Includes bibliographical references. / Fundamental research has been undertaken at the SASOL Advanced Fuels Laboratory to investigate the effects of the chemistry and physical properties of both conventional and synthetic jet fuels on threshold combustion. This research was undertaken using a purpose built low pressure continuous combustion test facility. Researchers at the laboratory now wish to examine these effects on an aviation gas turbine in service for which “off-map” scheduling of fuel to the engine would be required. A two phase project was thus proposed to develop this capability; the work of this thesis embodies Phase I of that project.

Includes bibliographical references. / This report sets out the results of a numerical investigation into the response of thin circular plates subjected to impulsive loads, using the commercial finite element code ABAQUS. iv Previous theoretical and numerical predictions of plate response have assumed a fully constrained boundary condition, while experiments have involved the use of both clamped and integral (fully built-in) boundary conditions. The current analysis employs 4-noded continuum elements in the finite element model, that allow the experimental boundary conditions to be modelled closely. Fully built-in plates are modelled by the inclusion of a material boundary, and clamped plates by the use of rigid clamping elements and a simple friction condition between the clamps and the plate surface. The inclusion of fillet radii at the integral boundary, and an edge radius at the clamped boundary, have been reported in additional experiments. These modifications are also modelled in the current investigation. The finite element model incorporates non-linear geometry and material effects, and strain rate sensitivity is included in the viscoplastic material definition. Impulsive loading is implemented through short duration pressure pulses, while the use of a uniform initial velocity profile is also shown to give good results. An explicit time integration scheme is used for the dynamic structural response.

Includes abstract. / Includes bibliographical references. / The world's dependency on fossil fuel will come to an end in the near future due to depletion of natural resources. Thus, research into alternative energy carriers becomes imperative. Furthermore, the use of fossil fuels is associated with environmental problems, which might be minimised by using alternatives. Hydrogen technology in the form of fuel cells can be a reliable and clean technology to minimise the problems associated with the use of fossil fuels. Fuel cells utilise hydrogen and air, and convert them to electrical energy through electro-catalysis with a co-product being water and heat.

This thesis explores the role of control, communication, and markets in the operation of smart buildings and microgrids. It develops models to study demand response (DR) alternatives in smart buildings using different communication and control protocols in building management systems. Moreover, it aims at understanding the extent to which smart buildings can provide regulation service reserves (RSR) by real time direct load control (DLC) or price-based indirect control approaches. In conducting a formal study of these problems, we first investigate the optimal operational performance of smart buildings using a control protocol called packetized direct load control (PDLC). This is based on the notion of the energy packet which is a temporal quantization of energy supplied to an appliance or appliance pool by a smart building operator (SBO). This control protocol is built on top of two communication protocols that carry either complete or binary information regarding the operation status of the appliances in the pool. We discuss the optimal demand side operation for both protocols and analytically derive the performance differences between them. We analyze the costs of renewable penetration to the system's real time operation. In order to strike a balance between excessive day-ahead energy reservation costs and stochastic real time operation costs, we propose an optimal reservation strategy for traditional and renewable energy for the PDLC in both the day-ahead and the real time markets to hedge the uncertainty of real time energy prices and renewable energy realization. The second part of the thesis proposes systematic approaches for smart buildings to reliably participate in power reserve markets. The problem is decomposed into two parts in the first of which the SBO starts by estimating its prior capacity of reserve provision based on characteristics of the building, the loads, and consumer preferences. We show that the building's reserve capacity is governed by a few parameters and that there is a trade off for smart buildings to provide either sustained reserve or ramping reserve. Based on the estimated capacity, we propose two real time control mechanisms to provide reliable RSR. The first is a DLC framework wherein consumers allow the SBO to directly modulate their appliances' set points within allowable ranges. We develop a feedback controller to guarantee asymptotic tracking performance of the smart building's aggregated response to the RSR signal. The second is a price controlled framework that allows consumers to voluntarily connect and consume electricity based on their instantaneous utility needs. Consumers' time varying dynamic preferences in providing RSR are studied by Monte Carlo simulation, in which such preferences are characterized by sufficient statistics that can be used in a stochastic dynamic programming (DP) formulation to solve for the optimal pricing policy.

