2021

 

 

B. D. Bock, 2021. " Surface influences on falling film boiling and pool boiling of saturated refrigerants"

Falling film evaporators that operate in the nucleate boiling regime in the refrigeration industry offer a number of advantages over their flooded counterparts such as lower refrigerant charge and at times improved heat transfer. Existing literature has not characterised the influence of surface characteristics on the falling film boiling process, and they are poorly understood for the pool boiling process. The purpose of this study was therefore to experimentally measure the influence of roughness, material and nanostructures on the heat transfer of falling film boiling and pool boiling of saturated refrigerants on the outside of horizontal tubes. The critical heat flux point was measured if it occurred, and the falling film heat transfer enhancement ratio, critical dryout threshold and general dryout characteristics were investigated in the study.

The tubes tested consisted of plain copper, stainless steel and mild steel tubes that were polished and roughened with various grades of sandpaper. Furthermore, three types of nanostructured surfaces were applied to polished copper tubes, namely a layer-by-layer (LbL) coating of silica nanoparticles, a copper oxide (CuO) nanostructure coating and a commercial nanocoating process termed nanoFLUX.

The nanoFLUX tube had the highest heat transfer coefficients of tubes tested under both pool boiling and falling film conditions, with between 40 and 200% higher heat transfer coefficients than those of a polished copper tube. The nanoFLUX surface outperformed the other surfaces due to a combination of rougher microstructure and a unique heat transfer mechanism, possibly linked to capillary wicking of liquid inside the nanochannels of the porous coating.

The falling film heat transfer enhancement ratio was found to increases as surface roughness was increased on plain tubes, suggested to be as a result of enhanced microlayer evaporation from the trapped sliding bubbles in the thin flowing film.

The nanoFLUX and CuO surfaces experienced lower critical heat flux as a result of departure from nucleate boiling under pool boiling and falling film boiling conditions compared with plain surfaces._. However, the nanoFLUX and CuO tubes performed well in terms of critical dryout at lower heat fluxes. The wicking capabilities of the nanoFLUX and CuO surfaces were thought to be the cause of their improved dryout capabilities at lower heat fluxes, but increased heat fluxes possibly led to dryout of the nanostructures resulting in operation in the Cassie-Baxter state and subsequent reduced wettability.

Supervisor: Prof. JP Meyer

Co-Supervisor: Prof. JR Thome

---

JC van der Walt, 2021. " Some challenges and solutions of steady-state model-based leak detection"

Leak detection of water networks poses significant concern worldwide. In this study, we investigate steady-state model-based leak detection. Three main limitations of this technique exist and include the sensitivity to noise, model calibration and sensor placement. The study investigates these three limitations.

Various strategies to identify leaks using the model-based leak detection technique were identified and the study shows that each of these strategies has its limitations. An ensemble of strategies is suggested instead of the development of new strategies. The identification of leaks on the considered networks shows how the leak detection accuracy and sensitivity to noise vary within a pipe or network depending on the leak size, location and sensor placement.

A model calibration technique is introduced, capable of calibrating a model with existing leaks. This is a significant weakness of current model-based leak detection methods. The introduced technique maps the measurements to a model with known parameters instead of the traditional prediction of model parameters. Sensor placement on a simple transportation network is investigated. The best sensor combination is identified, which together with the proposed model calibration technique allows for leak detection to be performed on an uncalibrated experimental network. The detection of leaks proved to be sufficiently accurate.

In this study, model-based leak detection is shown to be possible for the detection of leaks in simple steady-state transportation networks. However, further research is required to apply this technique on large water distribution networks.

Supervisor: Prof. PS Heyns

Co-Supervisor: Prof. DN Wilke 

 

---

2020

 

---

A. Bashir, 2020. " Single‐phase forced and mixed convection in the laminar and transitional flow regimes of inclined smooth tubes with inlet disturbances"

Laminar and transitional flow regimes in tubes have been extensively investigated in the literature. However, there are several gaps in the forced and mixed convection literature, especially for inclined tubes with different inlet disturbances. The purpose of this study was to experimentally investigate the effect of tube inclination and inlet contraction ratio on the single-phase heat transfer and pressure drop characteristics in the laminar and transitional flow regimes for pure forced and mixed convection conditions.

