Résumé en français non disponible

A base case modelling investigation was conducted to explore the chemical and physical behaviour of ground-level ozone (O3) and its precursor nitrogen dioxide (NO2) in Ontario using the U.S. Environmental Protection Agency (EPA) Community Multiscale Air Quality (CMAQ) model. Two related studies were completed to evaluate the performance of CMAQ in reproducing the behaviour of these species in both rural and urban environments by comparing to surface measurements collected by the Ontario Ministry of the Environment (MOE) network of air quality stations. The first study was a winter examination and the second study was conducted for a period during the summer of the same year. The municipality of North Bay was used to represent a rural setting given its smaller population relative to the city of Ottawa which was the base of the urban site. Statistical and graphical analyses were used to validate the model output. CMAQ was found to replicate the spatial variation of O3 and NO2 over the domain in both the winter and summer, but showed some difficulty in simulating the temporal allocation of the species. Validation statistics for North Bay and Ottawa showed overall O3 mean biases (MB) of 3.35 ppb and 2.25 ppb, respectively, and overall NO2 MB of -8.75 ppb and -4.37 ppb, respectively for the winter. Summer statistics generated O3 MB of 4.66 ppb (North Bay) and 10.05 ppb (Ottawa) while both MB for NO2 were between -2.20 ppb to -2.55 ppb. Graphical analysis showed that the model was not able to reproduce the lower levels of O3, especially at night, or the higher levels of NO2 during the day at the North Bay site for either season. This was expected since the comparisons were made between point measurements and 36 km grid-averaged model results. The presence of high amounts of NO2 emissions local to the monitoring sites compared to the levels represented in the emissions inventory may also be a contributing factor. The simulations for Ottawa demonstrated better agreement between model results and measurements as CMAQ provided a more accurate reproduction of both the higher and lower mixing ratios of O3 and NO2 during the winter and summer seasons. Results indicate that CMAQ is able to simulate urban environments better than rural ones.

Gallium nitride (GaN) has been targeted for use in high power (>30 W/mm) and high frequency (>160 GHz) application due to its wide band gap and its large break down field. One of the most significant advances in GaN devices has evolved from the AlGaN/GaN high electron mobility transistor (HEMT). As a result of the large power densities being applied to these devices there can develop intense hot spots near areas of highest electric field. The hot spot phenomenon has been linked to a decrease in device reliability through a range of degradation mechanisms. In order to minimize the effect that hot spot temperatures have on device reliability a detailed understanding of relevant transport mechanisms must be developed. This study focuses on two main aspects of phonon transport within GaN devices. The first area of focus was to establish an understanding of phonon relaxation times within bulk GaN. These relaxation times were calculated from an application of Fermi's Golden Rule and explicitly conserve energy and crystal momentum. This analysis gives insight into the details behind the macroscopic thermal conductivity parameter. Once relaxation times for GaN were established a multiscale phonon transport modeling methodology was developed that allowed the Boltzmann Transport Equation to be coupled to the energy equation. This coupling overcomes some computational limits and allows for nanoscale phenomena to be resolved within a macroscopic domain. Results of the transport modeling were focused on benchmarking the coupling method as well as calculating the temperature distribution within an operating 6 finger HEMT.

