<p>The objectives of this research is to determine the mechanisms involved in the development of electrohydrodynamic (EHD) flow structures, such as twisted liquid cones, twisted liquid columns and entrained liquid droplets, in a two-phase flow and to determine their impact on convective condensation heat transfer. EHD involves the application of high voltage to a dielectric fluid flow to induce electric forces that can create additional convection currents in single-phase and two-phase flows and also the redistribution of the different phases within the channel in a two-phase flow. The EHD phenomenon was investigated inside a smooth horizontal tube with a concentric rod electrode used to apply high voltage to a flow of refrigerant R-134a.</p> <p>The EHD flow structures were initially observed to be a transient phenomenon that appeared during the initial step input of high voltage and were not present after steady state. Investigations to sustain the EHD two-phase flow structures using various ±8kV pulse width modulated waveforms (PWM) are performed for a range of mass flux between 45 to 110 kg/m^2 s and a quality of 50%, conditions which correspond to a stratified/stratified wavy flow pattern. The effect of sustaining flow patterns consisting of these flow structures on condensation heat transfer and pressure drop is measured. The purpose of this research is to evaluate the potential of sustaining specific EHD two-phase flow patterns as a means of enhancement and control of heat transfer and pressure drop for industrial heat exchange applications.</p> <p>An evaluation of the mechanisms involved in the development of the EHD flow structures was first performed by determining the effect of applying positive and negative polarity high voltage to a single-phase flow to determine the mechanisms of charge injection in this particular geometry. A negative polarity high voltage applied to the concentric rod electrode was found to result in a larger heat transfer enhancement due to the charge injection occurring at the center electrode, whereas a positive polarity high voltage applied to the rod electrode resulted in negative charge injection at the grounded tube wall. This method of charge injection was used to explain the difference in the development of the twisted liquid cone and twisted liquid column flow structures that arise due to an applied positive and negative voltage respectively.</p> <p>An analysis of the development and sustainability of the EHD two-phase flow structures using various PWM waveforms was investigated using a high speed camera to visualize the flow. An image analysis of these high speed videos determined the suitable pulse width, duty cycle and voltage polarity conditions for the development of the twisted liquid cones/columns and in sustaining the structures during the pulse ON period. The sustainability of these flow structures were determined to be mainly influenced by the charge distribution within the flow and a means to sustain these flow structures for a range of duty cycles between 10%-90% was found using waveforms consisting of both positive and negative high voltage pulses. For duty cycles of 100%, an inverse annular flow pattern was observed, where liquid is extracted and encircles the concentric rod electrode. For the entrained droplets produced using EHD, the required high voltage pulse conditions for the production of the droplets was determined and the effect of EHD on the coalescence of droplets was also observed.</p> <p>Experiments were performed to measure the condensation heat transfer and pressure drop performance of flow patterns consisting of EHD flow structures. The results show that flow patterns consisting of different EHD flow structures exhibit different heat transfer and pressure drop characteristics. The heat transfer enhancement was determined to be a result of the extraction of liquid from the heat transfer surface and into the central core, the increase in convection within the phases and a disruption of the thermal boundary layer. The increase in pressure drop is due to the increase in frictional pressure drop between the liquid and electrode surfaces and an increase in the interfacial area resulting in greater liquid-vapour shear. The maximum enhancement of heat transfer and the maximum increase of pressure drop by using the EHD technique were determined to be 2.9-fold and 5.0-fold respectively. The results also indicate that EHD can be used to independently control the heat transfer and pressure drop as it was shown that for a fixed heat transfer or pressure drop performance, a range of corresponding pressure drop or heat transfer conditions can be achieved depending on the flow pattern established using PWM waveforms. The findings in this study show the promise in utilizing the EHD technique as an effective means of enhancement and control of heat transfer and pressure drop performance in advanced heat exchange systems.</p> / Master of Applied Science (MASc)

