyer
J.C. Smit, 2017 "Flow regime identification and compensation for solid flow measurement using concave capacitive sensors"
This dissertation describes research and experimental work done to improve the accuracy and applicability of capacitive sensors which are used to evaluate the mass flow rate of solid material within pneumatic conveying environments in power plant engineering applications. The research is focused on creating a measurement methodology on which solid mass flow rate can be evaluated without being dependent on variable flow regimes. The proposed research investigates the process of identifying four different types of flow regimes using a decision tree that utilises support vector machines and cut off values. During the study an investigation was also made into what electrode setup would provide proficient information regarding the flow distribution and orientation. Compensation according to the identified regime is then proposed by means of using a nonlinear fraction curves determined through calibration experiments.
Supervisor: Prof P.S. Heyns
CoSupervisor: Mr H Fourie
AJ Vogel, 2017 "Comparing direct and indirect methods for lowbudget tuning of heuristic optimisation algorithms"
Heuristic algorithms have parameters that control their performance on specific optimisation problems. By choosing optimal parameter values researchers can drastically reduce the time it takes to solve optimisation problems. Unfortunately, these optimal parameters depend on several factors such as the algorithm used, the problem being solved, and the fitness budget available. So, while some optimal parameter values are published for many algorithms, they are only applicable to a very specific case. For this reason researchers have spent some time developing methods to find these optimal parameter values, a process known as tuning. The performance of heuristic algorithms are stochastic and tuning methods must compensate for this in some way. Most algorithms do this by evaluating a single parameter set several times (usually 30 to 50) and taking the average performance of this parameter set as the nominal for comparison. We believe this approach is suboptimal as the tuning budget is wasted on parameter sets that are suboptimal.
In this thesis we propose an alternative approach that uses regression through spatially distributed points to compensate for the stochastic response of heuristic algorithms. This approach allows more efficient use of the tuning budget. In our approach we use a radial basis function response surface to create a regression surface through the points. This surface is then optimised to find a new candidate parameter set.
In a series of numerical experiments we show that our approach outperforms commonly used tuning methods, especially when tuning budgets are low.
Supervisor: Dr DN Wilke
D Kafka, 2017 "Investigation into regression strategies to address model errors in inverse analysis of creep models"
When solving inverse problems, the model error is a critical aspect to be considered, as it affects the validity of the solution. The forward method is the de facto standard technique used to solve inverse problems. It is, however, limited in the accuracy of its fit by the validity of the model used. If global convergence is achieved, it constitutes the best possible fit for the given model. This strategy can be computationally expensive in the context of engineering technologies such as the Finite Element Method. This is often alleviated using response surfaces to approximate the response of the desired analysis. An alternative approach is direct inverse mapping strategies or inverse regression. Inverse regression strategies offer a computationally efficient means of solving inverse problems in particular when they need to be solved multiple times. The absolute accuracy of these methods is dependent on the data used to construct the inverse regression. However, they offer the ability to extract features from the data used to construct the inverse regression. A linear combination of the extracted features is then used to construct the inverse regression. This study shows that in the case of using heterogeneous training data, with data from two different models, the mappings constructed are capable of accurately predicting the response, which is a linear combination of measured responses related to the models independently present in the training dataset. Thus, it is shown that inverse regression is capable of reducing the model error by combining multiple independent models to characterize a single problem.
Supervisor: Dr DN Wilke
Y Chae, 2017 "Optimal sensor placement approaches for the design of inverse experiments by simulation"
This dissertation serves to present the research conducted on sensor placement optimisation (SPO) using sensitivity analyses of virtual experiments in order to design virtual inverse problems. Two classes of SPO methods are considered namely modebased and modefree methods. The modebased methods make use of SIMPLS and SVD to extract useful data by examining the correlation between the target variables (characterising variables) and the sensor measurement variables, while the modefree methods eliminate the need of spending the extra time required to extract modes, which ultimately leads to successful sensor placement for solving inverse problems.
The aim of the modefree approach is to maximize the variance explained subject to uniqueness of the information of each sensor. Both approaches aim to maximize the potential of an experimental setup to solve an inverse problem by using the right number of sensors and placing them at the optimal spatial positions. SPO is not only capable of designing an experiment but it is also capable of classifying the wellposed or illposed nature of an existing experiment that can be modelled, which saves both the time and cost. The approach followed in this study was to design a simple virtual inverse problem for which the (well or ill)posedness of the problem can be controlled.
Numerous virtual experiments were conducted that varied from wellposed to severely illposed to allow for rigorous testing of the various approaches. The effect of model error and stochastic noise on ability to reliably place sensors is also investigated.
Supervisor: Dr DN Wilke
R. Strauss, 2017 "Brake Based Integrated Rollover Prevention and Yaw Control for an OffRoad Vehicle."
Sport utility vehicles typically feature high ground clearances that allow them to be used in offroad conditions. Their use is not limited to offroad conditions and they are often used as daytoday family vehicles. On the road, where high friction surfaces are prevalent, their high centres of gravity can make them susceptible to untripped rollovers during severe dynamic manoeuvers such as an emergency obstacle avoidance. The detection of a high risk of rollover and the avoidance thereof has great potential to improve vehicle safety, as the consequences of rollover incidents are generally quite severe.
Rollover mitigation systems are triggered when a rollover threshold index is exceeded, indicating a high risk of rollover. The metric implemented in this study is known as the zeromoment point method, which allows for vehicle parameters and terrain to be taken into account. Previous research has indicated that mitigation systems that trigger braking intervention are some of the most successful methods in reducing rollover risk, as it not only stabilises the vehicle, but also reduces the speed.
Brake based rollover prevention systems typically implement electronic stability program methods that use yaw rate reduction as the primary tool for reducing rollover risk, which often comes at the expense of the vehicle’s path following ability. This means that the stability control system may lead to the vehicle leaving the road and causing an even more severe accident. The control algorithm implemented in this study gives preference to reducing the forward speed of the vehicle which in turn reduces lateral acceleration, a major contributor to rollover propensity. Braking is however apportioned to all four wheels and distributed so as to achieve vehicle yaw rate targets. Emphasis is placed on maintaining good path following capability to prevent the vehicle from leaving the road.
The detection and mitigation system was tested on a Land Rover Defender 110 for a variety of manoeuvers in simulation as well as experimental testing. The results indicate that the rollover mitigation system managed to successfully reduce the rollover threshold index of the vehicle during the manoeuver whilst simultaneously maintaining the path following ability of the vehicle and improved the yaw rate tracking.
Supervisor: Prof. P. S. Els
GJ Howard, 2017 "Finite Element Modelling of Creep for an Industrial Application"
Thermal power stations operate at elevated temperatures and pressures in order to attain maximum available steam energy. At these high temperatures creep becomes a dominant mechanism that needs to be considered. However, for many components, the locations where peak stresses occur are unreachable to apply the commonly used NonDestructive Testing (NDT) techniques. This encourages the use of Finite Element Analysis (FEA) to better predict the creep state in these complex components.