Mobility on Demand (MoD) systems utilize shared vehicles to supplement or replace mass transit and private vehicles. Such systems include traditional taxis as well as Transportation Network Companies (TNCs) that offer bike and ride sharing. MoD systems face myriad operational challenges, but this dissertation focuses on the data-driven load balancing problem of redistributing vehicles among service regions. This is a difficult resource reallocation problem because customer demands follow a stochastic process subject to dynamic temporal-spatial patterns. The first half of this dissertation considers the load balancing problem for a bike sharing system in which bikes are redistributed among stations via trucks. The objective is to avoid situations in which a user wishes to rent (return) a bike to a station but cannot because the station is empty (full). First, a station and interval-specific inventory level is defined as a function of station capacity and interval demand rates as observed from analyzed data. Second, using a graph network framework, a receding horizon controller is proposed to determine the optimal paths -- over a short period of time -- for the fleet of trucks to take. When calculating the optimal paths the controller considers the current and projected inventory subject to the dynamically changing rent and return rates for every station in the network. The second half of this dissertation tackles the redistribution of an autonomous taxi fleet in which the vehicles themselves are capable of performing load balancing operations across service regions. The objective is to minimize the fraction of customers whose demands are dropped due to vehicle unavailability as well as the fraction of time the vehicles spend on load balancing operations (i.e driving empty). The system is represented by a queuing model and, as such, dynamic programming can find the optimal solution; however, the state-space of the model grows quickly rendering all but a minuscule system impossible to solve. To this end a parametric control is proposed that uses thresholds to dictate redistribution actions and well performing parameters are found via concurrent estimation methods of simulation.

Top-down fabrication is widely used for producing micro to nanometer scale features for biomedical applications. Opposed to bottom-up approach, it is superior to control geometry, structure and composition of the product. Traditional top-down fabrication techniques usually suffer from utility of clean room, harsh processing procedures and high cost. This dissertation is focused on developing a novel top-down fabrication technique based on soft lithography to fabricate functional microdevices and micropatterns for cell-borne drug delivery, biosensing and cell-tracking. The first chapter overviews the major top-down fabrication methods and their applications in biomedical science, as well as introduces recent advances in cell-borne drug delivery, biosensing and cell-tracking. The second chapter reports on the development of a novel type of particulate structures called microdevices for cell-borne drug delivery. The microdevices were fabricated by soft lithography with a disklike shape. Each microdevice was composed of a layer of biodegradable thermoplastic. One face of the thermoplastic layer was covalently grafted with a cell-adhesive polyelectrolyte such as poly-L-lysine. This asymmetric structure allowed the microdevices to bind to live cells through bulk mixing without causing cell aggregation. Moreover, the cell−microdevice complexes were largely stable, and the viability and proliferation ability of the cells were not affected by the microdevices over a week. In addition, sustained release of a mock drug from the microdevices was demonstrated. This type of microdevice promises to be clinically useful for sustained intravascular drug delivery. The third chapter extended the work in the second chapter to fabricate enzyme-laden microdevices for cell-borne enzyme delivery. Enzymes have been used to treat various human diseases and traumas. However, therapeutic utility of free enzymes is impeded by their short circulation time, lack of targeting ability, immunogenicity, and inability to cross biological barriers. Cell-mediated drug delivery approach offers unique capability for overcoming these limitations but the traditional cell-mediated enzyme delivery techniques suffer from drawbacks such as risk of intracellular degradation of and low loading capacity for the payload enzyme. This chapter presents development of a novel cell-mediated enzyme delivery technique featured by the use of micrometer-sized disk-shaped particles termed microdevices. The microdevices are fabricated by layer-by-layer assembly and soft lithography with catalase being used as a model therapeutic enzyme. The amount of catalase in the microdevices can be controlled with number of catalase layers. Catalase in the microdevices is catalytically active and active catalase is slowly released from the microdevices. Moreover, cell-microdevice complexes are produced by attaching the catalase-laden microdevices to the external surface of both K562 cells and mouse embryonic stem cells. This technique is potentially applicable to other enzymes and cells, and promises to be clinically useful. The fourth chapter describes fabrication of disk-shaped microdevices containing densely packed Carbon nanotubes (CNTs) for Raman labelling of macrophages. Capability to detect or track macrophages in vivo is important for developing macrophage-based therapies. Dispersed carbon nanotubes (CNTs) have been used for Raman labelling of cells and a top-down approach has been developed to fabricate disk-shaped microparticles for the same application. The fabrication is featured by the use of spray coating of CNTs to produce the microdevices. Raman detection of a single microdevice at a centimeter-scale working distance and the feasibility of using chemically modified CNTs for multiplexed Raman labelling were demonstrated. Macrophages were stably labelled with the microdevices by simply adding the microdevices to cultivated macrophages. The labelling slightly reduced viability of the macrophages and the labelled macrophages retained their ability to capture foreign particles. Moreover, Raman detection of a single macrophage was demonstrated. This technique promises to be useful for detection or tracking of macrophages in vivo as well as for other biomedical applications. The fifth chapter presents a simple strategy to construct a micrometer-sized self-referenced fluorescence sensor for detecting Cu2+. The method relies on microcontact printing of bovine serum albumin-stabilized gold nanocluster (BSA-AuNC) and poly(propyl methacrylate) (PPMA) stripes on a glass slide. The PPMA stripes are printed on the BSA-AuNC stripes to form a crossbar array, with its cell unit being composed of four distinct regions: BSA-AuNC, plain glass, PPMA, and BSA-AuNC covered by PPMA. The BSA-AuNC region is fluorescent and its fluorescence intensity is changeable upon contacting with analyte solution. The BSA-AuNC covered by PPMA is also fluorescent but insensitive to the analyte solution due to the presence of PPMA which prevents the analyte solution from contacting the BSA-AuNC. This region can thus be used as an internal reference for sensing. This self-referenced sensor is able to detect Cu2+ in a highly specific and concentration-dependent manner. The sixth chapter summarizes the major achievements of abovementioned studies. / A Dissertation submitted to the Department of Chemical and Biomedical Engineering in partial fulfillment of the Doctor of Philosophy. / Fall Semester 2016. / September 19, 2016. / Biosensing, Cell-borne Drug Delivery, Cell-tracking, Fabrication, Soft lithography / Includes bibliographical references. / Jingjiao Guan, Professor Directing Dissertation; Steven Lenhert, University Representative; Yan Li, Committee Member; Samuel C. Grant, Committee Member.