An experimental set-up was designed, constructed and validated against literature with the test section in a horizontal and different vertical orientation. The test section was 4.6 m long and was made from a smooth hard drawn copper tube with measured inner and outer diameters of 5.1 mm and 6.3 mm, respectively. Experiments were conducted at various inclination angles from vertical upward flow (+90º) to vertical downward flow (–90º), with horizontal flow (0º) and several other angles in between. A total of 2 679 mass flow rate measurements, 174 135 temperature measurements and 2 679 pressure drop measurements were conducted using water (Prandtl numbers between 3.5 and 8.1) as working fluid. The Reynolds number range covered were from 400 to 6 000 at constant heat fluxes varying from 1 to 8 kW/m². Four different types of inlets namely; square-edged and re-entrant inlet with different inlet contraction ratios (5, 11, 14 and 33), as well as hydrodynamically fully developed and 90º bend inlets were used.

It was found that an increase in the inclination angle from horizontal flow (0º) to vertical (±90º) flow, decreased the buoyancy effects which led to decreased laminar heat transfer coefficients and friction factors for both upward and downward flows. The onset of buoyancy effects was significant near the vertical inclination angles and caused a rapid increase in the laminar heat transfer coefficients and friction factors when the inclination angles moved from vertical to horizontal orientations. An inclined tube Grashof number which is a function of inclination angle was defined and used to express the laminar Nusselt numbers as a forced convection part plus an enhancement component owing to mixed convection. The laminar friction factors were expressed as a function of a forced convection/isothermal part multiplied by the mixed convection part. Furthermore, it was found that the critical Reynolds numbers at which transitional flow regime started increased as the inclination angles increased from horizontal to vertical, while the end of transitional flow regime were inclination angle independent. This caused the width of the transitional flow regime to decrease, as well as the transition gradients to increase, with increasing inclination angles at different heat fluxes. It was also found that the flow directions (upward and downward) had a negligible effect on the heat transfer coefficients and friction factors in the entire transition and quasi-turbulent regions. 

The fully developed laminar forced convection Nusselt numbers were not constant at 4.36, but were a function of Reynolds number for Reynolds numbers higher than 1 000. Therefore, a revised laminar Nusselt number correlation for smooth circular tubes was developed. The fully developed laminar forced convection friction factors were, as expected, equal to 64/Re. For both the forced convection heat transfer and pressure drop characteristics, transition occurred at the same mass flow rates for all the heat fluxes, including isothermal flow, but the critical Reynolds numbers increased with an increase in heat flux. For forced convection condition, the width of the transitional flow regime in the fully developed region remained constant for all heat fluxes.

For a square-edged inlet geometry, the transition from the laminar to the turbulent flow regimes occurred earlier as the inlet contraction ratio increased, while for the re-entrant inlet, transition was delayed. The transitional flow regime was significantly affected by smaller contraction ratios and this effect increased with increasing heat flux.  However, it was found that the critical Reynolds numbers were independent of inlet geometry for contraction ratios larger than 33.  For the 90º bend inlet, transition occurred earlier than all the other inlet geometries and contraction ratios.

Supervisor:  Prof JP Meyer

Co-Supervisor::  Dr M Everts

---

I. E. Kalu, 2020."Failure Assessment of Boiler Tubes under Localized External Erosion to Support Maintenance Decisions"

Boiler tubes used in power plants and manufacturing industries are susceptible to numerous failures due to the harsh environment in which they operate, usually involving high temperature, pressure and erosive-corrosive environment. Among the wide range of failures associated with the tubes, localized external erosion is prevalent. In spite of efforts made over the years to solve this problem, localized erosion of boiler tubes continues to be a leading cause of tube leakages and unscheduled boiler outages in power plants and other utilities. There is, therefore, a need to approach this problem systematically and engage in rigorous studies that will allow improved management of this persistent problem.