<p>Combustion at the conditions found in practical applications is extremely complex and must be well characterized to support the next generation of combustors. This task is extremely difficult both experimentally and numerically, where the fast timescales and small spatial scales restrict complete characterization. This complexity is mitigated for numerical simulations through the use of computationally efficient models that approximate problem physics. Limitations arise in validating these approximations, due to the inability to experimentally measure the required quantities combined with the prohibitive numerical cost of performing detailed simulations. Fundamental investigations address this difficulty&mdash;simplified flame geometries can be examined in more complete detail and can be used to support the validation of modeling approximations.</p> <p>Presented here is the experimental and numerical characterization of premixed laminar tubular flames fueled with dilute lean hydrogen mixtures. This fundamental flame geometry exhibits structure similar to practical turbulent flames (locally varying curvature with extinction zones), yet retains a time independent 2D planar structure. This aspect greatly reduces the experimental and computational requirements, permitting: (1) a more complete experimental and numerical characterization and (2) detailed numerical experiments to directly validate modeling approximations.</p> <p>Experimental measurements using non-intrusive laser diagnostic techniques are presented that provide 2D spatially resolved and quantitative measurements of temperature, major species, and two minor species (H and OH) through flame cross-sections. The detailed flame structure is then simulated using a 2D fully-implicit primitive variable finite difference formulation that includes multicomponent transport and detailed chemical kinetics. Comparisons between the experimental and numerical data sets are presented and show overall good agreement, though with significant quantitative discrepancies. These discrepancies are examined, and are further investigated with numerical experiments to better elucidate the dependence of cellular flame appearance on experimentally controlled variables. The cellular tubular flame is found to be highly sensitive geometry that may be used for validating diffusive transport modeling approximations. This capability is exemplified through the development of a simple and accurate approximation for thermal diffusion (i.e. Soret effect) that is suitable for practical combustion codes.</p>

The goal of this work was to determine if and how footwear properties could be modified to alter walking biomechanics. A test shoe was developed to allow heel and forefoot midsole cushioning to be quickly and independently varied. A systematic gait analysis study of 8 healthy human participants was performed to isolate and quantify the effects of shoe cushioning. Walking data were collected using a force-instrumented treadmill and infrared motion-capture system. Center-of-mass (COM) power, joint (ankle, knee, and hip) power, and deformable foot power were computed from these kinematic and kinetic data. Peak power and work summary measures were computed for key phases of the gait cycle: for push-off (the end of stance phase) and collision (immediately after foot contact). As hypothesized, forefoot cushioning primarily affected push-off power, with little effect on collision. Similarly, heel cushioning affected collision power, with minimal effect on push-off. COM peak push-off power was found to increase with softer forefoot cushioning and COM peak collision power became more negative with softer heel cushioning. Lower forefoot bending stiffness was also observed to increase peak COM push-off power. Therefore, a shoe intended to increase push-off power (e.g., for gait rehabilitation) might be designed with a soft and flexible forefoot midsole. These findings inform how shoe cushioning could be modulated to augment locomotor performance.

Surface plasmons, coherent oscillations of electrons in metals that can be excited with electromagnetic waves, are a key component in routing and manipulating light-matter interaction at nanometer length scales. Because of their deep-subwavelength mode volumes and strong field confinement, surface plasmons provide a means to realize numerous innovations such as metamaterials & metasurfaces, sub-diffraction limited imaging, sensing, and cloaking. While the non-radiative decay of plasmons has been long considered to be a parasitic loss, recent research has shown that it can be harnessed for a number of applications including photothermal heat generation, photodetection, photovoltaics and photocatalysis. Despite the significant advances, research in this area is still in its infancy with devices generally suffering from low efficiencies. This thesis focuses on understanding how plasmonic nanostructures can be properly engineered to take full advantage of the non-radiatively plasmon decay process for realizing new functionalities, as well as enhancing the efficiency of photothermal heating and photoelectrical energy conversion systems.

Several technical needs were identified and addressed for advancing the Strain Energy Accumulator (SEA), which is an energy storage device consisting of an expandable rubber bladder inside of a rigid shroud that stores energy in the form of pressure and strain. First, multiscale modeling methods were investigated to estimate the homogenized elastic modulus of carbon nanotube (CNT) rubber. The result is homogenized modulus estimates ranging from a few times to almost 80 times the elastic modulus of rubber, indicating the need for validation of existing models or development of new models to estimate the modulus for matrix and inclusion materials having drastically dissimilar moduli. Second, an analytical methodology was developed for simultaneously characterizing the energy storage in pneumatic and strain energy systems including component efficiency. By incorporating uncertainty analysis, the efficiencies of the strain energy accumulator are measured in over 2500 cycles of testing to be consistently over 93 %. Third, system state efficiency models were developed and expanded. Through experimentation, the model was determined to be favorably conservative with system efficiency projections ranging from 31 % to over 60 % depending on the system configuration. In addition, materials challenges in high pressure applications led to the conceptual investigation of CNT elastomers offering improved material strength properties and the potential for self-sensing. In previous research, carbon nanotube sensor thread was tested as a distributed sensor on carbon fiber reinforced composites and was able to monitor strain and detect damage in composite panels. The use of nanomaterials for self-sensing was extended in the current work with proof of concept tests performed on electrically conductive elastomers that exhibited the ability to monitor load and detect damage in specific directions. Each of these contributions in the areas of materials modeling, uncertainty analysis, and component and system efficiency quantification techniques has helped to advance the Strain Energy Accumulator technology.

