Le comportement thermique et thermo-mécanique des pieux énergétiques est étudié par plusieurs approches : mesures au laboratoire sur des éprouvettes de sol, modélisation physique en modèle réduit, expérimentations sur pieu en vraie grandeur, et calculs numériques/analytiques. D’abord, la conductivité thermique d’un loess à l’état non saturé est mesurée en fonction de la teneur en eau et de la succion. Les résultats montrent une relation univoque entre la conductivité thermique et la teneur en eau pendant un cycle d’humidification/séchage alors qu’une boucle d’hystérésis est observée pour la relation entre la conductivité thermique et la succion. Deuxièmement, des essais thermiques sont réalisés sur un pieu énergétique expérimental en vraie grandeur pour étudier le transfert thermique à l’échelle réelle. Troisièmement, une solution analytique est proposée pour simuler la conduction thermique d’un pieu énergétique vers le sol environnant pendant un chauffage. Les tâches mentionnées ci-dessus concernant le comportant thermique sont ensuite complétées par des études sur le comportement thermo-mécanique des pieux énergétiques. D’un côté, des expérimentations sont réalisées sur un modèle réduit de pieu installé dans un sable sec ou dans une argile saturée. Trente cycles thermiques, représentant trente cycles annuels, sont appliqués au pieu sous différentes charges axiales en tête. Les résultats montrent un tassement irréversible avec les cycles thermiques ; ce tassement est plus important sous une charge axiale plus grande. De plus, le tassement est plus marqué pendant les premiers cycles thermiques et devient négligeable pour les cycles suivants. De l’autre côté, les travaux expérimentaux sur le modèle réduit de pieu sont complétés par les calculs numériques utilisant la méthode des éléments finis. Cette approche est d’abord validée avec les résultats obtenus sur le pieu modèle avant d’être utilisée pour prédire les résultats des expérimentations en vraie grandeur / The thermal and thermo-mechanical behavior of energy piles is investigated by various approaches: laboratory measurement on small soil samples, physical modeling on small-scale pile, experiments on real-scale pile, and analytical/numerical calculations. First, the thermal conductivity of unsaturated loess is measured simultaneously with moisture content and suction. The results show a unique relationship between thermal conductivity and moisture content during a wetting/drying cycle while a clear hysteresis loop can be observed on the relationship between thermal conductivity and suction. Second, thermal tests are performed on a full-scale experimental energy pile to observe heat transfer at the real scale. Third, an analytical solution is proposed to simulate conductive heat transfer from an energy pile to the surrounding soil during heating. The above-mentioned tasks related to the thermal behavior are then completed by studies on the thermo-mechanical behavior of energy piles. On one hand, experiments are performed on a small-scale pile installed either in dry sand or in saturated clay. Thirty thermal cycles, representing thirty annual cycles, are applied to the pile under various constant pile head loads. The results show irreversible pile head settlement with thermal cycles; the settlement is higher at higher pile head load. In addition, the irreversible thermal settlement is the most significant during the first cycles; it becomes negligible at high number of cycles. On the other hand, the experimental work with small-scale pile is completed with numerical calculations by using the finite element method. This approach is first validated with the results on small-scale pile prior to be used to predict the results of full-scale experiments

A low-order discrete dynamical system (DDS) for finite-rate chemistry of H2-air combustion is derived in 3D. Fourier series with a single wavevector are employed to represent dependent variables of subgrid-scale (SGS) behaviors for applications to large-eddy simulation (LES). A Galerkin approximation is applied to the governing equations for comprising the DDS. Regime maps are employed to aid qualitative determination of useful values for bifurcation parameters of the DDS. Both isotropic and anisotropic assumptions are employed when constructing regime maps and studying bifurcation parameters sequences. For H2-air reactions, two reduced chemical mechanisms are studied via the DDS. As input to the DDS, physical quantities from experimental turbulent flow are used. Numerical solutions consisting of time series of velocities, species mass fractions, temperature, and the sum of mass fractions are analyzed. Numerical solutions are compared with experimental data at selected spatial locations within the experimental flame to check whether this model is suitable for an entire flame field. The comparisons show the DDS can mimic turbulent combustion behaviors in a qualitative sense, and the time-averaged computed results of some species are quantitatively close to experimental data.

