In this paper we develop a computational procedure to investigate linear fracture of two-dimensional problems in isotropic linearly elastic media. A symmetric Galerkin boundary element method (SGBEM), based on a weakly singular, weak-form traction integral equation, is adopted to model these fractures. In particular we consider multiple interacting cracks in an unbounded domain subject to internal pressure and remote stress. The growth of the cracks is driven by either linearly dependent injection pressures or volumes in each crack. A variety of crack geometries are investigated. / text

Eigenmannia virescens is a gymnotiform weakly electric fish which uses a quasi-sinusoidal electric organ discharge (EOD) to sense their environment. EOD frequency (EODF) is individual-specific. In conspecific interactions, each fish perceives the EODF of the conspecific as a periodic amplitude modulation (AM) of their own discharge. When both fish are stationary, the depth of this AM is constant, but it varies when fish are swimming. We hypothesized that AM variations during swimming act as a noise source that could have no effect on, hinder, or enhance EODF identification. To test this, we quantified the jamming avoidance response (JAR) (a natural behaviour wherein fish are required to accurately determine one another’s EODF) in response to stimuli of varying depths of noise. These experiments demonstrated that swimming noise does not impair the ability of E. virescens to identify conspecific EODF, and actually improves its ability to detect the presence of a neighbouring fish.

Detonation waves are supersonic combustion waves that have a complex three-dimensional cellular structure. There is growing experimental evidence that the cellular structure of detonations promotes their propagation in the presence of losses. In spite of that, the conventional model for the detonation structure, known as the Zeldovich - Von Neumann - Doring (ZND) model, neglects the existence of cellular structure for detonations and assumes the wave to consist of a strong leading planar shock coupled with trailing chemical reactions. Therefore, the influence of cellular structure on the dynamics and extinction limits of detonation waves has been of particular interest. Previous studies have investigated the influence of cellular structure on the dynamics of detonations with mass divergence in the framework of narrow tubes, porous-walled tubes and weak confinement. However, precise quantification of the loss mechanism in these frameworks has been associated with some difficulties. Complex flow in the boundary layers, inherent in thin tubes, or attenuation of the transverse waves in the porous-walled tubes has made the evaluation of the loss mechanism more difficult in such geometries. In this thesis, a novel well-posed problem is formulated for detonations with mass divergence. It is shown that detonations propagating in a channel with a cross-section area increasing exponentially have a constant mass divergence. The detonations were found to propagate at a quasi-steady speed below the ideal Chapman-Jouguet velocity. This permitted to make meaningful comparison with the theoretical models and simulations. The experiments were performed in two mixtures, one displaying characteristic weakly unstable detonations (2C2H2 + 5O2 + 21Ar), and the other displaying highly unstable detonations (C3H8 + 5O2). The dependence of the velocity deficits and limits on the amount of mass divergence for the two mixtures were compared with the predictions of the quasi-one-dimensional ZND model with lateral mass divergence. Since the ZND model neglects the cellular structure of the detonations, such comparison permitted to asses the influence of cellular structure on the dynamics of detonations with mass divergence. Comparisons were also made with the results of simulations of inviscid cellular detonations. These comparisons showed that the velocity deficits and critical rate of mass divergence in the weakly unstable mixture were reasonably well predicted by the quasi-one-dimensional model. For smaller values of mass divergence rate, a good agreement between the experiments and the predictions of the two-dimensional cellular simulations was observed for the weakly unstable mixture. For the highly unstable detonations, the quasi-one-dimensional model significantly over-predicted the effect of mass divergence.Detonations were observed for rates of mass divergence 93% higher than the critical predicted value, displaying more substantial velocity deficits than predicted. Such observations show conclusively that the ZND model cannot capture the dynamics of highly unstable detonations on large scales.