The research investigated the effect of aging altered driving posture on injury criteria of the head (HIC15) and chest (peak resultant chest acceleration) in low speed frontal impact. The aging altered driving postures were described in terms of body joint angles. Older drivers exhibited smaller body joint angles meaning older drivers tended to sitting closer to the steering wheels compared to their younger counterparts. A simplified anthropomorphic test dummy, sled rig, and impact pendulum were constructed as the necessary testing apparatus for the low speed frontal impact sled test experiment. A fractional factorial design screened out the significant factors that affect HIC15, and peak resultant chest acceleration. The results showed that for HIC15, torso angle, hip angle, right knee angle, right ankle angle, and seatbelt pillar loop anchor height were the significant factors. For peak resultant chest acceleration torso angle, right knee angle and left ankle angle were significant. Qualitatively, extended torso angles, meaning sitting more laid back, increased both HIC15 and peak resultant chest acceleration. Left and right side of lower extremities altered HIC15 and peak resultant chest acceleration separately. These findings may help to better understand biomechanical response of older drivers during car accidents and provide useful information regarding safer vehicle design, such as occupant seat and seatbelt in an effort to mitigate injuries. / A Thesis submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the Master of Science. / Fall Semester 2016. / November 22, 2016. / driving posture, low speed frontal impact, sled test / Includes bibliographical references. / Sungmoon Jung, Professor Directing Thesis; John O. Sobanjo, Committee Member; Eren Erman Ozguven, Committee Member.