<p>Ultrafine particles (UFP) emitted by traffic have been associated with health risks for people living and working near major roadways. Studies have shown that people living in near-roadway communities experience higher risk of aggravated asthma, respiratory diseases and even childhood leukemia. Sharp concentration gradients of UFP have been reported near major highways with the concentration decreasing rapidly away from the road. Dispersion of UFP downwind of a road depends on many parameters, such as the atmospheric stability and wind speed. Presence of different structures such as noise barriers and vegetation can greatly influence the dispersion and downwind concentrations of UFP. These structures can block the traffic emissions and increase vertical mixing. In addition, vegetation can reduce UFP by deposition processes. Two sets of experiments were conducted in this thesis to investigate the effect of barriers on UFP deposition and dispersion. </p><p>The first set of experiments was performed in a wind tunnel facility to address UFP deposition to vegetation barriers solely. Two analytical models were proposed to characterize UFP dry deposition to vegetation measured during the wind tunnel experiment. The first model was derived from the filtration theory to explain UFP dry deposition to pine and juniper branches. The model agrees well with the experimental data indicating that pine and juniper branches can be treated as fibrous filters. The fiber diameters of pine derived from the experimental data were also similar to the physical diameters of pine needles; thus, providing further evidence that vegetation can be regarded as fibers. The second model was derived from the continuity equation and can predict the branch-scale dry deposition of UFP using conventional canopy properties such as the drag coefficient and leaf area density. Both models agree with the measurement results to within 20%.</p><p>The second set of experiments was done in three near-roadway environments to investigate the effects of barriers on the dispersion and dry deposition of UFP. We used mobile and stationary measurements to obtain the spatial and temporal variability of UFP. Both mobile and stationary measurements indicated that vegetation and noise barriers can reduce downwind UFP concentrations through dispersion and dry deposition by 20-60 %. </p><p>In conclusion, the effect of barriers on UFP dispersion and deposition has been characterized in this thesis. Two analytical models were also proposed from the wind tunnel experiments to characterize dry deposition and agreed well with the measurement results. The analytical model could benefit future climate and air quality models.</p> / Dissertation

<div>Electronics have now become a part of our daily life and therefore the reliability of microelectronics cannot be overlooked. As the Moore's law era comes to an end, various new system-level innovations (e.g., 3D packaging, evolution of packaging material to molding compounds) with constant scaling of transistors have resulted in increasingly complicated integrated circuits (ICs) configurations. The reliability modeling of complex ICs is a nontrivial concern for a variety of reasons. For example, ever since 2004, self-heating effect (SHE) has become an important reliability concern for ICs. Currently, many groups have developed thermal predictive models for transistors, circuits, and systems. In order to describe SHE self-consistently, the modeling framework must account for correlated self-heating within the ICs. This multi-scales nature of the self-consistency problem is one of the difficult factors poses an important challenge to self-consistent modeling. In addition, coupling between different physical effects within IC further complicates the problem.</div><div><br></div><div>In this thesis, we discuss three challenges, and their solutions related to an IC's reliability issues. We (i) generalize the classical effective medium theory (EMT) to account for anisotropic, heterogeneous system; (ii) develop computationally efficient a physics-based thermal compact model for a packaged ICs to predict junction temperature in the transistor based on the EMT model, and image charge theory. Our thermal compact model bridges different length scales among the sources and rest of the system. Finally (iii) propose the modeling framework of electrical chip package interaction (CPI) due to charge transport within mold compounds by coupling moisture diffusion, electric distribution, and ions transport. The proposed modeling framework not only addresses the three major modeling challenges discussed earlier, but also provides deep and fundamental insights regarding the performance and reliability of modern ICs. </div>