<p> Small scale unmanned underwater vehicles provide an opportunity to safely and efficiently complete tasks such as boat hull inspection and subsea development survey. These Remotely Operated Vehicles (ROVs) can be made more efficient if navigating underwater autonomously. This requires the development of highly accurate navigation and control algorithms, which, in turn, require a high-fidelity dynamic model of the vehicle based on first principles and validated by empirical data.</p><p> In this thesis, a simulation of a dynamics model for a commercially available ROV is developed. Empirical data from open-loop testing is used to generate a second-order transfer function using system identification to validate the simulation model. The transient response characteristics of the experimentally generated transfer function are then utilized to fine-tune the physical parameters in the simulation model. Finally, autopilot systems are designed using classical control theory to enable autonomous control over the attitudes and depth of the underwater vehicle.</p>

<p> This thesis describes the development of a two-dimensional, high-order, fluid-structure interaction (FSI) solver. The well-established spectral difference (SD) method is used for spatial discretization of the Euler equations over deforming, unstructured quadrilateral grids. The Geometric Conservation Law (GCL) is incorporated into the conservative Euler equations, before discretization. After simplification, the equations reduce to a form, in the computational domain, identical to the equations in the physical domain. In this form, the equations can be integrated implicitly in time, without the requirement of any additional source term, to guarantee free-stream preservation. The fluid and structure sub-systems are individually integrated in time using the explicit first stage, single diagonal, diagonally implicit Runge-Kutta (ESDIRK) method. As the first step to solving the coupled, non-linear Euler equations, implicit in time, we linearize the governing equations. The resulting linearized simultaneous equations are then solved sequentially using lower-upper symmetric Gauss-Seidel (LU-SGS) relaxation sweeps. The fluid and structure sub-systems are loosely coupled and the coupling term is integrated in time using an explicit RK method, resulting in an implicit-explicit (IMEX) RK coupling. The spatial accuracy and the free-stream preserving ability of the solver are demonstrated by testing a supersonic, isentropic vortex in a curved channel. Next, the temporal accuracy of the solver is established using an Euler vortex propagation test case. It is also demonstrated that the four-stage ESDIRK is capable of handling time-steps 50 times larger than the four-stage explicit RK. In each of these cases, third- and fourth-order SD for spatial discretization and second-order backward difference (BDF2) and third-order, four stage ESDIRK for time integration were tested. Since the loose (explicit) FSI coupling restricts permissible structural deformation, we limit ourselves to small harmonic oscillations resulting from linearized perturbed Euler equations. The interaction between a linear piston and an inviscid, compressible fluid is simulated to demonstrate that the IMEX coupling does not contaminate the spatial or temporal accuracy of the implemented high-order methods. Through rigorous testing, this development is expected to lay a foundation for a powerful computational framework for various fluid-structure interaction problems.</p>

The influence of pulse energy on nanosecond pulse driven dielectric barrier discharge (ns-DBD) plasma actuation applied to the turbulent shear layer downstream of a backward facing step (BFS) is examined experimentally. The ns-DBD control mechanism, which is believed to be primarily thermal in contrast to most other flow control actuators, has been demonstrated in various high speed shear flows yet questions on fundamental physics and scaling remain unanswered. In this work, ns-DBD pulse amplitude is varied between 0.13mJ/cm-0.88mJ/cm per pulse in order to understand its effects on the excitation of a turbulent shear layer. This work is carried out at a fixed actuation frequency of 1000Hz which corresponds to St(θ) ≈ 0.005 based on initial momentum thickness. Both mean and phase-averaged data indicate a preference for the 0.33mJ/cm and 0.56mJ/cm pulse amplitudes. However, further analysis concludes that 0.33mJ/cm is the most favorable as seen from momentum thickness growth and TKE distribution. Further analysis through the use of schlieren imaging suggests that depreciating control authority for the highest pulse amplitude of 0.88mJ/cm may be a result of either increased plasma three dimensionality resulting in non-uniform heating, or a thermal saturation mechanism (overheating). Additional theoretical studies are required to substantiate these claims and to decipher between the two.