Commonly, creep damage models are used in conjunction with accelerated creep tests to develop material models that can be implemented into a FEA to determine failure. These approaches are often infeasible for industrial decisionmaking, leaving a gap for more accessible commercially available models to be developed. This paper focuses on using openly available creep data from the Japanese National Institute for Material Science (NIMS). A creep strain model capable of modelling only the primary and secondary creep regimes was then chosen from the ANSYS database to fit this data. In order to fully characterise the experimental data a multicreepmodel approach was adopted that uses a family of creep models, instead of a single creep material model, to characterise the probable range of responses. This methodology was applied to an industrial application, namely an Intermediate Pressure (IP) valve operating under creepprone conditions. The multicreepmodel approach was incorporated into FEA to analyse the variation in stress distributions. It was interesting to see that a variation of 153% in the creep strain models only resulted in a 21% variation in the relaxed stress. Worst case scenario life time calculations were then conducted using both a timebased LarsonMiller approach and a strainbased ASME code approach. Both sets of results showed that, for the specific component of interest, creep rupture lifetimes were in excess of 3000 years. It was therefore noted that, for the IP valve of interest, the operating temperature and pressure combination were such that no worrisome creep damage occurred. In conclusion, for the specific component analysed, the operating conditions are such that creep based failure will not occur.
Supervisor: Dr HM Inglis
CoSupervisor: Mr F Pietra
2016
SB Leith, 2016 "An Investigation into the External Flow Boiling Phenomena on the Surface of Water Cooled Zircaloy4 and Silicon Carbide Nuclear Fuel Cladding"
The materials in use inside modern nuclear reactors have all been subjected to a large amount of research and development. This is necessary due to the challenging environment found inside the core of a nuclear reactor as well as the stringent safety requirements imposed on their construction. The materials need to be able to handle a multitude of different circumstances and there are many attributes which a core material needs to have in order to be sustainable. Nowadays, the focus of most reactor material research is on safety. For this reason, coupled with the advanced materials available today, a new cladding material has been proposed. This material is Silicon Carbide (SiC). Therefore the purpose of this study was to experimentally investigate, and compare, the heat transfer characteristics of SiC versus current nuclear grade Zircaloy4 hereinafter called Zircaloy®. The experiments were thus focused on external flow boiling. The setup used for the experimentation was designed by a colleague and the test section was designed purposefully for surface temperature measurements to be taken at various mass flow rates, inlet temperatures and pressures. The setup was constructed so that the cladding rods are positioned vertically with the heat transfer medium flowing vertically upwards around it. In order to replicate the heat generated by nuclear fuel, a 3 kW electric resistance cartridge heater was inserted inside both the SiC as well as the Zircaloy®. This resulted in heat fluxes ranging from 20 kW/m^{2} to 145 kW/m^{2} for Zircaloy® and from 12 kW/m^{2} to 90 kW/m^{2} for silicon carbide. A total of twenty four tests were conducted each with six heat flux set points and three flow rates. The test specimens were two Zr4 and SiC tubes, with diameters of 9.8 mm and 15.5 mm, supported inside a square test section. Water at 100°C, 11.37°C subcooling, was pumped through the test section as the heat transfer medium with the test section held at 1.5 bar. An uncertainty analysis indicated that the heat flux uncertainty varied between 20% and 52% while the heat transfer coefficient uncertainty varied between 25% and 52%. Overall the results indicate the Zircaloy® specimen is superior to the silicon carbide specimen. The results differ from previous research indicating that a higher roughness and thermal conductivity do not necessarily lead to a greater heat transfer coefficient. When the results are placed in specific boiling equations from previous research it is found that the silicon carbide is closely predicted by the Shah and Kandlikar correlations while the Zircaloy® is more closely predicted using the Gungor and Winterton correlation. These correlations were developed for internal flow though and were simply extended to external flow boiling for predictive purposes. In conclusion it can be stated that the Zircaloy® cladding material outperforms the silicon carbide test specimen evaluated in this study. It remains however a promising replacement. To fully evaluate silicon carbide’s readiness for nuclear applications many unknowns still need to be answered though. For future research the power supply needs to be upgraded to provide a constant input power as well as a new data logging system to provide a constant readout of data. The inlet water temperature is crucial and needs to remain constant for all sets of tests. The test section should also be made smaller for single rod experiments and the heater power needs to be greatly increased to higher surface temperatures and possibly film boiling.
Supervisor: Prof J. Slabber
CoSupervisor: Prof J.P. Meyer
H. Ghodsinezhad, 2016 "Experimental Investigation on Natural Convection of Al2O3water Nanofluids in Cavity Flow"
The thermophysical properties of nanofluids have attracted the attention of researchers to a far greater extent than the heat transfer characteristics of nanofluids have. Contradictory results on the thermalfluid behaviour of nanofluids have been numerically and experimentally reported on in the open literature. Natural convection has not been investigated experimentally as much as the other properties of nanofluids. In this study, the characteristics and stability of Al_{2}O_{3}water nanofluids
(d = 20–30 nm) were analysed using a Malvern zetasizer, zeta potential and UVvisible spectroscopy. The natural convection of Al_{2}O_{3} water nanofluids (formulated with a singlestep method) was experimentally studied in detail for the volume fractions 0, 0.05, 0.1, 0.2, 0.4 and 0.6% in a rectangular cavity with an aspect ratio of 1, heated differentially on two opposite vertical walls for the Rayleigh number (Ra) range 3.49 x 10^{8} to 1.05 x 10^{9}. The viscosity of Al_{2}O_{3}water nanofluids measured between 15 and 50 °C. The effect of temperature and volume fraction on viscosity was also investigated. A detailed study of the nanoparticle concentration effect on the natural convection heat transfer coefficient was performed. It was found that increasing the concentration of nanoparticles improves the heat transfer coefficient by up to 15% at a 0.1% volume fraction. Further increasing the concentration of nanoparticles causes the natural convection heat transfer coefficient to deteriorate. This research also supports the idea that “for nanofluids with thermal conductivity – more than the base fluids – an optimum concentration may exist that maximises heat transfer in an exact condition as natural convection, laminar force convection or turbulence force convection”.
Supervisor: Dr, M Sharifpur
CoSupervisor: Prof J.Meyer
J Otto, 2016 "Nuclear Fusion of Li6 H2 Crystals"
The inception of this study lies in mankind’s need for energy in order to drive commerce, industry and development. Although several sources have been exploited in the production of energy, it is the personal opinion of the author that the most novel of these is atomic energy. It is known that humanity has become quite proficient in using nuclear fission, as is evidenced in the world’s nuclear fleet; however the use of fusion for controlled energy production has not yet developed sufficiently for commercial use. It is, however, a popular anecdote that the production of energy from nuclear fusion is only forty years away – and always will be. During the literature survey, several forms of fusion producing processes are considered ranging from stellar bodies and magnetic confinement based reactors to percussive wave driven fusion as is used in fissionfusion warheads. This study focusses on the prospect of fusion at room temperature. The aim of this study is to further investigate a specific fusion process in order to determine if this may become a viable source for the production of energy. In the literature survey it is determined that the atoms in certain moleculessuch as Li6 H2 and H2O can spontaneously overcome the Coulomb barrier and fuse to form new elements; releasing energy in the process. The mechanism for these two atoms are perceived to be similar, and the Li6 H2 reaction was selected based on the ease of observability, of the two αparticles that would form , should a fusion reaction take place. Furthermore, it is theorised in the literature survey that the reaction rate of the Li6 with H2 can be increased by exciting the molecules with Xray radiation. In this study, an experiment is devised and conducted in order to determine if the reaction rate can indeed be accelerated as proposed. The results are compared to studies where the molecules were not excited in this manner as a baseline, and an additional determination lies in the practicality of energy production for commercial use in this manner. In the experiment, an Xray source is used to bombard several samples of L6 H2 with wide spectrum Bremsstrahlung for several hours. Mathematical approximations, as well as simulations, are used in order to determine the energy deposition of the Xrays into the samples. This is done in order to determine if the process has viability as an energy production source, and in an attempt to determine if there is a specific wavelength that the process is partial to. In order to record possible fusion reactions; the samples are layered with several plastic detectors. These detectors are chemically etched and studied using microscopy, the appendix deals with experiments aimed at calibrating the process used to study these detectors. It is successfully shown that the reaction rate is increased by introducing Xrays to the Li6 H2 powder, however, the magnitude is far lower than was hoped for. Additionally, due to the low number of fusion reactions that took place, enough data is not available to determine an electromagnetic wavelength that is of particular interest. Finally, it is thus shown that the configuration used in this study is not a viable assembly for the production of power.