This research focuses on studying the feasibility of placing large wind turbines on deep-ocean platforms. Water tank studies have been conducted using the facilities at Alden Research Laboratories (ARL) on 100:1 scale Tension Leg Platform (TLP) and Spar Buoy (SB) models. Froude scaling was used for modeling the offshore wind turbine designs. Primary components of the platform turbine, tower, and cable attachments were fabricated in ABS plastic using rapid prototyping. A wireless data acquisition system was installed to prevent umbilical data cables from affecting the behavior of the platform when exposed to wave loading. In Phase I testing, Froude-scaled TLP and Spar Buoy models at a 100:1 scale were placed in a water flume and exposed to periodic waves at amplitudes ranging from 0.5 cm - 7.5 cm and frequencies ranging from 0.25 Hz - 1.5 Hz. The testing was conducted on simple tower and turbine models that only accounted for turbine weight at the nacelle. In Phase II testing, emphasis was placed on further testing of the tension leg platform as a more viable design for floating offshore wind turbines. The tension leg platform scale model was improved by adding a disc to simulate drag force incident at the top of the tower, as well as a rotor and blades to simulate the gyroscopic force due to turbine blade rotation at the top of the tower. Periodic wave motions of known amplitude and frequency were imposed on the model to study pitch, heave, roll, surge, sway motions and mooring cable tensions (in Phase II only) using accelerometers, inclinometers, capacitance wave gage, and load cells. Signal analysis and filtering techniques were used to refine the obtained data, and a Fourier analysis was conducted to study the dominant frequencies. Finally, Response Amplitude Operators (RAO's) were plotted for each data set to standardize the results and study the overall trend with respect to changes in wave amplitude and frequency. For Phase I testing, it is shown that surge motion of the platform dominates other motions for both the tension leg platform and spar buoy, and varying tether pretension has little effect on response amplitude operator values. For phase II testing, it was found that the introduction of thrust and gyroscopic forces increases sway and pitch motions as well as upstream tether forces. Coupling effects of pitch motion with roll and sway due to the presence of gyroscopic forces were also seen. The present experimental results can be used to validate the hydrodynamic kernels of linear frequency-domain models, time-domain dynamics models, and computational simulations on floating wind turbines. Numerical analysis and simulations have been conducted in a separate study at WPI. These simulations are comparable to the experimental results.

This dissertation consists of two parts, i.e. dynamic approaches for subgrid-scale (SGS) stress modelling for large eddy simulation and advanced assessment of the resolved scale motions related to turbulence geometrical statistics and topologies. The numerical simulations are based on turbulent Couette flow. The first part of the dissertation presents four contributions to the development of dynamic SGS models. The conventional integral type dynamic localization SGS model is in the form of a Fredholm integral equation of the second kind. This model is mathematically consistent, but demanding in computational cost. An efficient solution scheme has been developed to solve the integral system for turbulence with homogeneous dimensions. Current approaches to the dynamic two-parameter mixed model (DMM2) are mathematically inconsistent. As a second contribution, the DMM2 has been optimized and a modelling system of two integral equations has been rigorously obtained. The third contribution relates to the development of a novel dynamic localization procedure for the Smagorinsky model using the functional variational method. A sufficient and necessary condition for localization is obtained and a Picard's integral equation for the model coefficient is deduced. Finally, a new dynamic nonlinear SGS stress model (DNM) based on Speziale's quadratic constitutive relation [J. Fluid Mech., 178, p.459, 1987] is proposed. The DNM allows for a nonlinear anisotropic representation of the SGS stress, and exhibits a significant local stability and flexibility in self-calibration. In the second part, the invariant properties of the resolved velocity gradient tensor are studied using recently developed methodologies, i.e. turbulence geometrical statistics and topology. The study is a posteriori based on the proposed DNM, which is different than most of the current a priori approaches based on experimental or DNS databases. The performance of the DNM is further validated in terms of its capability of simulating advanced geometrical and topological features of resolved scale motions. Phenomenological results include, e.g. the positively skewed resolved enstrophy generation, the alignment between the vorticity and vortex stretching vectors, and the pear-shape joint probability function contour in the tensorial invariant phase plane. The wall anisotropic effect on these results is also examined.

This study proposes simulations which present optimized methods for producing fatty acids, fatty alcohols and alkanes using Zymomonas mobilis bacterium by the energy efficient β-oxidation reversal pathway, an eco-friendly alternative to the present petroleum based processes. Zymomonas has advantages of higher carbon intake, higher ethanol tolerance and higher ethanol production efficiency than other organisms. I have improved an earlier Zymomonas genome scale model and used Constraint Based Reconstruction and Analysis (COBRA), a linear optimization based computational tool in Matlab, and to perform flux balance analysis (FBA) based simulations. FBA accounts for formation, consumption, accumulation and removal rate or flux of each metabolite. The results present solution spaces of cell growth rate and product formation rate, which trend with products and their carbon chain length. I have analyzed these solution space trends gaining insight into the Zymomonas’ metabolism, enabling efficient product formation and opening a way for future improvement.