<p> Composite materials are receiving increasing attention and broadly used in aerospace industry due to their superior strength-to-weight ratio, corrosion resistance and design flexibility. The need for rapid nondestructive evaluation (NDE) techniques for composites is growing rapidly as the complexity and dimensions of the structures are increasing significantly. Structural health monitoring (SHM) has been attracting much attention as a means of providing in-service and in-situ monitoring of various critical structures. Due to their capability of long-range and through-the-thickness interrogation of the structures for small defects, guided waves have been studied extensively in damage detection for plate-like structures. </p><p> However, a few challenges exist when Lamb wave-based SHM/NDE techniques are employed. For example, the dispersion effect decreases the accuracy of many damage imaging algorithms; damage severity quantification is always a difficult problem. To provide possible solutions to above challenges, two damage imaging algorithms were developed and utilized for Lamb-wave based damage imaging. </p><p> The first algorithm is reverse-time migration (RTM), which was first used in geophysics to provide proper solutions to complex wave phenomena. The traditional imaging condition utilized in SHM is called excitation-time imaging condition, which used ray tracing and group velocity corresponding to the center frequency of the input signal. Due to the dispersion effect, the time-of-flight (ToF) estimation cannot always be accurate, especially for the situations that the Lamb waves propagate for a long distance. In this thesis, new imaging conditions are proposed to form enhanced zero-lag cross-correlation reverse-time migration (E-CCRTM) techniques. The proposed damage imaging technique takes into account the amplitude, phase, and all the frequency content of the Lamb waves propagating in the plate; thus, the severity of multiple sites of damage can be non-biasedly imaged regardless of the damage locations in comparison with using existing imaging conditions. The other imaging algorithm is called &lsquo;DORT-MUSIC&rsquo;. A Lamb wave-based, subwavelength imaging algorithm is developed for damage imaging in large-scale, plate-like structures based on a decomposition of the time-reversal operator (DORT) method combined with the multiple signal classification (MUSIC) algorithm in the space-frequency domain. The physics of wave propagation, reflection, and scattering that underlies the response matrix in the DORT method is mathematically formulated in the context of guided waves. Singular value decomposition (SVD) is then employed to decompose the experimentally measured response matrix into three matrices, detailing the incident wave propagation from the linear actuator array, reflection from the damage, and followed by scattering waves toward the linear sensing array for each small damage. The SVD and MUSIC-based imaging condition enable quantifying the damage severity by a &lsquo;reflectivity&rsquo; parameter and super-resolution imaging. </p><p> The two algorithms were also integrated with a hybrid system mainly comprised piezoelectric actuators mounted onto the structure and a laser Doppler vibrometer (LDV) for reception. The flexibility of the proposed system was used for inspection of various plate-like structures. The experimental results show that the 2-D E-CCRTM has robust performance to image and quantify multiple sites of damage in large area of the plate using a single PZT actuator with a nearby areal scan using LDV, and the DORT-MUSIC (TR-MUSIC) imaging technique can provide rapid, highly accurate imaging results as well as damage quantification with unknown material properties.</p>

<p> The ability to quantify particle mass and number concentrations in the exhaust stream of combustion engines during in-use operation is of critical importance for continuously monitoring and diagnosing the particulate matter removal efficiency of modern exhaust gas after-treatment systems. Extensive literature survey suggested a sensor operating on the diffusion-charging principle being optimally suited for particle measurements due to their proportional response towards particle surface area. This study was designed to determine and assess the possibility of quantifying particle emissions during on-road measurements using a prototype diffusion-charging type sensor. Such a sensor would not only allow for continuous monitoring capabilities of the exhaust particulate filters integrity, but moreover provide for a simplified tool to assess real-world particle number emissions to verify in-use emissions compliance of engines. </p><p> Evaluation of the sensor followed a three tier process, starting with fundamental sensor response analysis using a particle generator in order to develop and parameterize the underlying physical phenomena of the measurement principle. Next, examine the sensor in engine dynamometer experiments under controlled environment, and sampling from test vehicles during chassis dynamometer testing aimed at real-world like test conditions. Finally, the sensor was installed on vehicles while operated on the road over diverse driving conditions. This allowed for comparison to laboratory-grade measurement systems and the standard regulatory gravimetric particulate matter measurement method. The diffusion-charging type sensor employed in this study was observed to exhibit a response proportional to particle size <i>D_p</i>^1.09 and a measurement variability below 2% over consecutive tests. The sensor&rsquo;s sensitivity allowed for distinguishing between Diesel particulate filter efficiencies due to soot cake layer build-up on the substrate walls. In summary, the study concluded that the diffusion-charging type sensor provided a viable method to quantify in-use particle number emissions.</p>