The dorsodorsal division (DD) of the teleost telencephalon has been implicated in memory processes similar to those associated with the mammalian hippocampus. The network connectivity and neural activity underlying this involvement have remained unclear. This thesis attempts to elucidate both. Attempts have been made to record the neural activity of DD neurons, but none have succeeded in correlating the recorded firing with any meaningful stimuli. In this thesis, I present single-unit electrophysiological recordings of DD neurons that reveal persistent activity in the form of up-states which are evoked by two modalities of naturalistic sensory stimuli – visual and electrosensory. The anatomy of DD was a little better understood than the neural activity. Recent anatomical work has shown that DD is strongly inter-connected with the cortical-like dorsolateral division (DL) of the pallium, re-inforcing its similarity to mammalian hippocampal structures. This same work has also revealed much of DD’s extrinsic connectivity. It was not, however, of a resolution fine enough to disambiguate the connections of the various DD subregions, nor to clarify the existence and structure of intrinsic DD connectivity. In this thesis, I further elucidate the connectivity of DD, by isolating its subregions. This was done by means of very small and precise neurotracer injections. These injections revealed strong recurrent connectivity within individual DD subregions, multiple pathways between DD and DL, and striking similarities between the connectivities of DL and DD and those of the mammalian dentate gyrus and CA3 respectively. From the results of these investigations I propose a model of homology between the teleost DD-DL loop and the putative pattern separation and completion networks contained in the mammalian cortico-hippocampal circuitry, as well as a role for the observed persistent activity in DD within this model. I further propose the dorsal teleost telencephalon as an excellent model system for the further study of the network mechanisms of pattern separation and completion.

abstract: Feature embeddings differ from raw features in the sense that the former obey certain properties like notion of similarity/dissimilarity in it's embedding space. word2vec is a preeminent example in this direction, where the similarity in the embedding space is measured in terms of the cosine similarity. Such language embedding models have seen numerous applications in both language and vision community as they capture the information in the modality (English language) efficiently. Inspired by these language models, this work focuses on learning embedding spaces for two visual computing tasks, 1. Image Hashing 2. Zero Shot Learning. The training set was used to learn embedding spaces over which similarity/dissimilarity is measured using several distance metrics like hamming / euclidean / cosine distances. While the above-mentioned language models learn generic word embeddings, in this work task specific embeddings were learnt which can be used for Image Retrieval and Classification separately. Image Hashing is the task of mapping images to binary codes such that some notion of user-defined similarity is preserved. The first part of this work focuses on designing a new framework that uses the hash-tags associated with web images to learn the binary codes. Such codes can be used in several applications like Image Retrieval and Image Classification. Further, this framework requires no labelled data, leaving it very inexpensive. Results show that the proposed approach surpasses the state-of-art approaches by a significant margin. Zero-shot classification is the task of classifying the test sample into a new class which was not seen during training. This is possible by establishing a relationship between the training and the testing classes using auxiliary information. In the second part of this thesis, a framework is designed that trains using the handcrafted attribute vectors and word vectors but doesn’t require the expensive attribute vectors during test time. More specifically, an intermediate space is learnt between the word vector space and the image feature space using the hand-crafted attribute vectors. Preliminary results on two zero-shot classification datasets show that this is a promising direction to explore. / Dissertation/Thesis / Masters Thesis Computer Engineering 2019

No description available.

Side-polished fiber allows access to the evanescent field propagating in the cladding of a few-mode fiber. This cladding mode is analyzed and experimentally validated to further the design of a novel class of fiber optic devices. To do this, specific modes are excited in the polished fiber using a phase-only spatial light modulator to determine spatial mode distribution. Each mode is excited and compared to the expected field distribution and to confirm that higher order modes can propagate through side-polished fiber. Based on each mode’s distribution, a side-polished fiber can be designed so that perturbations on the polished portion of the fiber effect each mode independently. By carefully analyzing the effects of identical perturbations on each mode, it is determined that each mode can be isolated based on the geometry of the polished fiber and careful alignment of the mode field. This research has the potential to advance the development of novel fiber-based sensors and communications devices utilizing mode-based interferometry and mode multiplexing. / M.S. / Fiber optic devices have seen significant advancement since the realization of the laser and low-loss optical fiber. Modern day fiber optics are commonly utilized for high-bandwidth communications and specialized sensing applications. Utilizing multiple modes, or wave distributions, in a fiber provides significant advantages towards increasing bandwidth for communications and provides potential for more accurate sensing techniques. Significant research has been conducted in both the sensing and communication field, but mode-domain devices have the capability to significantly advance the field of fiber optic devices. This thesis demonstrates the potential for side-polished fiber geometry to effect each mode independently, thus allowing side-polished fiber to be utilized for realizing novel devices such as multiplexing devices and fiber optic sensors.

