• Canada
• Publicationsclose
• Open Access close
• Thesis close
• FR close
• MY close

In recent years, aircraft manufacturers have been making progress in the design of more efficient aircrafts to reduce the environmental footprint. To attain this target, aircrafts manufactures work on the replacement of the hydraulic and bleed systems for electrical systems leading to a “More Electrical Aircraft”. However, the expected mass gain is a challenge, as previous technologies have been developed and optimized for decades. The new electrical solutions need to be look into detail to be competitive with previous technologies. All degrees of freedom must be considered, that is, new technologies and architectures. In particular, an HVDC network that reduces the number of rectifier stages seems a promising solution. From the HVDC network, the different three phase AC loads will be supplied by a series of power generic inverters. As the power consumption of the different loads change during the flight mission, the same inverter is used to supply different loads. The connection between the inverters and the loads is managed by a matrix of contactors. The proposed solution also considers redundant configurations, thus increasing system robustness. The design of the innovative system is presented in this document. That is, determining the optimal trade-off between the number of power inverters and the nominal power of each generic inverter that will also impact the size of the matrix of contactors. However, to assess the combinatory problem, the mass of the different components as a function of the nominal power needs to be calculated. A design environment is therefore created to perform automatic and optimized design of power converters. The different components are described using a “direct modelling” approach and coded using “object-oriented” programming. The components are validated experimentally or by numerical simulations. The different models are coupled to an optimization environment and to a frequency solver allowing a fast calculation of the steady-state waveforms. The optimization environment performs the precise design of the different parts of the power inverter: heatsink, power module, DC filter and coupling inductor. The power inverter is designed for different values of nominal power and switching frequency. The optimization assesses as well the usage of different technologies. Finally, the results are used to determine the optimal trade-off between the number of inverters and the nominal power of each inverter using a heuristic algorithm.

The exchange stiffness, Aex, is one of the key parameters controlling magnetization reversal in magnetic materials but is very difficult to measure, especially in thin films. We developed a new technique for measuring the exchange stiffness of a magnetic material based on the formation of a spin spiral within two antiferromagnetically coupled ferromagnetic films [1]. Using this method, I was able to measure the exchange stiffness of thin film Co alloyed with Cr, Fe, Ni, Pd, Pt and Ru. The results of this work showed that the rate at which a substituent element reduces the exchange stiffness is not directly related to its effect on the magnetization of the alloy. These measured trends have been understood by combining measurements of element specific magnetic moments obtained using X-ray magnetic circular dichroism (XMCD) and material specific modeling based on density functional theory (DFT) within the local density approximation (LDA). The experimental results also hint at significant reduction of the exchange stiffness at the interface that can account for the difference between our results and those obtained on bulk materials.

Cette thèse est consacrée à l'étude des processus stochastiques décentralisés. Parmi les exemples typiques de ces processus figurent la dynamique météorologique, la circulation automobile, la façon dont nous rencontrons nos amis, etc. Dans cette thèse, nous exploitons une large palette d'outils probabilistes permettant d'analyser des chaînes de Markov afin d'étudier un large éventail de ces processus distribués : modèle des feux de forêt (réseaux sociaux), balls-into-bins avec suppression, et des dynamiques et protocoles de consensus fondamentaux tels que Voter Model, 2-Choices, et 3-Majority. This thesis is devoted to the study of stochastic decentralized processes. Typical examples in the real world include the dynamics of weather and temperature, of traffic, the way we meet our friends, etc. We take the rich tool set from probability theoryfor the analysis of Markov Chains and employ it to study a wide range of such distributed processes: Forest Fire Model (social networks), Balls-into-Bins with Deleting Bins, and fundamental consensus dynamics and protocols such as the Voter Model, 2-Choices, and 3-Majority.