Supervisor: Prof J. Slabber
CoSupervisor: Prof J.P. Meyer
Saboura Yousefiaboksari, 2016 "Experimental investigation and theoretical analysis on the effects of nanolayer on nanofluids’ thermophysical properties"
Nanofluids, which are suspensions of nanoparticles in conventional heat transfer fluids, attracted research studies on different heat transfer applications, while they enhance thermal transport properties in comparison with conventional base fluids.
Recently, the use of these new fluids has been growing increasingly. However, the ambiguities of their thermophysical properties cause them to function inefficiently in industrial design. The recognised important parameters that affect the properties of nanofluids include the volume fraction of the nanoparticles, temperature, nanoparticle size, nanolayer, thermal conductivity of the base fluid, pH of the nanofluid and the thermal conductivity of the nanoparticles. However, there is a distinct lack of investigation and reported research on the nanolayer and its properties.
In this study, the effect of uncertainty of the nanolayer properties on the effective thermal conductivity and viscosity of nanofluids, and heat transfer are discussed in detail. The results show that the uncertainties can cause 20% error in the calculation of the Nusselt number and 24% for the Reynolds number. Therefore, more research needs to be conducted on nanolayer properties in order to identify them accurately.
The density of some nanofluids, such as SiO2water, SiOxEGwater, CuOglycerol and MgOglycerol, has also been investigated experimentally. Therefore, the effects of nanolayer thickness and density on nanofluid properties are discussed in detail. The results show that nanolayer density and thickness have a significant effect on nanofluid density, and nanolayer density is found to be between void and base fluid density.
Consequently, by analysing experimental results and performing a theoretical analysis, a model has been derived to calculate the density of nanofluids.
Specific heat capacity is the other nanofluid property that is discussed in this study. Experimental data from literature, available formulae and the presented model for nanofluid density have been used to identify nanofluidspecific heat capacity, while nanofluid density is one of the parameters in calculating specific heat capacity. This investigation was performed using a model – used by different authors – that also considers the nanolayer. The specific heat capacity of nanofluids that resulted from two methods of calculation has been compared with available experimental data. This investigation shows that the proposed model for the density of nanofluids provides better agreement for specific heat capacity in comparison to experimental data.
Supervisor: Dr Mohsen Sharifpur
CoSupervisor Prof Josua P Meyer
A. Jami, 2016. "Impeller Fault Detection under Fluctuating Flow Conditions using Artificial Neural Networks"
Maintenance of equipment at the required condition to ensure a reliable performance, as well as improvement of safety, are major concerns in the field of asset integrity management. Condition monitoring is a procedure that allows one to identify early signs of failures and implement efficient maintenance plans to eliminate the uncertainties in machine operation. In addition, vibration monitoring is known as a detection tool for early detection of degradation from the expected performance. It is often superior to other condition monitoring techniques, due to its high sensitivity and simplicity of implementation. Vibration analysis provides substantial information regarding the operating condition of components and aids to remedy problems. Therefore, it can be used to detect a wide range of fault conditions in rotating machinery, such as imbalance, misalignment of internal shafts, looseness, cracked shaft, gear failures, rolling element bearing damages, motor faults and impeller issues.
The primary intention of the research reported in this dissertation is to investigate the applicability of a neural network methodology for the detection and diagnosis of mechanical defects of impellers in centrifugal pumps. The study focuses on extracting appropriate features from vibration signals associated with pump impellers and the performance of artificial neural networks (ANNs) using these features. The second intention is to enhance maintenance decisions regarding the actual impeller condition. This leads to a transition from time based preventive maintenance to condition based maintenance, and also improving the safety and reliability of pumping systems, as well as reducing unexpected and catastrophic failures. Hence, vibration analysis techniques are used as a principal tool to characterise the impeller conditions under flow variation, with the requirements of data collection, data processing, transformation and selection of essential features corresponding to the running condition.
This dissertation presents a study of current vibration analysis techniques to extract the required features, namely time based features, frequency based features and wavelet based features. An experimental setup is developed to measure the impeller vibration. The experiment is performed using seven impeller fault conditions such as crack and imbalance under fluctuating
flow conditions to simulate nonstationary conditions in the system. Also, the evolution of features over varying flow rates are evaluated in order to identify features that contain fundamental information corresponding the fault characteristics. Moreover, the collected features form nondimensional training data sets are used to train ANNs. Comparisons of different training algorithms, network hidden nodes and effectiveness of different transfer functions are performed to select the most appropriate parameters of networks.
Validation of the results prove that the accuracy of ANN prediction improves considerably by using decomposed vibration signals and energy based features. Comparison of the network accuracy based on wavelet packet transform (WPT) features with time analysis and frequency analysis based features, indicate that WPTANN lead to lower mean square errors and higher correlation coefficients, as well as shorter training times. The WPTANN model can save computational time and provides better diagnostic information, which can be effectively used for classification of impeller defects under nonstationary conditions.
Supervisor:Prof Stephan Heyns
DA Ramatlo, 2016."Optimal Design of a Guided Wave Rail Web Transducer using Numerical Modelling"
Ultrasonic Guided Waves can propagate over long distances, and are thus suitable
for the interrogation of long structural members such as rails. A recently developed
Ultrasonic Broken Rail Detection (UBRD) system for monitoring continuously welded
train rail tracks, primarily detects complete breaks. This system uses a guided wave
mode with energy concentrated in the head of the rail, which propagates large distances and which is suitable for detecting defects in the rail head. Exploiting a second mode, with energy concentrated in the web section, would allow us to e_ectively detect defects in the web of the rail.
The objective of this study is to develop an ultrasonic piezoelectric transducer that can excite a guided wave mode with energy concentrated in the web of the rail. It is required that the transducer must strongly excite such a mode at the operational frequency of the UBRD system. The objective is thus to obtain a design with optimal performance.
A recently developed numerical modelling technique is used to model the interaction
of the transducer with the rail structure. The technique employs a 2D SemiAnalytical
Finite Element (SAFE) mesh of the rail crosssection and a 3D _nite element mesh of the transducer; and is thus referred to as SAFE3D. The accuracy of the SAFE3D method was validated though experimental measurements performed on a previously developed transducer.