High strength secondary hardening lath martensitic steel is a strong candidate for high performance and reliable transmission systems in aircraft and automotives. The fatigue resistance of this material depends both on intrinsic microstructure attributes, such as fine scale (M2C) precipitates, and extrinsic attributes such as nonmetallic primary inclusions. Additionally, the aforementioned attributes are affected by processing history. The objective of this research is to develop a computational framework to quantify the influence of both extrinsic (primary inclusions and residual stresses) and intrinsic (martensite laths and carbides) microstructure attributes on fatigue crack formation and the early stage of microstructurally small crack (MSC) growth that dominate high cycle fatigue (HCF) lifetime. To model the fatigue response at various microstructure scales, a hierarchical approach is adopted. A simplified scheme is developed to simulate processing effects such as shot peening that is suitable to introduce representative residual stresses prior to conducting fatigue calculations. Novel strategies are developed to couple process route (residual stresses) and microstructure scale response for comprehensive analysis of fatigue potency at critical life-limiting primary inclusions in gear steels. Relevant microstructure-scale response descriptors that permit relative assessment of fatigue resistance are identified. Fatigue crack formation and early growth is highly heterogeneous at the grain scale. Hence, a scheme for physically-based constitutive models that is suitable to investigate crack formation and early growth in martensitic steel is introduced and implemented. An extreme value statistical/probabilistic framework to assess the influence of variability of various microstructure attributes such as size and spatial distribution of primary inclusions on minimum fatigue crack formation life is devised. Understanding is sought regarding the relative role of microstructure attributes in the HCF process, thereby providing a basis to modify process route and/or composition to enhance fatigue resistance. Parametric studies are conducted to assess the effect of hot isostatic pressing and introduction of compliant coatings at debonded inclusion-matrix interface on enhancement of fatigue resistance. A comprehensive set of 3D computational tools and algorithms for hierarchical microstructure-sensitive fatigue analysis of martensitic gear steels is developed as an outcome of this research; such tools and methodologies will lend quantitative and qualitative support to designing improved, fatigue-resistant materials and accelerating insertion of new or improved materials into service.

Helicopter performance is greatly influenced by its drag. Pylons, fuselage, landing gear, and especially the rotor hub of a helicopter experience large separated flow regions, even under steady level flight conditions the vehicle has been designed for, contributing to the helicopter drag. Several passive and active flow control concepts have been studied for reducing helicopter drag. While passive flow control methods reduce drag, they do so at one optimized design condition. Therefore, passive drag reduction methods may not work for helicopters that operate under widely varying flight conditions. Active flow control (AFC) methods overcome this disadvantage and consequently are widely being pursued. The present investigator has studied some of these AFC methods using computational fluid dynamics (CFD) techniques and has found synthetic (or pulsed) jets as one of the more effective drag reduction devices. Two bluff bodies, representative of helicopter components, have been studied and the mechanism behind drag reduction has been analyzed. It was found that the increase in momentum due to the jet, and a resultant reduction in the separated flow region, is the main reason for drag reduction in these configurations. In comparison with steady jets, synthetic jets were found to use less power for a greater drag reduction. The flow inside these synthetic jet devices is incompressible. It is computationally inefficient to use compressible flow solvers in incompressible regions. In such regions, using Lattice Boltzmann equations (LBE) is more suitable compared to solving the incompressible Navier-Stokes equations. The length scales close to the synthetic jet devices are very small. LBE may be used to better resolve these small length scale regions. However, using LBE throughout the whole domain would be computationally expensive since the grid spacing in the LBE solver has to be of the order of the mean free path. To address this need, a coupled Lattice Boltzmann-Navier-Stokes (LB-NS) methodology has been developed. The LBE solver has been successfully validated in a standalone manner for several benchmark cases. The solver has also been shown to be of second order accuracy. This LBE solver has been subsequently coupled with an existing Navier-Stokes (NS) solver. Validation of the coupled methodology has been done for analytical problems with known closed form solution. This LB-NS methodology is further used to simulate the flow past a cylinder where synthetic jet devices have been used to reduce drag. The LBE solver is used in the cavity of the synthetic jet nozzle while the NS solver is employed in the rest of the domain. The cylinder configuration was chosen to demonstrate drag reduction on helicopter hub shape geometries. Significant drag reduction is observed when synthetic jets are used, compared to the baseline no flow control case.