In this thesis, comprehensive studies were first carried out on modelled variants of localized external eroded boiler tubes with conceptualized flaw geometries, such as could be seen in real situations. The outcome of these investigations provided insights into the factors that influence the failure of these tubes while in use. The stress concentration, plasticity and flaw geometry all play critical roles in influencing the failure of tubes. Also, the failure pressures of the modelled tubes were analyzed in relation with several other failure criteria, to determine which failure criteria will be most suitable for the failure assessment of the localized tubes. Based on the result of the analysis, plastic strain in the range 5%-7% is recommended as a compromise between the extreme benchmark failure criterion of 20%, and the overly conservative 2%.

The insights gained from the studies carried out on conceptualized variants of localized thinned tubes were extended to real localized external eroded tubes obtained from the industry and used to develop an improved and efficient failure assessment methodology framework for heat resistant seamless tubes while in service. This was done by treating the tubes as an inverse problem and using an optimization technique to obtain the flaw geometric properties of the tubes so as to effectively replicate them on the conceptualized geometries. Using two Material Properties Council (MPC) models generated based on the properties of the tubes as a function of their operating temperatures, comprehensive nonlinear finite element analyses (NLFEA) were conducted on the 160 finite element models. These tubes were assessed based on the maximum equivalent plastic strain and Von Mises stress produced at the deepest point of the flaw area within each of the tubes when subjected to their respective operating pressures at which they failed. The failure assessment outcome revealed that most of the heat resistant tubes while in service will remain intact and not fail if their remaining tube thicknesses were within (0.7 t_min to t_min), where t_min is the minimum remaining thickness of the tube based on allowable stress. In addition, a 5% plastic strain (P_5%) and equivalent Von Mises stress criteria of 0.8 σ_uts were deduced as failure criteria to guard against the failure of these tubes while in service, and also avoid their early replacement. The developed methodology framework was checked and compared with the API-ASME FFS standard and found to be in good agreement with it, also more efficient and with reduced conservatism.

Finally, sensitive studies were conducted based on the developed methodology to examine how the combination of the flaw geometry and material factors could possibly influence the failure of the tubes while in use. The study outcome shows that there were no appreciable changes in the normalized Von-Mises stress ratios and the plastic strain response for the normalized remaining thickness of the tubes. The proposedP_5% and 0.8 σ_uts limits accurately predicted the failure for all the tubes and were reasonably safe limit for the tubes. Insights gained from the strain hardenability of the tubes studied will also provide guidance with taking proactive measures for the maintenance of the tubes.

In summary, all the insights gained from this research and the developed failure assessment methodology framework will be helpful in categorizing the severity of localized external erosion on tubes while in use, and also support maintenance decisions on these critical assets.

Keywords: Boiler tubes, localized external erosion, plastic deformation, stress concentration, flaw geometry, failure criteria, plastic strain, conceptualized finite element models, nonlinear finite-element analysis, equivalent Von Mises stress, API-ASME FFS Standard.

Supervisor: Dr. H.M.Inglis 

Co-Supervisor: Prof S. Kok

---

D Kafka, 2020."Automated learning rates in machine learning for dynamic mini-batch sub-sampled losses"

Learning rate schedule parameters remain some of the most sensitive hyperparameters in machine

learning, as well as being challenging to resolve, in particular when mini-batch subsampling is considered. Mini-batch sub-sampling (MBSS) can be conducted in a number of ways, each with their own implications on the smoothness and continuity of the underlying loss function. In this study, dynamic MBSS, often applied in approximate optimization, is considered for neural network training. For dynamic MBSS, the mini-batch is updated for every function and gradient evaluation of the loss and gradient functions. The implication is that the sampling error between mini-batches changes abruptly, resulting in non-smooth and discontinuous loss functions. This study proposes an approach to automatically resolve learning rates for dynamic MBSS loss functions using gradient-only line searches (GOLS) over fifteen orders of magnitude. A systematic study is performed, which investigates the characteristics and the in ability of GOLS to resolve learning rates. GOLS are shown to compare favourably against the state-of-the-art probabilistic line search for dynamic MBSS loss functions. Matlab and PyTorch 1.0 implementations of GOLS are available for both practical training of neural networks as well as a research tool to investigate dynamic MBSS loss functions.