A design objective function representative of the energy transmitted by the transducer to the web mode was selected. The identi_ed design variables were the dimensions of the transducer components. The performance of the transducer was optimized using a response surfacebased optimization approach with a Latin Hypercube sampled design of experiments (DoE) that required SAFE3D analyses at the sampled points. A Nelder Mead optimization algorithm was then used to an optimal transducer design on the response surface.
The performance of the optimal transducer predicted by the response surface was found to be in good agreement with that computed from SAFE3D. The optimum transducer was manufactured and experimental measurements veri_ed that the transducer model was exceptionally good. The design method adopted in this study could be used to automate the design of transducers for other sections of the rail or other frequencies of operation.
KEYWORDS: Ultrasonic guided wave; Piezoelectric transducer; SAFE3D; Optimization
Supervisor: Dr. D.N. Wilke
Cosupervisor: Dr. P.W Loveday
Cosupervisor: Dr. C.S Long
KyoungYeoll (John) Lee, 2016 "Cavity natural convection of zinc oxidewater nanofluid experimental work"
Nanofluids are recognized to have great potential for conventional heat transfer fluids that could benefit industries. However, many indepth numerical studies and fewer experimental studies have been conducted on natural convection of nanofluid and their results are inconsistent. In this study, natural convection heat transfer characteristics of zinc oxide (ZnO)water nanofluid is investigated in rectangular enclosure through cavity flow experimental measurements. The ZnOwater nanofluids are prepared with different volume fractions of 0.09, 0.18, 0.36, 0.5 and 1 (vol.%) (0.5, 1, 2, 3 and 5.67 weight percentage) and for Rayleigh number (Ra) varies from 7.9E+7 to 8.9E+8. The stability of the ZnO nanofluid is verified using a spectrophotometer and zeta potential measurement at various temperatures and concentrations. Zeta potential values are measured within the stable range, and no sedimentation of nanoparticles is indicated within 24 hours. The viscosity of ZnOwater nanofluid is also measured experimentally, which is 20% higher than the use of the traditional Einstein viscosity model at 1 vol.%. Consequently, the suspension of ZnO nanoparticles in water does not enhance the natural convection heat transfer coefficient. The average Nu increases as the Ra increases, but the average Nu decreases as the volume fraction of the nanofluid increases. The systematic deterioration of the natural convection heat transfer coefficient is observed as increasing in the concentration of nanoparticles.
Supervisor: Dr M Sharifpur
Cosupervisor: Prof JP Meyer
C. Grobler, 2016 "MultiObjective Parallelization of Efficient Global Optimization"
Design optimization is a subject field where mathematical algorithms are used to improve designs. Analyses of designs using computational techniques often require significant com puting resources, and for these problems, an efficient optimization method is needed. Efficient Global Optimization (EGO), first proposed by Jones et al. [2] is an optimization method which aims to use few function evaluations when optimizing a design problem. In this study, we use a multiobjective strategy to parallelize EGO.
EGO is part of a set of algorithms called surrogate optimization methods. A set of initial designs are analyzed and then a response surface is fitted to the evaluated designs. In each iteration, EGO selects the set of design variables for which the next analysis will be performed. It makes this decision based on two opposing criteria. EGO will either decide to sample where the predicted objective function value is low, an exploitation approach, or where there is high uncertainty, an exploration approach.
In each iteration, the classical EGO only selects one design per iteration. This selected design vector is either a result of exploitation or exploration based on a measure referred to as maximum Expected Improvement (EI). However, the modern day computing envi ronment is capable of running multiple different analyses in parallel. Thus, it would be advantageous if EGO would be able to select multiple designs to evaluate in each iteration.
In this research, we treat EGO’s inherent selection criteria to either exploit or explore as
a multiobjective optimization problem, since each criterion can be defined by a separate objective function. In general multiobjective optimization problems don’t only have one solution, but a set of solutions called a Pareto optimal set. In our proposed strategy multiple designs from this Pareto optimal set are selected by EGO to be analyzed in the subsequent iteration. This proposed strategy is referred to as Simple Intuitive Multi objective ParalLElization of Efficient Global Optimization (SIMPLEEGO).
We start our study by investigating the behaviour of classical EGO. During each iteration of EGO, a new design is selected to be evaluated. This is performed by finding the maximum of the Expected Improvement (EI) function. Maximizing this function initially proved challenging. However, by exploiting information regarding the nature of the EI function, the maximization problem is simplified significantly, and the robustness of finding the maximum is enhanced. More importantly, solving this maximization problem robustly, dramatically improves the convergence behaviour once a local basin has been found.
We compare our SIMPLEEGO method to a multiobjective optimization algorithm (EGO MO) published by Feng et al. [1]. We first investigate the behaviour of EGO, EGOMO, and SIMPLEEGO. Thereafter the convergence performance of these methods is quanti fied.
As expected the parallelization of both SIMPLEEGO and EGOMO lead to faster con vergence on a range of test functions compared to classical EGO, which only sampled one point per iteration. The convergence characteristics of SIMPLEEGO and EGOMO are also markedly different. We conclude with a discussion on the advantages and disadvan tages of the investigated methods.
References
[1] Z. Feng, Q. Zhang, Q. Zhang, Q. Tang, T. Yang, and Y. Ma. A multiobjective opti mization based framework to balance the global exploration and local exploitation in expensive optimization. Journal of Global Optimization, 61:677–694, 2015.
[2] D. R. Jones, M. Schonlau, and W. J Welch. Efficient global optimization of expensive blackbox functions. Journal of Global Optimization, 13:455–492, 1998.
Supervisor:Prof. S. Kok
CoSupervisor: Dr D. N. Wilke
H. J. Theron, 2016. "Modelling and characterization of a modified 3DoF pneumatic GoughStewart platform"
Stabilised line of sight optical payloads for maritime vessels require variable platform conditions during the development, test and evaluation phases. A ship deck motion simulator is one means of generating such conditions in a controlled laboratory environment. This dissertation describes the aspects of the modelling, identification and validation of a ship motion simulator, in the form of a pneumatically actuated 3DOF modified GoughStewart manipulator, to generate a realistic simulation environment for controller design. The simulation environment is a Matlab supervised MSC ADAMS/Matlab Simulink cosimulation in which Simulink houses the pneumatic model, the friction model, and the controller, and ADAMS runs the dynamic model of the physical hardware. A similar simulator cannot be found in published literature forcing a development of the model from the ground up, using published information as a foundation. The simulator model is broken up at the subsystem level which comprises the valve mass flow model, the piston chamber and force model, the complete actuator model and finally the complete ship simulator model. Each of these is derived, identified, and validated. The requirements of the simulator as well as the simulation environment is derived from reallife measurements done on seafaring vessels. An inverse kinematic solution is presented as a set of lookup tables which are generated from the outputs of MSC ADAMS by manipulating the simulator platform over the whole range of movements through Matlab. The reverse of the process is then used to ensure that actuator extensions generate the correct platform attitude — the attitude errors as shown to be infinitely small. Two valve mass flow models are proposed, a classical model and an ISO model, the first derived from thermodynamic principles and the second based on the ISO6358 standard. The parameters of the two models are identified through experimental charging and discharging of a constant volume pressure chamber and sampling the temporal pressure and temperature outputs. The mass flow is calculated from the measured data through parameter estimation. Validation is done by comparing the temporal pressure outputs of the models with the actual measured pressure signals. The mean absolute error for the best fit ISO model is less than half of the Classic model at 0.4 MPa (MAE < 2 kPa) and the temporal pressure relationships in the closedloop and openloop tests shows a 93% correlation against measured pressure signals. The combination of the derived actuator chamber model and the valve mass flow model produces a realistic actuator model. The force equation of each of the actuators makes provision for a nonlinear friction component. The actuator friction model is based on a simple stickslip relation with an acceleration dependent Stribeck function and an exponential viscous friction component. This model is also identified with data from the actual hardware. The complete ship motion simulator model is validated through openloop as well as closedloop tests. The openloop tests are performed with chirp or sinusoidal signal excitation from a stable elevated offset starting condition. The ratio of the measured and simulated extension amplitudes in the openloop is larger than 0.95 while the ratio of the rise times (t_{m}/t_{s}) is approximately 0.85. The closedloop validation tests are conducted with both heave and roll inputs and compared well with the real system. A 14% difference in the actuator position amplitude (between the simulated and measured systems), and a 20% slower extension rate at 0.05 Hz that increases at 1 Hz to match the measured rate are observed. The maximum large signal bandwidth is 0.617 Hz, and is only limited by the mass flow. A simplified plant model is derived and compared with the high performance model and is subsequently used for a state feedback controller design and evaluation. The final controller gains deliver a stable system with the same 0.617 Hz bandwidth limitation and a controller that is insensitive to loop gain changes from 0.5 to 15.