The thesis presents research on multiobjective optimization based on memetic computing and its applications in engineering. We have introduced a framework for adaptive multiobjective memetic optimization algorithms (AMMOA) with an information theoretic criterion for guiding the selection, clustering, and local refinements. A robust stopping criterion for AMMOA has also been introduced to solve non-linear and large-scale optimization problems. The framework has been implemented for different benchmark test problems with remarkable results. This thesis also presents two applications of these algorithms. First, an optimal image data hiding technique has been formulated as a multiobjective optimization problem with conflicting objectives. In particular, trade-off factors in designing an optimal image data hiding are investigated to maximize the quality of watermarked images and the robustness of watermark. With the fixed size of a logo watermark, there is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. We propose to use a hybrid between general regression neural networks (GRNN) and the adaptive multiobjective memetic optimization algorithm (AMMOA) to solve this challenging problem. This novel image data hiding approach has been implemented for many different test natural images with remarkable robustness and transparency of the embedded logo watermark. We also introduce a perceptual measure based on the relative Rényi information spectrum to evaluate the quality of watermarked images. The second application is the problem of joint spectrum sensing and power control optimization for a multichannel, multiple-user cognitive radio network. We investigated trade-off factors in designing efficient spectrum sensing techniques to maximize the throughput and minimize the interference. To maximize the throughput of secondary users and minimize the interference to primary users, we propose a joint determination of the sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is used again in the form of AMMOA. This algorithm learns to find optimal spectrum sensing times, decision threshold vectors, and power allocation vectors to maximize the averaged opportunistic throughput and minimize the averaged interference to the cognitive radio network. / February 2016

This thesis explores multiscale approaches to describe structural and chemical defects in metals. Particular emphasis is placed on investigating processes involving grain boundaries (GBs) in combination with impurity and vacancy defects. The defects and their interactions are calculated to very high accuracy using density functional theory (DFT) and connected to the macroscopic behavior within the two multiscale formalisms presented here. We begin with a sequential approach to address chemical embrittlement of nickel by sulfur impurities. Effects at both a \(\Sigma 5 (012)\) symmetric tilt GB and in the bulk are studied by considering competing mechanisms for ductile and brittle behavior. For the bulk, this takes the form of Rice’s theory, where the ratio of the surface and unstable stacking energy is used as a measure of ductility. This is generalized to the GB by considering GB sliding (GBS) and intergranular decohesion. Clear evidence that chemical embrittlement of nickel by sulfur is a GB driven effect is found. Next, a concurrent multiscale approach is described. A small region, containing the defects, is treated with Kohn-Sham DFT and coupled to the bulk, described with the embedded atom method. We apply this novel method to elucidate the chemical embrittlement of a copper \(\Sigma 5 (012)\) symmetric tilt GB. Intergranular decohesion for three substitutional impurities, bismuth, lead and silver, is investigated by considering the work of separation \((W_s)\) and the tensile strength \((\sigma_t)\). Bismuth and lead show a significant decrease in \(W_s\) and \(\sigma_t\), consistent with embrittlement, whilst silver has only a minor effect. Then, the concurrent multiscale method is applied to the process of GBS in copper. It is found that the resistance against sliding increases significantly for bismuth, lead and silver impurities. The underlying mechanisms for this increase are found to be dominated by size effects for bismuth and lead. For silver, chemical effects are of greater importance. Similar results are found for the underlying mechanisms of intergranular decohesion. The effect of a mono-vacancy on GBS is studied for copper. The multiscale approach enables improved decoupling of the mono-vacancy. It is found that the monovacancy enhances GBS by 22%. / Engineering and Applied Sciences

