archives@tulane.edu / 1 / Joseph Skelton

Vertical axis wind turbine, developed as one of the main methods to utilize the wind energy, has a promising future; however, the major issue to limit its performance is the uneven loading on the blade during operation. Flow control mechanisms have been employed in the aerodynamic field to improve the performance of airfoils. In this study, two types of leading-edge structures, including flexible leading-edge and leading-edge roughness, are experimentally investigated to analyze their effects on altering the aerodynamic characteristics of NACA 0018 airfoil under steady flow condition and dynamic pitching condition. Current experimental results indicate that 1) during the steady flow condition, both of leading-edge structures contribute to the delay of the static stall; 2) for the dynamic pitching process, the leading-edge structures either delayed the dynamic stall angle or increased the area of the coefficient of pressure loop as a function of angle of attack.

This thesis presents the Deriche edge detector as an interesting alternative to the commonly used edge detectors. The Deriche edge detector's design is presented to the reader as well as its strengths and weaknesses. Performance issues of the Deriche edge detector are described in comparison with the Canny edge detector together with recommendations for using the Deriche detector. Finally, edge detection quality of the Deriche edge detector is compared to the Canny edge detector using robust subjective evaluation method.

A set S of vertices in a graph G is a connected dominating set of G if S dominates G and the subgraph induced by S is connected. We study the graphs for which adding any edge does not change the connected domination number.

A set S of vertices in a graph G is a total dominating set if every vertex of G is adjacent to some vertex in S. The minimum cardinality of a total dominating set of G is the total domination number of G. A graph is total domination edge addition stable if the addition of an arbitrary edge has no effect on the total domination number. In this paper, we characterize total domination edge addition stable graphs. We determine a sharp upper bound on the total domination number of total domination edge addition stable graphs, and we determine which combinations of order and total domination number are attainable. We finish this work with an investigation of claw-free total domination edge addition stable graphs.

An aeroacoustic investigation was performed to assess the effects of adding mass flow at the trailing edges of four stators upstream of an aircraft engine simulator. By using trailing edge blowing to minimize the shed wakes of the stators, the flow into the rotor was made more uniform. In these experiments a reduced number of stators (four) was used in a 1/14 scale model inlet which was coupled to a 4.1 in (10.4 cm) turbofan engine simulator with 18 rotors and 26 downstream stators. This study is a preliminary step toward a more in depth investigation of using trailing edge blowing to reduce unsteady rotor-stator interaction. Steady-state measurements of the aerodynamic flow field and acoustic far field were made in order to evaluate the aeroacoustic performance at three simulator speeds: 40%, 60%, and 88% of the design speed. The lowest test speed of 40% design speed showed the most dramatic reduction in radiated noise. Noise reductions as large as 8.9 dB in the blade passing tone were recorded at 40% design speed, while a tone reduction of 5.5 dB was recorded at 60% design speed. At 88% design speed a maximum tone reduction of 2.6 dB was recorded. In addition, trailing edge blowing reduced the overall sound pressure level in every case. For both the 40% design speed and the 60% design speed, the fan face distortion was significantly reduced due to the trailing edge blowing. The addition of trailing edge blowing from the four upstream stators did not change the total pressure ratio, and the mass flow added by the blowing was approximately 1%. The results of these experiments clearly demonstrate that blowing from the trailing edges of the stators is effective in reducing unsteady rotor-stator interaction and the subsequent forward radiated noise. / Master of Science

Periodic disturbance in CP33 habitat buffers is required to maintain early-successional plant communities for grassland birds. However, effects of disturbance on habitat suitability and nesting success of grassland birds are unknown. Furthermore, the grassland bird community and reproductive performance in CP33 habitat buffers may be influenced by the landscape context (e.g., edge habitat, surrounding land cover). My objectives for this study were to determine how periodic disturbance and the agricultural landscape mosaic influence the breeding grassland bird community in buffers, and how these same factors influence grassland bird nesting success and density in buffers. Data collected during line-transect surveys conducted in CP33 habitat buffers in northeast Mississippi during the 2007-2009 breeding seasons demonstrate periodic disturbance through prescribed burning and light strip-disking does not influence breeding bird diversity or density in buffers. Buffers with woodland at the non-crop edge, however, had the least grassland bird diversity. Dickcissels, one of the most abundant grassland bird species detected in buffers, had a lesser, though non-significant, density in woodland-bordered than in grassland-bordered buffers. I estimated nest success of Dickcissels and Red-winged Blackbirds using maximum likelihood approaches modeling daily survival rates of nests in Program MARK. Analysis results indicated periodic disturbance did not inform models of nest success for either species to an extent requiring inclusion in the best approximating model. Both species, however, had greater nest densities in control and burned buffers than in disked buffers, as well as adjacent to grassland and crop edges. Dickcissel nest success was associated negatively to proximity of developed areas and amount of land cover in crop production, and associated positively with nest height. Red-winged Blackbird nest success decreased with increasing distance from a crop edge and decreasing height of surrounding vegetation, but increased with greater amounts of ground cover. Results from this study will assist land managers with selecting land for enrollment in CP33 that will maximize producers’ conservation goals while minimally impacting crop production, as well as assist with development and refinement of USDA-NRCS Practice Standards, documentation of ecological benefits of federally subsidized conservation practices, and enhancement of wildlife benefits in agriculture-dominated landscapes.