Supervisor: Prof. N. J. Theron
G Stephens ,2016. "Characterisation of Filling Stage Models for Vacuum Infusion"
The development and proposal of models to simulate the vacuum assisted infusion processes have received significant attention over the last decade. In this study four permeability and four compaction models are considered in a standard resin infusion simulation. This results in 16 models that follow from the combination of the four permeability and four compaction models. All 16 models are characterised in this study, with the aim to investigate the extent to which the models are able to represent an experimentally measured pressure time response for the filling stage. This study therefore investigates the innate ability of the models to represent the experimental response. Each model is characterised using an exhaustive robust inverse analysis approach.
The need for the proposed robust inverse strategy was in response to the multimodal nature of the inverse problem. Brute force inverse approaches were found to be too computationally expensive. The robust strategy calculates the model response for a Latin hypercube sampled set of parameters in the domain. A partial least squares regression then uses the parameters and response sets to generate an initial starting point for an interior point optimisation algorithm to solve the conventional least squares fit. This method was first shown to give reliable results on two example problems before using it on the vacuum infusion problem.
Using this robust strategy it was found that all 16 models have the same ability to represent the experimentally measured response. The response corresponds well to the simulated response of a more sophisticated resin infusion simulation that predicts the response from experimentally estimated parameters.
This study will hopefully stimulate future researchers to also characterise newly proposed models to investigate the innate ability to represent a desired response in addition to the predictive capability using physically justifiable parameters.
supervisor: Dr DN Wilke
M.A. Rahiman, 2016. "Title: An Exploration of the Relationship between Maintenance Performance and Resource Productivity"
As a consequence of their relative magnitude with respect to overall organisational expenditure, potential sources for significant cost savings involve maintenance costs, raw material costs and energy consumption. Previously conducted but inconclusive research indicates that there may be a relationship between maintenance activities and resource productivity. If this is the case, knowledge of such a relationship may unveil opportunities for direct productivity enhancement. Moreover, it may also serve as an aid in making improved measurements of the true value of the maintenance function. This in turn may enable practitioners to recognise when resource reallocation may be required to achieve greater levels of productivity.
The objective of this research is to explore the relationship between maintenance activities and resource productivity. It aims in part to assess if opportunities for productivity enhancement exist as a result of such a relationship. It also aims to establish if resource productivity can serve as a representative measure of maintenance performance. This study is based on rigorously proven theoretical propositions which are tested empirically on data procured from a metallurgical plant in South Africa.
The conclusion of this study is that the maintenance function enables equipment to process resources productively. Resource productivity may thus have the propensity to serve as an encompassing and cost effective measure of maintenance performance. In terms of its potential in this regard, decreases in resource productivity may offer valuable signals which indicate that corrective action is warranted.
In terms of productivity enhancement, this study elucidates the fact that machinery should always be kept in the best operating condition possible. When machinery malfunctions are discovered, it should be repaired in a timely manner to prevent unnecessary wastage from occurring.
Supervisor: Prof JL Coetzee.
HansRudolf Björn Bosch, 2016. "FTire model parameterization and validation of an allterrain SUV tyre"
Tyre modelling has been a focal point of vehicle dynamics modelling since the beginning of vehicle dynamics research. Many tyre models are based on single point contact models which utilize some form of the Pacejka Magic Formula curve fit. The Pacejka Magic Formula approach was formulated in the 1980s and has certain advantages such as high computational efficiency and easily obtainable parameterization data. However, the Pacejka Magic Formula is limited to function on smooth roads and a finite number of well defined, long wavelength discrete obstacles.
A high fidelity approach in the form of Cosin’s FTire tyre model was developed, in which the tyre is modelled as a three dimensional object populated with bending, tangential, lateral and radial stiffnesses as well as damping. The tyre is numerically approximated with a predetermined number of elements. The disadvantages of using FTire include its low computational efficiency and the large number of parameters prescribed to parameterize the tyre model. However, FTire is claimed to be capable of accurately predicting the forces and moments generated by the tyre on smooth as well as uneven road surfaces for onroad tyres.
The focus of this study lies on parameterizing and validating an FTire model of an allterrain SUV tyre. The aim is to verify whether a parameterized FTire model is able to predict the tyre behaviour of an allterrain SUV tyre for lateral and longitudinal forces on smooth road surfaces and vertical forces on uneven but hard terrain.
Static laboratory and dynamic field tests are conducted to acquire parameterization and validation test data to parameterize the FTire model. An Adams model of the tyre testing equipment is implemented to simulate the FTire model and validate it against dynamic validation test results.
It is found that the FTire model is able to predict the lateral tyre behaviour well on a smooth road surface. The longitudinal tyre behaviour on a smooth road surface and vertical tyre behaviour on an uneven road surface are predicted very well by the parameterized FTire model.