In recent years, concern has grown over the consequences of global warming. The arctic region is thought to be particularly vulnerable to increasing temperatures, and warming is occurring here substantially more rapidly than at lower latitudes. Consequently, assessments of the state of the Arctic are a focus of international efforts. For the terrestrial Arctic, large datasets are generated by remote sensing of above-ground variables, with an emphasis on vegetation properties, and, by association, carbon fluxes. However, the terrestrial component of the carbon (C) cycle remains poorly quantified and the below-ground distribution and stocks of soil C can not be quantified directly by remote sensing. Large areas of the Arctic are also difficult to access, limiting field surveys. The scientific community does know, however, that this region stores a massive proportion (although poorly quantified, soil C stocks for tundra soils vary from 96 to 192 Gt C) of the global reservoir of soil carbon, much of it in permafrost (900 Gt C), and these stocks may be very vulnerable to increased rates of decomposition due to rising temperatures. The consequences of this could be increasing source strength of the radiatively forcing gases carbon dioxide (CO2) and methane (CH4). The principal objective of this project is to provide a critical evaluation of methods used to link soil C stocks and fluxes at the usual scales spanned by the field surveys (centimetre to kilometre) and remote sensing surveys (kilometre to hundreds of kilometres). The soil C distribution of two sub-arctic sites in contrasting climatic, landscape/geomorphologic and vegetation settings has been described and analysed. The transition between birch forest and tundra heath in the Abisko (Swedish Lapland) field site, and the transition between mire and birch forest in the Kevo (Finnish Lapland) field site span several vegetation categories and landscape contexts. The natural variability of below-ground C stocks (excluding coarse roots > 2 mm diameter), at scales from the centimetre to the kilometre scale, is high: 0.01 to 18.8 kg C m-2 for the 0 - 4 cm depth in a 2.5 km2 area of Abisko. The depths of the soil profiles and the soil C stocks are not directly linked to either vegetation categories or Leaf Area Index (LAI), thus vegetation properties are not a straightforward proxy for soil C distribution. When mapping soil or vegetation categories over large areas, it is usually necessary to aggregate several vegetation or soil categories to simplify the output (both for mapping and for modelling). Using this approach, an average value of 2.3 kg C m-2 was derived both for soils beneath treeless areas and forest understorey. This aggregated value is potentially misleading, however, because there is significant skew resulting from the inclusion of exposed ridges (with very low soil C stocks) in the ‘treeless’ category. Furthermore, if birch trees colonise tundra heath and other ‘open’ plant communities in the coming decades, there will likely be substantial shifts in soil C stocks. This will be both due to direct climate effects on decomposition, but also due to changes in above- and below-ground C inputs (both in quantity and quality) and possibly changes in so-called root ‘priming’ effects on the decomposition of existing organic matter. A model of soil respiration using parameters from field surveys shows that soils of the birch forest are more sensitive to increases in mean annual temperature than soils under tundra heath. The heterogeneity of soil properties, moisture and temperature regimes and vegetation cover in ecotone areas means that responses to climate change will differ across these landscapes. Any exercise in upscaling results from field surveys has to indicate the heterogeneity of vegetation and soil categories to guide soil sampling and modelling of C cycle processes in the Arctic.