The purpose of the edge beam is to support the railing and the pavement, function as part of the drainage system and in the case it is integrated into the bridge deck it can serve to distribute concentrated loads. It is located in road environment and therefore exposed to water and salt with chlorides as well as subject to impacts during accidents. It deteriorates in a greater pace than the rest of the bridge and therefore has a shorter lifespan than the bridge in full. A deteriorated edge beam put the safety of the bridge users in jeopardize and increases the need of maintenance, repair and replacement work. These activities affect the surrounding traffic flow due to reduced speed limits as well as closure of traffic lanes. A literature study has been performed to get an understanding of how edge beams are designed and constructed. A great part of this was done by examining codes and regulations. By meeting engineers from different building companies it has been possible to obtain a picture of how it is done in real life and how the path to the final design looks like. Building site visits were carried out to see the process from design to construction i.e. how it is applied in real life. A design study was performed, including a check of crack width in an integrated edge beam over a support, height of bridge deck when a pre-fabricated (brokappa) is used and a comparison in the magnitude of the clamping moment in a steel-concrete bridge with and without an edge beam. All proposals are presented by the Edge Beam Group (EBG, in Swedish, Kantbalksgruppen), which is composed of experienced engineers that works within the frame of the project social optimal edge beam systems governed by the Swedish Transport Administration. The literature research showed that even if the edge bean is prone to deteriorate its lifespan does not have to be governed by its condition. Planned expansion of bridge width and maintenance strategies including the replacement of waterproofing layer could also be a reason for replacement in some cases. A significant increase of reinforcement in the edge beam and top part of the bridge deck over support is needed to obtain an acceptable crack width of 0.15mm. This would however aggravate the casting phase. The use of a pre-fabricated edge beam result in an increase of the bridge deck height. A solution could be to strengthen the anchoring capacity but this could in turn give an over reinforced structure. When it comes to the clamping moment in a steelconcrete composite bridge the integrated edge beam leads to a better distribution of the traffic load. On the other hand, due to the higher dead weight, a bridge deck without an edge beam would result in a lower total moment in the cantilever.

This thesis explores the potential of edge devices in three applications: indoor localization, urban traffic prediction, and multi-modal representation learning. For indoor localization, we propose a reliable data transmission network and robust data processing framework by visible light communications and machine learning to enhance the intelligence of smart buildings. The urban traffic prediction proposes a dynamic spatial and temporal origin-destination feature enhanced deep network with the graph convolutional network to collaboratively learn a low-dimensional representation for each region to predict in-traffic and out-traffic for every city region simultaneously. The multi-modal representation learning proposes using dynamic contexts to uniformly model visual and linguistic causalities, introducing a novel dynamic-contexts-based similarity metric that considers the correlation of potential causes and effects to measure the relevance among images. To enhance distributed training on edge devices, we introduced a new system called Distributed Artificial Intelligence Over-the-Air (AirDAI), which involves local training on raw data and sending trained outputs, such as model parameters, from local clients back to a central server for aggregation. To aid the development of AirDAI in wireless communication networks, we suggested a general system design and an associated simulator that can be tailored based on wireless channels and system-level configurations. We also conducted experiments to confirm the effectiveness and efficiency of the proposed system design and presented an analysis of the effects of wireless environments to facilitate future implementations and updates. This thesis proposes FedForest to address the communication and computation limitations in heterogeneous edge networks, which optimizes the global network by distilling knowledge from aggregated sub-networks. The sub-network sampling process is differentiable, and the model size is used as an additional constraint to extract a new sub-network for the subsequent local optimization process. FedForest significantly reduces server-to-client communication and local device computation costs compared to conventional algorithms while maintaining performance with the benchmark Top-K sparsification method. FedForest can accelerate the deployment of large-scale deep learning models on edge devices.

Yes / Edge detection is widely believed to be an important early stage in human visual processing. However, there have been relatively few attempts to map human edge detection filters. In this study, observers had to locate a randomly placed step edge in brown noise (the integral of white noise) with a 1/𝑓2 power spectrum. Their responses were modelled by assuming the probability the observer chose an edge location depended on the response of their own edge detection filter to that location. The observer’s edge detection filter was then estimated by maximum likelihood methods. The filters obtained were odd-symmetric and similar to a derivative of Gaussian, with a peak-to-trough width of 0.1–0.15 degrees. These filters are compared with previous estimates of edge detectors in humans, and with neurophysiological receptive fields and theoretical edge detectors.