This thesis considers the character and development of clerical social criticism in England between c.1540 and c.1640. It draws principally on a number of sermons and treatises that offered critiques of the prevailing structures of wealth and power or exhortations to the fulfilment of charitable obligation. The paradigm through which these texts were constructed was that of ‘complaint’, a genre that was particularly vibrant in medieval discourse and in the sermons and ‘commonwealth’ tracts of the 1540s. It will be argued that rather than eschewing this tradition, late sixteenth-century preachers appropriated and refashioned its structures, themes and authorial positioning in response to far reaching economic, social and religious change.\ud \ud Particular aspects of socio-economic change, and of their effects on the clergy in particular, are examined in the introduction. Among the themes that are particularly germane to this thesis are the history of the enclosure movement; increasing commercialisation; and changing attitudes towards the poor. The first chapter assembles a number of printed texts in order to re-examine the trajectory of clerical complaint literature in the context of these developments. The second chapter considers the potential for social and political criticism in sermons preached at the county assizes, a sub-genre of ‘occasional’ sermons that until recently has received little attention from literary scholars or historians. The latter half of the thesis offers three case studies of selected sermons by three different authors. The intention of these chapters is primarily to examine the interaction between a text and its particular local context, although attention is also paid to broader social, political and discursive developments that help shed light on the historical meaning of these sermons.\ud \ud It is thus hoped that this study will contribute particularly to the ongoing interdisciplinary work of ‘contextualising’ the early modern English sermon and of reconstructing the role and status of the parish minister. Rather than a ‘voice in the wilderness’, it is concluded, the clerical moralist was an active agent in the discursive interpretation of economic change, and in the fashioning and communication of the reputation of powerful individuals.

The aim of this thesis is to expand on and improve the existing techniques used for detecting and identifying bacterial pathogens in clinical specimens with laser-induced breakdown spectroscopy (LIBS). Specifically, the existing experimental procedures, including bacterial sample preparation and data acquisition, as well as the data analysis with chemometric algorithms were investigated. Substantial reductions in LIBS background signal were achieved by implementing rigorous cleaning steps and the introduction of the use of ultrapure water. Following this, a database of LIBS spectra was acquired from specimens of E. coli, S. aureus, E. cloacae, M. smegmatis, and P. aeruginosa. The use of both discriminant function analysis (DFA) and partial least squares discriminant analysis (PLSDA) were compared. A PLSDA model built using the sum of all spectra acquired from 21 filters of E. coli and deionized water resulted in a sensitivity and specificity of 100% and 100%, respectively, in an external validation. To optimize the classification accuracy of the single-shot spectra for E. coli, E. cloacae, and S. aureus, outlier rejection schemes and data pre-processing methods were investigated. Classification errors of 30% motivated the use of artificial neural network analysis with principle component analysis pre-processing (PCA-ANN). The average sensitivity and specificity obtained using a randomized 80:20 split validation of the data was 94.18% and 97.01%, respectively. External validation was done on 52 filters of E. coli, E. cloacae, and S. aureus, giving an average sensitivity of 65.7%, and on 49 filters of E. coli, S. aureus, and M. smegmatis giving an average sensitivity of 87.2%. Samples of blood and urine were obtained from a hospital and spiked with the same species listed above. 98.9% sensitivity and 100% specificity were achieved for detection of bacteria in urine. 96.3% sensitivity and 98.6% specificity were achieved for detection of bacteria in blood. Discrimination using PCA-ANN on species in urine using an 80:20 split resulted in an average sensitivity and specificity of 97.2% and 98.6%, respectively. External validation on 16 filters gave an average sensitivity 77.5%. Applying PCA-ANN using an 80:20 split on species in blood resulted in 100% sensitivity and specificity. External validation on 19 filters gave an average sensitivity of 82.3%. These results indicate the potential usefulness of LIBS in the clinical setting.

Plans to reintroduce Atlantic salmon in Lake Ontario tributaries consists of stocking hatchery-reared fish yearly which will help to achieve a self-sustaining population. The issue with reintroduction remains in understanding the distribution of fishes after stocking. Environmental DNA (eDNA) provides a sensitive approach for monitoring that can offer inferences into fish distribution. I determined the distribution of stocked Atlantic salmon downstream from stocking sites using qRT-PCR and metabarcoding. I found that Atlantic salmon eDNA detection was more sensitive using qRT-PCR (51%) versus metabarcoding HTS (18.3%). However, metabarcoding provided data on fish community assemblages, which can help to monitor ecological interactions. I also found that eDNA and microsatellite markers genotyped and assigned an estimate number of individuals to 68.3% of the positive Atlantic salmon NGS data. This data indicates that eDNA and microsatellites can be used as a non-invasive method to quantify and monitor communities.