<p>The Discontinuous Galerkin time domain (DGTD) method is promising in modeling of realistic multiscale electromagnetic systems. This method defines the basic concept for implementing the communication between multiple domains with different scales.</p><p>Constructing a DGTD system consists of several careful choices: (a) governing equations; (b) element shape and corresponding basis functions for the spatial discretization of each subdomain; (c) numerical fluxes onto interfaces to bond all subdomains together; and (d) time stepping scheme based on properties of a discretized</p><p>system. This work present the advances in each one of these steps.</p><p> </p><p>First, a unified framework based on the theory of differential forms and the finite element method is used to analyze the discretization of the Maxwell's equations. Based on this study, field intensities (<bold>E</bold> and <bold>H</bold>) are associated to 1-forms and curl-conforming basis functions; flux densities (<bold>D</bold> and <bold>B</bold>) are associated to 2-forms and divergence-conforming basis functions; and the constitutive relations are defined by Hodge operators.</p><p>A different approach is the study of numerical dispersion. Semidiscrete analysis is the traditional method, but for high order elements modal analysis is prefered. From these analyses, we conclude that a correct discretization of fields belonging to different p-form (e.g., <bold>E</bold> and <bold>B</bold>) uses basis functions with same order of interpolation; however, different order of interpolation must be used if two fields belong to the same p-form (e.g., <bold>E</bold> and <bold>H</bold>). An alternative method to evaluate numerical dispersion based on evaluation of dispersive Hodge operators is also presented. Both dispersion analyses are equivalent and reveal same fundamental results. Eigenvalues, eigenvector and transient results are studied to verify accuracy and computational costs of different schemes. </p><p>Two different approaches are used for implementing the DG Method. The first is based on <bold>E</bold> and <bold>H</bold> fields, which use curl-conforming basis functions with different order of interpolation. In this case, the Riemman solver shows the best performance to treat interfaces between subdomains. A new spectral prismatic element, useful for modeling of layer structures, is also implemented for this approach. Furthermore, a new efficient and very accurate time integration method for sequential subdomains is implemented.</p><p>The second approach for solving multidomain cases is based on <bold>E</bold> and <bold>B</bold> fields, which use curl- and divergence-conforming basis functions, respectively, with same order of interpolation. In this way, higher accuracy and lower memory consumption are obtained with respect to the first approach based on <bold>E</bold> and <bold>H</bold> fields. The centered flux is used to treat interfaces with non-conforming meshes, and both explicit Runge-Kutta method and implicit Crank-Nicholson method are implemented for time integration. </p><p>Numerical examples and realistic cases are presented to verify that the proposed methods are non-spurious and efficient DGTD schemes.</p> / Dissertation

This work presents the first known attempt to model the dairy business from a multiscale modelling perspective. The multiscale nature of the dairy industry is examined with emphasis on those key decision making and process scales involved in production. Decision making scales identified range from the investor level to the plant operator level, and encompass business, production, plant, and operational levels. The model considers scales from the production manager to the unit operation scale. The cheese making process is used to demonstrate scale identification in the context of the important phenomena and other natural levels of scrutiny of interest to decision makers. This work was a first step in the establishment of a multiscale system model capable of delivering information for process troubleshooting, scheduling, process and business optimization, and process control decision-making for the dairy industry. Here, only material transfer throughout a process, use of raw materials, and production of manufactured product is modelled. However, an implementation pathway for adding other models (such as the precipitation of milk protein which forms curd) to the system model is proposed. The software implementation of the dairy industry multiscale model presented here tests the validity of the proposed: • object model (object and collection classes) used to model unit operations and integrate them into a process, • mechanisms for modelling material and energy streams, • method to create simulations over variable time horizons. The model was implemented using object oriented programming (OOP) methods in conjunction with technologies such as Visual Basic .NET and CAPE-OPEN. An OOP object model is presented which successfully enabled the construction of a multiscale model of the cheese making process. Material content, unit operation, and raw milk supply models were integrated into the multiscale model. The model is capable of performing simulations over variable time horizons, from 1 second, to multiple years. Mechanisms for modelling material streams, connecting unit operations, and controlling unit operation behaviour were implemented. Simple unit operations such as pumps and storage silos along with more complex unit operations, such as a cheese vat batch, were modelled. Despite some simplifications to the model of the cheese making process, the simulations successfully reproduced the major features expected from the process and its constituent unit operations. Decision making information for process operators, plant managers, production managers, and the dairy business manager can be produced from the data generated. The multiscale model can be made more sophisticated by extending the functionality of existing objects, and incorporating other scale partial models. However, increasing the number of reported variables by even a small number can quickly increase the data processing and storage demands of the model. A unit operation’s operational state of existence at any point of time was proposed as a mechanism for integrating and recalculating lower scale partial models. This mechanism was successfully tested using a unit operation’s material content model and is presented here as a new concept in multiscale modelling. The proposed modelling structure can be extended to include any number of partial models and any number of scales.