Supervisor: Prof P. S. Els
Marc Robert Greenland , 2016. "Analysis of Conjugate Heat Transfer and Pressure Drop in Microchannels for Different "
In this study the heat transfer and hydrodynamic parameters were experimentally investigated for a single microchannel housed in a stainless steel solid base material for different aspect ratios in the laminar regime with water as the working fluid. The stainless steel base material had a low thermal conductivity (15.1 W / mK) which magnified the conjugative effects in order to better understand the heat transfer. Rectangular microchannels with a height and width of 0.64 mm x 0.41 mm for Test Section 1, 0.5 mm x 0.5 mm for Test Section 2 and 0.43 mm x 0.58 mm for Test Section 3 were considered. The overall width of the solid substrate was 1.5 mm and the length was 50 mm for all of the test sections. The aspect ratio of the channel and the solid substrate was kept equal. A constant heat flux of 10 W / cm^{2} was applied to the bottom outer wall of the test section. A sudden contraction inlet and a sudden expansion outlet manifold contained pressure ports, to measure the pressure drop across the test sections, and thermocouples measured the mean inlet and outlet fluid temperatures. Thermocouples were used to measure the outer top and side wall temperatures at four equally spaced positions along the axial direction. The amount of axial heat conduction was below 0.6 % for all of the test sections and therefore warranted the use of a twodimensional conduction model to determine the heat transfer parameters at the fluid to solid interface based on the outer measured wall temperatures. The local Nusselt number decreased, along the axial direction but increased towards the exit for all of the test sections. The average Nusselt number increased with the flow rate and the critical Reynolds number for fully turbulent flow Test Section 1 was 1950, for Test Section 2 was 2250 and for Test Section 3 was 1650. The average Nusselt number was directly related to the perimeter of the microchannels’ two side walls and the bottom wall (not the top wall), and thus decreased as the aspect ratio of the channel increased. The experimentally determined Nusselt numbers were larger for all three test sections when compared to common acceptable correlations. The friction factor decreased with the flow rate and was smaller in magnitude when compared to conventional theories. The diabatic friction factor magnitudes were smaller than the adiabatic friction factors. The friction factor decreased as the aspect ratio decreased, where the aspect ratio was calculated by taking the maximum of the microchannels width or height, divided by the minimum of the two. The possibility of a relationship could exist between the Colburn jfactor and the friction factor when considering the results for Test Section 1 and Test Section 2 but the results for Test Section 3 were significantly different.
Keywords: microchannel, heat transfer, pressure drop, single phase, laminar, stainless steel, water
Supervisors: Dr J. Dirker and Prof J.P. Meyer
Thomas Christian Montgomery, 2016. "Optimal distribution of discrete heat sources in a twodimensional
data centre"
In the study, the optimal distribution of discrete heat sources in a twodimensional data centre was
investigated. The optimal placement of the cool supply air inlet and outlet was also investigated.
The governing equations were solved by using the finite volume method. The computational fluid
dynamics code Fluent was used to solve the governing equations. Optimisation was achieved using a
goaldriven optimisation approach and a response surface methodology.
The numerical model was validated using past experimental work and the results were in good
agreement with each other, showing an error of less than 6%. The realisable kε turbulence model
was used as closure equations to solve the Reynoldsaveraged NavierStokes equations. Additionally,
the viscosity affected nearwall regions were modelled using a wall treatment method.
The optimum distribution of constant height (42 U) server racks was established for three different
configurations of inlet and outlet locations. After these optimal placements were established, the
effect of varying the height of the server racks was investigated for the same inlet and outlet
placements and the optimum locations were determined. By means of a sensitivity analysis, it was
found that the placement of the first and last servers as well as their respective heights had the most
influence on the heat transfer between the server panels and the ambient surroundings.
It was concluded that the inlet and outlet should be placed on opposing walls of the data centre and
variable server rack heights should be used in order to achieve maximum heat transfer.
Keywords: twodimensional, discrete heat source, optimisation, response surface, data centre
Supervisors: Prof T. BelloOchende and Prof J.P. Meyer
E Miles, 2016. "Optimal Control Surface Mixing of a RhomboidWing UAV"
This thesis describes the development of an openloop control allocation function – also
known as a ‘mixing function’ – for aircraft with an unconventional control surface setup (i.e.
not consisting of a conventional elevator, rudder and ailerons) by using mathematical
optimisation. The techniques used to design the control allocation and mixing used on the
unconventional configuration when flying it without artificial stability or control
augmentation is provided. A typical application of this control mixing would be to enable a
pilot to operate an unconventional unmanned aerial vehicle (UAV) as if it was a conventional
model aircraft during flight testing or as a backup mode should any sensor failures occur
during a typical flight test program. The allocation can also be used to simplify the inner
control structure of a UAV autopilot or stability augmentation system. Although this type of
mixing would be straightforward on a conventional airframe, an unconventional
configuration has several unique characteristics that complicate the modelling and design
process.
A custom six degree of freedom (6DOF) formulation for flight simulation was made
available to model the aircraft and run various scripts to evaluate the aircraft response when
the control allocation function is implemented. The simulation model was used to develop the
mixing function that maps conventional input commands to the unconventionally situated
control surfaces in the most optimal way.
The design process was formulated as a multiobjective optimisation problem, which was
solved using a custom sequential quadratic programming and custom leapfrog programming
method. A methodology was proposed to define the constraints, which can be customised for
a particular aircraft or application.
The control allocation function was implemented in two different simulation environments to
investigate the suitability of candidate designs. A robustness study was performed to evaluate
the impact of actuator failures on the aircraft control response using the designed control
allocation system. The proposed control allocation design methodology can also be used to
design the inner control loops of more sophisticated control systems such as stability
augmentation and automatic flight control, which is also briefly discussed in this thesis.
Supervisor: Dr B.A. Broughton
Cosupervisor: Prof J.P. Meyer
Alan Glenn Guthrie, 2016. "3D COMPUTER VISION CONTACT PATCH MEASUREMENTS INSIDE
OFFROAD VEHICLE TYRES"
The interaction between tyres and terrain is one of the most studied areas in the vehicle
dynamics and terramechanics research communities because it is the only region where excitation
forces act on the vehicle if aerodynamics is not considered. The tyre area which
deforms against the ground is called the contact patch. Measuring the contact patch has
been accomplished statically in the past; however, measurement while a wheel rotates has
proven difficult.
A number of attempts to measure carcass deformation from inside the tyre have succeeded in
measuring small areas or single points but full field measurement has never been attempted.
This study describes the design and testing of a system which uses stereo cameras and
image correspondence to measure the deformation of the inside of the tyre carcass in the
contact region completely. The system includes a stabilisation mechanism which prevents
the cameras from rotating, ensuring that the cameras view the inside of the contact region
at all times.
Software to capture and process the images captured is developed and tested to ensure measurement
accuracy. The 3D results produced by the software are compared to one another
where possible and any trends or problems are discussed. Results indicate that the full 3D
displacement field in the contact region can be measured accurately. The information produced
is expected to be extremely valuable for development and validation of tyre and vehicle
dynamics models.
Supervisor: Prof P. S. Els
Ibrahim Garbadeen, 2016. "Natural convection of multiwalled carbon nanotubes with water mixtures in a square enclosure".
The use of nanofluids in buoyancydriven heat transfer can be very useful in enhancing the performance of various heat transfer applications. In this thesis, natural convection by multiwalledcarbon nanotubes (MWCNT) was studied in a square enclosure with differential heating by two opposite walls. Low particle concentrations of 0 – 1% based on volume were considered at Rayleigh numbers of 10^{4} – 10^{8}. Thermal conductivities and viscosities of the nanofluids were experimentally determined. It was found that thermal conductivity and viscosity increased with increasing concentration by 6% and 58%, respectively. Models based on these experimental results were obtained and subsequently used in a numerical study of a twodimensional simulation of natural convection in a square cavity using a commercial code. Results revealed an initial enhancement in the Nusselt numbers to a maximum of 22% which occurs at 0.14 % particle concentration and a Rayleigh number of 10^{8}. Beyond the maximum, the Nusselt number deteriorated. This was true for the different Rayleigh numbers studied with percentage enhancement in the Nusselt number increasing with increasing Rayleigh numbers. Further analysis was done to predict heat transfer performance of higher particle concentrations up to 8% which showed a general decline in the Nusselt numbers by increasing particle concentration. An experimental setup was subsequently used to study natural convection in an insulated square cavity with different temperature differences between the two opposite sides for particle concentrations of 0–1% at Rayleigh numbers between 2.1x10^{8} and 6x10^{8}. Results from the experimental and numerical studies were subsequently compared and the validity of projected results for higher particle concentration was therefore assessed. The experimental results supported the overall behaviours of the nanofluids obtained from the numerical analysis. However, the experimental results of maximum enhancement in the Nusselt number was 42% at particle concentration 0.1% and a Rayleigh number of 6x10^{8}. Nevertheless, both results indicated the existence of an optimum particle concentration at which heat transfer in MWCNT nanofluids is maximised. The variation in the performance nanofluid was attributed to the counteracting, nonlinear effects of thermal conductivity and viscosity both of which increases by increasing particle concentration. The thermal conductivity effect which improves heat transfer performance was observed to be more dominant for a very narrow range of low particle concentration up to 0.1 % while the viscous effect which diminishes heat transfer performance was found to be more dominant at higher particle concentration.