In DSP applications such as speech, image and video processing, receiving the output signal values in order is necessary for faster processing. For the same reason, Decimation-In-Time Fast Fourier Transform (DIT-FFT) is a method of implementation where the output signals are received in sequential order which avoids the extra circuitry required for re-ordering at the receiver. Modular FFT and Conventional FFT algorithms are methods of implementations to compute the Discrete Fourier Transform (DFT). An N-point Fast Fourier Transform requires v = log[superscript N][subscript r] butterfly stages to compute, for radix r. Implementation of the Modular Pipeline FFT (MP-FFT) algorithm differs from the Conventional FFT in terms of the storage of center elements after the v/2 stage. The Modular method of implementation reduces the computational requirements by nearly half without significant change in performance. We present a comparison between a Conventional FFT and Modular Pipeline FFT implementations in terms of the number of computations, latency and hardware utilization, which are substantiated by our implementations using Xilinx Virtex 5, Virtex 6 devices and Quartus Stratix IV, Stratix V devices. The output simulations are performed using Modelsim software. As the size of FFT increases from 16 to 1024 points, the efficiency in terms of number of multiplications required increases from 21.8% to 31.23% for Radix-2 and from 12.5% to 25% for Radix-. Estimated delay to compute the Modular algorithm shows an increased efficiency of 37.7% for Radix-2 and 24.08% for Radix-4 implementation when compared to Conventional FFT. Keywords: DFT, FFT, DIT-FFT, MP-FFT / A Thesis submitted to the Department of Electrical & Computer Engineering in partial fulfillment of the Master of Science. / Spring Semester 2017. / April 12, 2017. / Includes bibliographical references. / Linda DeBrunner, Professor Directing Thesis; Victor DeBrunner, Committee Member; Bruce A. Harvey, Committee Member.

The in-plane torsion test is a well-established test used for the characterisation of sheet metals. The specimen is intended to deform in planar simple shear and is designed to be machined with a continuous annular shear zone. As a result, there are no “edge effects” or geometric discontinuities to generate instabilities, thus large true strains up to 1 can be achieved. Before this research, the specimen had only been used for material characterisation in the quasi-static regime. The aim of this research was to conduct further quasi-static testing using the in-plane torsion test and to extend its use into the dynamic regime. Quasi-static tests were performed on a quasi-static torsional (QST) system that was designed to be integrated onto a Zwick universal testing machine. Dynamic tests were performed on a modified torsional split Hopkinson bar (TSHB) system. The TSHB system adopted a nested configuration which allowed for a longer incident bar, and thus larger obtainable strains. Two quick-release mechanisms were used, one using a novel reusable wedge and the other using fracture-pins. All specimens were manufactured from Al 1050 H14. Typical results agreed with material test data available in the literature. Both systems attained large strains at near-constant strain rates and together, allowed for material characterisation over a large range of strain rates. Near-uniform deformations were observed for specimens with lower strain gauge widths. An added feature of the specimen was the flat reverse face, which together with the nested configuration of both systems allows for the possibility for full-field DIC measurement in the future. An estimation method for steady-state flow stress is presented with the steady-state flow stress found to be rate dependant. Finally, a relationship between the steady-state flow stress and strain rate for all experimental results is proposed.

For robotics to be useful to the public in a multifaceted manner, they need to be both legged and agile. The legged constraint arises as many environments and systems in our world are tailored to ablebodied adults. Therefore, a practically useful robot would need to have the same morphology for maximum efficacy. For robots to be useful in these environments, they need to perform at least as well as humans, therefore presenting the agility constraint. These requirements have been out of reach of the field until recently. The aim of this thesis was to design a planar monopod robot for rapid acceleration manoeuvres, that could later be expanded to a planar quadruped robot. This was achieved through a hybrid electric and pneumatic actuation system. To this end, modelling schemes for the pneumatic cylinder were investigated and verified with physical experiments. This was done to develop accurate models of the pneumatic system that were later used in simulation to aid in the design of the platform. The design of the platform was aided through the use of Simulink to conduct iterative testing and multivariate evaluations using Monte Carlo simulation methods. Once the topology of the leg was set, the detail design was conducted in Solidworks and validated with its built in simulation functions. In addition to the mechanical design of the platform, a specialist boom was designed. The design needed to compensate for the forces the robot exerts on the boom as well as the material constraints on the boom. This resulted in the development of a cable-stayed, four bar mechanism boom system. An embedded operating system was created to control the robot and take in and fuse sensor inputs. This was run using multiple sensors, sub-controllers and microcontrollers. Sensor fusion for the system was done using a Kalman Filter to improve readings and estimate unmeasured states of the robot. This Kalman Filter took LiDAR and accelerometer readings as inputs to the system to produce a subcentimetre accurate position measure for the system. Finally, the completed platform was validated using fixed-body forward hopping tests. These tests showed a significant degree of similarity to the simulated results and therefore validated the design process.