Supervisor: Prof DN Wilke

---

S M M Osman, 2020. "Experimental investigation into convection heat transfer in the transition flow regime by using nanofluids in a rectangular channel"

The growing demand for energy worldwide requires attention to the design and operating of heat exchangers and thermal devices to utilise and save thermal energy. There is a need to find new heat transport fluids with better heat transfer properties to increase convective heat transfer, and nanofluids are good alternatives to conventional heat transport fluids. Although extensive research has been done on the properties of nanofluids in recent decades, there is still a lack of research on convection heat transfer involving nanofluids, particularly in the transitional flow regime. This study focused on the application of nanofluids in heat exchangers as heat transport fluids by investigating forced convective heat transfer of alumina-water and titanium dioxide-water nanofluids prepared by using the one-step method. The particle size used was 46 nm and 42 nm for the aluminium oxide and the titanium dioxide respectively. Uniform heat flux boundary conditions were used by uniformly heating the rectangular channel electrically. Nanofluids with volume concentrations of 0.3, 0.5 and 1% were used for the alumina-water nanofluids, and volume concentrations of 0.3, 0.5, 0.7 and 1% were used for the titanium dioxide-water nanofluids. The viscosity of the nanofluids under investigation was determined experimentally, while the thermal conductivity and other properties were predicted by using suitable correlations from the literature. A Reynolds number range of 200 to 7 000 was covered, and the investigated flow rates included the laminar and turbulent flow regimes, as well as the transition regime from laminar to turbulent flow. Temperatures and pressure drops were measured to evaluate heat transfer coefficients, Nusselt numbers and pressure drop coefficients. Heat transfer and hydrodynamic characteristics in the transition flow regime were carefully studied and compared with those in the transition regime when flowing pure water in the same test section.

The study also investigated another approach of enhancing heat transfer in heat exchangers by increasing the heat transfer area of the heat exchanger itself, and this was done by filling the rectangular test section with porous media to increase the heat transfer surface area and thus enhance heat transfer. Hence in this study, the effect of using porous media was also studied by filling the rectangular test section with high-porosity nickel foam. The permeability of the used nickel foam was determined by conducting pressure drop measurements through the nickel foam in the test section, and heat transfer and pressure drop parameters were measured and compared with those in the empty test section.

The results showed that all the nanofluids used enhanced heat transfer, particularly in the transition flow regime. The 1.0% volume concentration alumina nanofluid showed maximum enhancement of the heat transfer coefficient, with values of 54% and 11% in the turbulent regime. The maximum enhancement of the heat transfer coefficient was 29.3% in the transition regime for the 1.0% volume concentration titanium dioxide-water nanofluid. The thermal performance factor in the transition flow regime was observed to be better than that in the turbulent and laminar flow regimes for all the nanofluids.

The results of the nickel foam test section showed that the values of the friction coefficient were 24.5 times higher than the values of the empty test section, and the Nusselt number was observed to be three times higher when using nickel foam than without foam in the test section. No transition regime was observed for the foam-filled test section on either the heat transfer results or the pressure drop results; however, transition from laminar to turbulent was found for the test section without foam. The results of the thermal factor of the foam-filled test section showed a thermal performance factor higher than unity through the entire Reynolds number range of 2 000 to 6 500, with better thermal performance factor at lower Reynolds number.