The term packing refers to the arrangement of multiple geometrical structures or shapes such as circles, squares, triangles, or polygons into a fixed and finite set of points. The geometric structures to be packed can also be trees and paths. Packing is also possible in a 3-dimensional space with geometric structures such as spheres, cylinders, and cubes. The concept of packing was introduced more than half a century ago. Since then, many researchers have studied the packing strategies of different geometric structures in different configurations of point-set. Packing strategies help to construct and arrange multiple geometric structures in a predetermined bounded space; hence, it can be classified as an optimization problem, as we are trying to allocate the optimal space for resources in a finite bounded space. The better the efficiency of the algorithm, the greater number of items that can be packed. Packing geometrical structures have applications in the storage, transportation, and transmission of objects in fields like automobile, aerospace, and naval industries. Since, in real-life scenarios, resources are finite, and space is limited; thus it raises the question, how to efficiently use limited space for accommodating multiple resources. However, packing multiple geometric structures can raise some design considerations. In our research, we have studied the packing of non-self-crossing, edge-disjoint plane spanning paths and have obtained some promising results. We further address some design considerations and provide a different approach on packing at least two non-self-crossing, edge-disjoint plane spanning paths into a point-set.

The internet has transformed the way we shop. Previously, most of our purchases came in the form of shopping trips to a nearby store. Now, it’s as easy as clicking a mouse. But with great convenience comes great responsibility. We have to be constantly vigilant about our personal information. This work presents a novel approach for online buyers to have a trusted, decentralized, privacy-preserved physical assets delivery solution. The proposed solution focuses on privacy-preserving personal information in delivering physical assets between sellers and buyers. Our primary approach is to prevent sellers and agents (responsible for ensuring the asset delivery is carried out) from accessing the buyer's shipping information. Those data include personally identifiable information such as full name, phone number, email address, etc. Our framework encrypts the information printed on the physical packages (containing personal information) and stores that information into a decentralized database as encrypted data rather than keeping them in centralized databases using technologies like Blockchain, Ethereum, smart contracts and a commit-reveal scheme. In our framework, we used Blockchain, smart contracts to increase the transparency, accessibility, and integrity of the data and the commit-reveal scheme to preserve the confidentiality of the data. We present, implement and assess a blockchain-based system using Ethereum smart contracts. We present detailed algorithms that explain the details of our smart contract, security, cost and performance analysis of the proposed method. Our work indicates that the proposed solution is economically attainable and provides data integrity, security, transparency, and traceability.

The development of deep learning algorithms has tremendously helped computer vision applications, image processing methods, Artificial Intelligence, and Natural Language Processing. One such application is image synthesis, which is the creation of new images from text. Recent techniques for text-to-image synthesis offer an intriguing yet straight forward conversion capability from text to image and have become a popular research topic. Synthesis of images from text descriptors has practical and creative applications in computer-aided design, multimodal learning, digital art creation, etc. Non-Fungible Tokens (NFTs) are a form of digital art that is being used as tokens for trading across the globe. Text-to-image generators let anyone with enough creativity can develop digital art, which can be used as NFTs. They can also be beneficial for the development of synthetic datasets. Generative Adversarial Networks (GANs) is a generative model that can generate new data using a training set. Diffusion Models are another type of generative model which can create desired data samples from the noise by adding random noise to the data and then learning to reverse the diffusion process. This thesis compares both models to determine which is better at producing images that match the given description. We have implemented the Vector-Quantized GAN (VQGAN) - Connecting Text and Images (CLIP) model. It combines the VQGAN and CLIP machine learning techniques to create images from text input. The diffusion model that we have implemented is Guided Language to Image Diffusion for Generation and Editing (GLIDE). For both models, we use text input from the MS-COCO data set. This thesis is an attempt to assess and compare the images generated using text for both models using metrics like Inception Score (IS) and Fréchet Inception Distance (FID). The semantic object accuracy score (SOA) is another metric that considers the caption used during the image generation process. We compute and compare the results for each label in the MS COCO data set. We highlight the potential causes of why the models may not be able to generate images through analysis of the results obtained. Our experimental results indicate that the GLIDE model outperforms the VQGAN - CLIP for our task of generating images from text.