Keywords: Nanofluids, MWCNT, natural convection, cavity flow, volume fraction, viscosity.
Supervisors

:

Dr Mohsen Sharifpur; Prof Johan Slabber; Prof Josua Meyer

Pieter C. B. Luyt, 2016. "A leak tight design methodology for large diameter flanges based on nonlinear
modelling and analysis"
There is currently a need for large diameter flanges for the supply of water in South Africa. These large
diameter pipe flanges are required to accommodate pipes with nominal bores of up to 4 m and should
successfully withstand internal pressures of up to 8 MPa. No current relevant standard / code contains
prescribed design values for flanges which either operate at such high pressures or have such large
diameters. Due to this an alternative method of design, by means of nonlinear finite element modelling,
is proposed. Three types of integral flange designs are considered, namely: flat face, raised face, and
a modified raised face with an Oring groove. The effects of creeprelaxation, flange rotation, and the
bolting sequence are considered.
For each of these designs a finite element model was created and compared to a small scale experiment
which included strain and contact pressure measurements. The proposed nonlinear finite element
models were capable of accurately predicting the strains in the flanges as well as the contact pressures
between the faces of the flange and the surfaces of the packing material. Finally, a comparison between
the ASME design method and the proposed nonlinear finite element modelling design method was
done for the large diameter flanges. It was found that the ASME design code did not have the ability to
accurately predict the stresses in the flanges. It was also found that by using the maximum equivalent
Von Mises stress as failure criteria for the flanges and fasteners, and contact pressure for the sealing
ability, circular bolted flange connections which are lighter, safer, and leak tight could be designed by
means of the proposed nonlinear finite element models.
Keywords: flat face flange; raised face flange; raised face flange with an Oring groove; flange rotation;
creeprelaxation; contact pressure
Supervisor: Prof. Nico J. Theron
Cosupervisor: Mr Francesco Pietra
C.B. Church, 2016. 'Turbomachine Internal Pressure and Blade Response Modelling"
Blades are critical components of turbomachines, failure of a single blade may result in catastrophic
failure of the entire machine. One study found that blade failure was the third largest cause of power
generation unit unavailability. Their condition during operation is therefore of interest to monitor.
Various intrusive and nonintrusive blade vibration measurement (BVM) techniques have been
developed for this purpose. Intrusive techniques such as strain gauge approaches and the frequency
modulated grid method require expensive and complex alteration of the actual blades and/or casing.
Further, they are prone to failure due to operation in harsh working environments. Therefore the use
of intrusive techniques has been predominantly limited to design verification, testing and research.
Blade tip timing approaches are currently at the forefront of BVM. The practicality, accuracy and ease
of implementation of these approaches have limited their commercial roll out. An alternative nonintrusive
source of blade vibration information was found in the internal casing pressure signal (CPS).
As the machine operates the blade movement excites the fluid in the casing, producing a measureable
response. Unlike BTT approaches which deal with a scarcity of information, CPS based methods must
identify blade vibration from a complex signal which contains multiple other sources of information.
The issue of how to model the blades’ response and fluid interaction is the topic of this investigation.
An available single stage turbomachine mock setup was modified for internal pressure and direct
blade vibration measurements. Pressure measurements were taken in line with a redesigned hub and
rotor blade assembly. Strain gauges (SG) were applied to blades in order to capture their response.
The blades’ response was modelled as the combination of a forcing function and a multiple degree of
freedom transfer function. Repurposed experimental modal analysis frequency response
reconstruction techniques were used to model the blades’ transfer function. It was found that this
technique was able to capture the blades’ underlying behaviour to a high degree. The forcing function
was modelled in the time domain as a series of Gaussian shaped force distributions. It was found that
the model was able to capture many important aspects of the forcing behaviour. Both the forcing
function and blade transfer function were explored using constrained optimisation techniques.
The bladefluid interaction was modelled as a Fourier series. It was shown that the blade behaviour
cannot be extracted from a pressure signal using standard frequency analysis techniques. The viability
of an inverse problem solution methodology, for the purpose of blade behaviour extraction, was
investigated. This was achieved by solving reduced components of the model with SG measurements
and observations from pressure measurements. Further the need to isolate the pressure field about
individual blades was motivated and a novel time domain windowing technique provided.
Keywords: Turbomachine, blade vibration, casing pressure, signal processing, optimisation.
Supervisor: Prof. P.S. Heyns
MulockHouwer, 2016. "The effect of adjacent tubes on the diabatic friction factors in the transitional flow regime"
Heat exchangers are used throughout the world in important processes such as the generation of electrical energy. Modern heat exchangers are often forced to operate in the transitional flow regime, where flow can be unpredictable. Most of the research that has been done on the transitional flow regime has focussed on the influence of heat transfer and the inlet effects. However, all these studies made use of only a single tube, while most heat exchangers would typically have a bundle of tubes such as in shellandtube type heat exchangers. The purpose of this study was to investigate the effect of adjacent tubes on the transitional flow regime during diabatic conditions. An experimental setup was purposefully built for this investigation and two test sections were investigated. A singletube test section was built for validation purposes, since similar work has been done. A tripletube test section was built with three tubes spaced at a pitch distance of 1.4 outer diameters. The mass flow rate, as well as the pressure drops over the fullydeveloped section was measured for each tube. From the pressure drop data the friction factors were calculated. Furthermore, a heat flux of 3 kW/m^{2} was applied to each tube and the inlet, outlet and wall temperatures were measured, to ensure that specifically the diabatic friction factors were determined. Water was used as the working fluid and tests were run over a Reynolds number range of 1 000  6 500. An uncertainty analysis showed the maximum uncertainty of the friction factors to be 8.3%. The laminar, transitional and turbulent flow regimes could be identified from the friction factor data. The results from the singletube test section correlated well to the literature with transition starting at a Reynolds number of 2 380 and ending at 3 050. The results from the tripletube test section showed the start of transition to be initiated by the presence of adjacent tubes, with the Centretube entering transition at a Reynolds number of 1 970. The outer tubes experience a delayed start in transition at Reynolds numbers of 3 000 and 2 800 for the Lefttube and Righttube respectively. The end of transition occurred at approximately the same Reynolds number (3 100) for all three tubes of the tripletube test section. Since the Centretube entered transition earlier than the outer tubes, maldistribution was evident, with the water taking the path of least resistance. The flow rate in the Centretube showed an average difference of 2.8% in the Reynolds number range of 1 970 to 3 150. Maldistribution proved to be negligible when all three tubes were in the laminar or turbulent flow regimes.