Supervisor: Prof Mohsen Sharifpur

Co-supervisor: Prof Josua Meyer

---

A.S. Shote, 2020, "Aerodynamic losses and endwall heat transfer in a linear vane cascade with endwall film-cooling and endwall-leading edge contouring"

Elevated inlet temperature is desirable in turbine passages. This is not without a whole lot of aerodynamic loss, endwall thermal stresses consequences and pressure loss penalties most especially downstream of the turbine passage. Film cooling and endwall modification are needed to increase the efficiency and prolong the material life of the turbine endwall and components. The main objective of this work is to investigate the heat transfer influence at the near endwall region by employing the leading edge contouring and film-cooling at the endwall.

The hub-side blade profile and passage geometry of the 1st stage nozzle guide vane of a GE-E 3 turbine engine are scaled up and employed in a linear cascade for the experimentation at a Reynolds number of 2.1E+05 based on turbine vane- blade chord and reference velocity. The hot endwall of the turbine engine was replicated experimentally by supplying constant heat flux at the endwall. A numerical model of the cascade is employed in a commercial computational fluid dynamics (CFD) code STAR CCM+ TM to optimize fillet shape. CFD is employed to predict the baseline case (no film-cooling, no endwall modification) and Fillet-1 endwall modification case (no film-cooling). Experimentally, the efficacy of four unique upstream film-cooling schemes and geometries are investigated with two linear fillets employed at the endwall-blade junction. The film-cooling is employed at five different inlet blowing ratios (M = 1.0, 1.4, 1.8, 2.2 and 2.8). The interaction of the film-cooling jets with the main flow inside the cascade are captured and analyzed. The leading-edge film cooling schemes tested include the flush slots, the slot-discrete holes combination, the discrete holes and the linear inlet beveled/curved holes. The distance travelled and spread of the film cooling along the endwall are assessed with respect to the effectiveness and non-dimensional temperature. Two linear contoured endwall geometries known as the fillets were employed at the blade-endwall junction along with the film-cooling configuration at the endwall for the experimental measurement.

The computational results of static pressure on the vane blade surface at the mid- span of the baseline and Fillet-1 cases (M = 0) shows good agreement with the experimental cases. The simulation also captures appropriately the passage vortex as presented by the experimental result of the coefficient of the total pressure loss. The location and the magnitude of the total pressure loss distribution of the computational result are comparable with that of the experiment. The linear variations of Fillet-1 in the axial and spanwise directions have significant effect on the approaching endwall boundary layer of the horse shoe vortex. The
optimized contoured endwall reduces the size and magnitude of the horse shoe vortex at the near endwall of the vane-blade. High Nusselt number (Nu) is recorded at the throat region of the endwall for the baseline without fillet case in the experiment. This is attributed to the flow acceleration effect. However, with the installation of Fillet-1 and Fillet-2, the magnitude of the Nusselt number at the throat region reduced by (20 – 40)% from that of the baseline case. Fillet-2 has the lowest reduction effect on the Nu at the throat region of the endwall. The non-dimensional temperature of the flow field near the endwall shows that Fillet-1 and Fillet-2 improve the endwall film cooling coverage in both pitchwise and axil directions. In general, high film cooling jet flux, provides better cooling and endwall coverage as the momentum of jet reduces the pitchwise cross flow that is responsible for high heat transfer and aerodynamic loss. With the injection of the high momentum film cooling flow (M = 2.8), the result shows excellent improvement in the adiabatic effectiveness for all the leading-edge film cooling configurations. The curved holes produced the best effectiveness distributions at the endwall. However, the slot-discrete hole configuration with the fillet or without the fillet are having the best reduction influence on the passage vortex at the exit plane of the turbine cascade passage. Therefore, the magnitude and size of the passage vortex are reduced significantly. At blowing ratio ≤ 2.2, with the fillet and without the fillet, the main flow has more influence on the film-cooling jets thereby reducing the adiabatic effectiveness coverage of the endwall. Generally, Mass fractions are found to increase as the blowing ratio increases and decrease with endwall modification. Fillet-1 recorded the least mass fraction for all blowing ratios investigated.

Supervisor: Dr G.I. Mahmood

Co-Supervisor: Prof. J.P. Meyer