Comprendre le fonctionnement des écosystèmes est un enjeu crucial, en particulier dans un contexte de changements globaux. Afin de mieux prédire les processus écosystémiques, j'ai testé la précision et les limites des hypothèses du biomass-ratio de Grime (HBMR) et de l'annulation idiosyncratique (HAI), cette dernière étant une hypothèse originale de cette thèse. Pour cela, j'ai appliqué le principe du biomass-ratio aux traits fonctionnels, en employant la méthode des traits agrégés en communauté, pour estimer la réponse globale des espèces en mélange. La décomposition des litières plurispécifiques constitue un bon modèle biologique, pour lequel je me suis posé les questions suivantes : (1) est-ce que l'HBMR prédit bien les taux de décomposition en mélanges plurispécifiques ? ; (2) est-ce que le degré de variabilité de ces taux diminue pour des raisons biologiques avec l'augmentation de la richesse spécifique (RS) des mélanges (HAI) ? ; (3) est-ce que la variabilité des taux entre mélanges diminue quand les conditions abiotiques du site deviennent plus limitantes ? ; (4)considérant que les mélanges plus contrastés fonctionnellement sont susceptibles de développer plus d'interactions, est-ce que la déviation à la prédiction augmente avec la dispersion fonctionnelle des mélanges(" FDis ", La liberté & Legendre 2010) ? Cette thèse inclut deux expériences de décomposition en sachets à litières : (1) à Sherbrooke (QC, Canada) avec des microcosmes, impliquant des litières de six espèces d'arbres, décomposant seules et en mélanges et (2) sur trois sites au climat contrasté dans la région de Clermont-Ferrand (France) avec des litières de quatre espèces d'herbacées, décomposant seules et en mélanges. Les résultats montrent des déviations positives et négatives par rapport aux taux prédits, mais l'HBMR décrit bien la réponse moyenne des litières plurispécifiques. Bien que l'HAI ait été rejetée, les résultats montrent une convergence des taux observés vers les taux prédits quand (1) la RS des mélanges augmente, (2)l'échelle spatiale augmente et (3) le climat est plus limitant pour la décomposition. Enfin, malgré des corrélations entre FDis et interactions entre espèces dans les litières, cette relation n'est pas généralisable et l'hypothèse de corrélation positive entre FDis et déviation à l'HBMR a été rejetée. Understanding ecosystem functioning is a key goal in ecology, especially in the context of global changes. To better predict ecosystem processes, I tested the accuracy and the limits of Grime's biomass-ratio (BMRH) hypothesis and a novel idiosyncratic annulment (IAH) hypothesis. I applied the biomass-ratio to functional traits, using the community-weighted means (CWM) to estimate the global response of species in mixtures. I studied the decomposition of litter species mixtures as a biological model and asked the following questions : (1) does the BMRH predict well the decomposition rates of mixed species litters ? ; (2) does the degree of variability of these rates decrease with increasing species richness (SR) beyond that expected from purely mathematical causes (IAH) ? ; (3) does the variability of rates between mixtures decrease with less favourable abiotic conditions for decomposition ? ; (4) as more functionally contrasted mixtures are expected to develop more interactions, does the deviation from prediction increase with increasing functional dispersion in mixtures (" FDis ", Laliberté & Legendre 2010) ? This study involves two decomposition experiments using litterbags : (1) at Sherbrooke (QC, Canada), in microcosms, involving litters from six tree species, decomposed alone and in mixtures and (2) in three climatically contrasted sites in the region of Clermont-Ferrand (France) with litters from four herbaceous species, decomposed alone and in mixtures. Despite both positive and negative deviations from expectation occurring at all levels of SR, the BMRH well described the average response of mixed species litters. Although I rejected the IAH, the results showed a convergence to the predicted values based on CWM with (1) increasing the SR in mixtures, (2) increasing the spatial scale of the study and (3) a less favourable climate to decomposition. Finally, although there was a correlation between litter interactions and functional divergence, this relationship was not generalizable and I rejected the hypothesis of a positive correlation between FDis and the deviations from BMRH.