Simplifications regarding calculations of fatigue life due to vibrational loading is based on weakly stationarity assumptions which is a time independent method. The hypothesis was based on the uncertainty of these assumptions. The aim of this study was to examine whether the analysed data fulfilled the assumptions of weakly stationarity. It was determined that the assumption was not valid for most signals and a comparison of time dependent methods should be performed to evaluate the difference compared with the time independent method. Two time dependent methods were constructed and implemented on the signals based on the results of performed stationarity tests. The result determined that a decrease in fatigue life of an investigated weld might occur for the two time dependent methods compared with the time independent method. The method which was considered to produce the most accurate results was also the most constrained as to the amount of data that fulfilled its requirements. A conclusion was drawn that signals containing more data was necessary to achieve conclusive results of the fatigue life. The hypothesis was proven to be mostly true since most of the analysed signals were found to be piecewise weakly stationary.

In this thesis we are studying some properties of weakly sequential homeomorphisms between Banach spaces. First, we show some results that summarize how are some clas- ses of Banach spaces (specifically separable spaces, spaces with separable dual, Asplund spaces, reflexive spaces, weakly compactly generated spaces and spaces not containing isomorphic copy of ℓ1) determined by weak topology of the space. Then we show that to preserve some properties (separability, reflexivity and being weakly compactly gene- rated) it is enough for the spaces to be weakly sequentially homeomorphic. Furthermore we show that if two spaces are weakly sequentially uniformly homeomorphic then one contains isomorphic copy of ℓ1 if and only if the other spaces has this property. Finally we construct weakly sequential homeomorphisms between some class of Banach spaces.

The weakly electric fish Sternopygus macrurus emits an electric organ discharge (EOD) composed of a series of pulses. The EOD pulse is mainly shaped by sodium currents. There are two sodium channel α subunits orthologs of the mammalian Nav1.4 expressed in the EO of Sternopygus. Previous studies identified a novel splice variant of the Nav1.4b (Nav1.4bL), in which an extra 51-amino acid occurs in the N terminal end. Nav1.4bL currents inactivate and recover from inactivation significantly faster than Nav1.4bS. The voltage-dependence of steady-state inactivation of smNav1.4bL shifts to hyperpolarized potential. Structural analysis predicts an α-helix in the middle of the extended N terminus. Removal of a proline right after the α-helix significantly slows down current decay but has no effect on channel recovery from inactivation, suggesting inactivation and recovery have independent mechanism. Mutagenesis analysis of the extended N terminus showed that the short helical region, especially the positive charges in the helix, is an important determinant for channel voltage-dependence of steady-state inactivation. However, other residues outside the helical region are required for regulation of fast inactivation and recovery form inactivation. Functional and structural analysis provides evidence for the importance of the C terminus in fish Nav1.4b channel properties. Chimera in which the C terminus of smNav1.4bS was substituted by the human Nav1.4 C terminus, shows an 11 mV positive shift in voltage-dependence of activation and a -16 mV negative shift in inactivation. Deletion of the distal half of smNav1.4bS negatively shifted voltage-dependence of inactivation and significantly accelerated channel recovery from inactivation. In the deletion mutant, the regulation by the N segment is missing. Substitution of the C terminus mutant retains wild type channel inactivation and recovery properties and can be regulated by N segment again. My study provides evidence that the extended N terminus of smNav1.4bL binds the distal part of C terminal tail to modulate channel inactivation properties. This is the first time to show the distal C terminus is involved in channel recovery from inactivation. Studies in the fish sodium channel properties provide useful information to understand function and structure of voltage-gated sodium channels. / text