Supervisor: Professor J. P. Meyer
Ntumba Tshimanga, 2016. "Experimental investigation and model development for thermal conductivity of glycerolbased nanofluids"
Glycerol has historically been used as an antifreeze fluid to facilitate heat transfer in the automotive and air conditioning and refrigeration industries. It has also been used as a lubricant in the processing of food and the production of pharmaceuticals and cosmetics. Although a lot of work has been done recently to evaluate the potential to enhance heat transfer using nanoparticles mixed with a base fluid to form a nanofluid, no work has been done on using glycerol as a base fluid. Therefore the purpose of this study was to investigate the effect of nanoparticle volume fraction, nanoparticle size and temperature on the thermal conductivity of stable glycerolbased nanofluids. Two types of metal oxide nanoparticles were considered namely MgO and αAl_{2}O_{3}. The particle sizes of the MgO ranged from 21 nm to 119 nm and for the αAl_{2}O_{3 }it ranged from 31 nm to 134 nm. The thermal conductivities were determined by experimental measurements and with analytical and empirical models. The thermal conductivity measurements were taken at temperatures ranging from 20˚C to 45˚C, for nanofluids prepared at volume fractions ranging from 0.5% to 4%. The nanofluids were prepared with a twostep method that included ultrasound mixing to ensure the nanoparticles were fully dispersed and deagglomerated in the glycerol. The experimental results showed that both the αAl_{2}O_{3}glycerol and MgOglycerol nanofluids had substantially higher thermal conductivity than the base fluid. It was also found that at room temperature, the effective thermal conductivity remains almost constant for at least 50 hours. The maximum thermal conductivity enhancement for the αAl_{2}O_{3}glycerol nanofluids was observed for a 4% volume fraction to be 19.5% for a nanoparticle size of 31 nm. For the MgOglycerol nanofluids the maximum thermal conductivity enhancements were also for a volume fraction of 4%, however, the enhancement was 18% for a particle size of 21 nm. Furthermore, the thermal conductivities as function of nanoparticle size, volume fraction and temperature, of the two nanofluids were investigated. It was found that the thermal conductivities of the α‑Al_{2}O_{3}glycerol nanofluids were significantly more dependent on particle size than the MgOglycerol nanofluids. Furthermore, it was found that no equations exist at present that can accurately predict the thermal conductivity of glycerol based nanofluids and therefore new empirical equations correlations were developed.
Keywords: Nanofluids, thermal conductivity, glycerol, nanoparticle size, volume fraction, temperature, stability.
Supervisors: Dr Mohsen Sharifpur and Prof Josua P Meyer
Wietsche Clement William Penny, 2016. "Antilock Braking System performance on rough terrain
Wietsche"
The safety of motor vehicles is of primary concern in the modern age as the death rate of road users are still at unacceptably high numbers and is the second largest cause for unnatural death worldwide. Consumers often expect unrealistic performance and comfort levels from their vehicles regardless of terrain or conditions, and the Sport Utility Vehicle class is often under the most pressure to meet these high expectations.
Literature reveals that the performance of Antilock Braking Systems (ABS) deteriorates on rough offroad terrains due to a number of factors such as axle oscillations, wheel speed fluctuations and deficiencies in the algorithms. This leads to complications such as loss of vertical contact between the tyres and the terrain and poor contact patch generation that eventually results in reduced longitudinal force generation.
In this study, an ABS modulator is retrofitted on a test vehicle to perform brake pressure control. The hydraulic modulator is controlled by an embedded computer, running the Linux operating system, onto which a slightly modified version of the Bosch ABS algorithm is coded in Clanguage. Brake tests are conducted with the vehicle on hard concrete terrains for both smooth roads and rough Belgian paving. The algorithm is also implemented in Matlab/Simulink using cosimulation with a validated nonlinear full vehicle ADAMS model employing a validated FTire tyre model. The cosimulation model was validated with the test data on both flat and rough terrains and experimental results correlate well with simulation results when the recorded brake pressures from the test data are given as input to the simulation model.
Test data and simulation results indicate that wheel speed fluctuations can cause inaccuracies in the estimation of vehicle velocity and excessive noise on the derived rotational acceleration values. This leads to inaccurate longitudinal slip calculation and poor control decisions respectively. Although possible solutions to the identified problem are not explored in detail, the developed simulation model and test vehicle can be used to test improved ABS algorithms and suspension control strategies to solve the deterioration of ABS performance on rough terrain.
Supervisor : PS Els
Erin Sarah Vause, 2016. "THE INLET EFFECTS OF MULTIPLE TUBES ON THE ADIABATIC PRESSURE DROP OF SMOOTH, HORIZONTAL TUBES, IN THE TRANSITIONAL FLOW REGIME"
There exists many investigations in the field of pressure drop inside smooth tubes; however, there is a research gap where this flow is in the transitional flow regime. All work in the transitional flow regime thus far has been concerned with a single tube and in limited cases the effect of different types of inlets were investigated. Many heat exchangers such as shellandtube heat exchangers consist of a number of closely packed tubes leading from a common header and in some cases may operate in or close to the transitional flow regime. However, it is not known what effect adjacent tubes will have on pressure drop and the flow distribution in the transitional flow regime. The purpose of this study is therefore to investigate the effect that the presence of adjacent tubes will have on the pressure drop in the transitional flow regime. The study is limited to smooth and circular, horizontal tubes in the fully developed, transitional flow regime and adiabatic pressure drops. The transition effects were investigated experimentally by developing and building an experimental setup on which, firstly, the pressure drops could be measured of one tube to be used as a reference and then secondly, the pressure drop of three tubes in parallel, equally spaced, 1.4 diameters apart. The internal tube diameters of all tubes were 3.97 mm and the tube lengths were 6 m. The pressure drops were measured over a length of 1.97 m, at the end of the tubes where the flow was fully developed. The pressure drops were measured with pressure transducers while the inlet and outlet temperatures of the water were measured with PT100 probes. All the tubes were connected to a calming section to ensure a squareedge inlet. Experiments were conducted with water at Reynolds numbers from 700 to 5 100 to ensure that the pressure drops and thus friction factors could be determined for fully developed flow throughout the laminar, transitional and turbulent flow regimes. The uncertainty of the friction factors were all less than 1%. It was found that the centre tube experienced an earlier onset of transition than that previously seen in single tube tests, at a Reynolds number of 1 840. In the outer tubes, transitional flow was delayed well beyond that previously seen in literature. The flow in all three tubes underwent transition into fully developed turbulent flow by the maximum Reynolds number of 3 340. The effect of having multiple, adjacent inlets caused a maldistribution in the mass flow rate, with a 5.8% difference in the flow rates of the outer tubes in the transitional flow regime. New correlations were developed to predict the friction factor for transitional flow in each of three adjacent tubes at an inlet pitch distance of 1.4 times the inner tube diameter. Overall, it can be concluded that multiple tube entrances have an effect on the transitional flow in all of the tubes and should be further investigated for other pitch distance and tube arrangements.
Supervisor: Prof J.P. Meyer
Share this page
Last edited by Bradley Bock
Edit