This dissertation consists of two parts, i.e. dynamic approaches for subgrid-scale (SGS) stress modelling for large eddy simulation and advanced assessment of the resolved scale motions related to turbulence geometrical statistics and topologies. The numerical simulations are based on turbulent Couette flow. The first part of the dissertation presents four contributions to the development of dynamic SGS models. The conventional integral type dynamic localization SGS model is in the form of a Fredholm integral equation of the second kind. This model is mathematically consistent, but demanding in computational cost. An efficient solution scheme has been developed to solve the integral system for turbulence with homogeneous dimensions. Current approaches to the dynamic two-parameter mixed model (DMM2) are mathematically inconsistent. As a second contribution, the DMM2 has been optimized and a modelling system of two integral equations has been rigorously obtained. The third contribution relates to the development of a novel dynamic localization procedure for the Smagorinsky model using the functional variational method. A sufficient and necessary condition for localization is obtained and a Picard's integral equation for the model coefficient is deduced. Finally, a new dynamic nonlinear SGS stress model (DNM) based on Speziale's quadratic constitutive relation [J. Fluid Mech., 178, p.459, 1987] is proposed. The DNM allows for a nonlinear anisotropic representation of the SGS stress, and exhibits a significant local stability and flexibility in self-calibration. In the second part, the invariant properties of the resolved velocity gradient tensor are studied using recently developed methodologies, i.e. turbulence geometrical statistics and topology. The study is a posteriori based on the proposed DNM, which is different than most of the current a priori approaches based on experimental or DNS databases. The performance of the DNM is further validated in terms of its capability of simulating advanced geometrical and topological features of resolved scale motions. Phenomenological results include, e.g. the positively skewed resolved enstrophy generation, the alignment between the vorticity and vortex stretching vectors, and the pear-shape joint probability function contour in the tensorial invariant phase plane. The wall anisotropic effect on these results is also examined.

The purpose of this study was to develop a questionnaire to measure academic self-efficacy for undergraduates. The theoretical framework of the questionnaire was based on Bandura¡¦s self-efficacy theory. A total of 409 participants were selected by judgment sampling from the first- and second-grade undergraduate of six colleges in national Sun Yat-sen university (Liberal Arts, Science, Engineering, Management, Marine Sciences, and Social Sciences). The newly developed Academic Efficacy and Motivation Questionnaire (AEMQ) was modified and derived from eight related scales. Expert content validity was conducted by three experts in the field, and the data were analyzed by rating scale model (RSM) in ConQuest. The findings were as follows: 1. All the 69 items in AEMQ have good fit values (MNSQ between 0.6 and 1.4). 2. The items of the AEMQ tended to be too easy for participants. 3. The reliability of each dimension of AEMQ was poor. The highest reliability value was 0.575, and the lowest was 0.382. 4. The first dimension ¡§actual performance¡¨ and the second dimension ¡§vicarious experience¡¨ have negative correlation with the forth dimension ¡§psychological index¡¨. The result was consistent with past studies.

Success of acid fracturing depends on uneven etching along the fracture surfaces caused by heterogeneities such as variations in local mineralogy and variations in leakoff behavior. The heterogeneities tend to create channeling characteristics, which provide lasting conductivity after fracture closure, and occur on a scale that is neither used in laboratory measurements of acid fracture conductivity, which use core samples that are too small to observe such a feature, nor in typical acid fracture simulations in which the grid block size is much larger than the scale of local heterogeneities. Acid fracture conductivity depends on fracture surface etching patterns. Existing acid fracture conductivity correlations are for random asperity distributions and do not consider the contribution of channels to the conductivity. An acid fracture conductivity correlation needs the average fracture width at zero closure stress. Existing correlations calculate average fracture width using dissolved rock equivalent width without considering the effect of reservoir characteristics. The purpose of this work is to develop an intermediate-scale acid fracture model with grid size small enough and the whole dimension big enough to capture local and macro heterogeneity effects and channeling characteristics in acid fracturing. The model predicts pressure field, flow field, acid concentration profiles, and fracture surface profiles as a function of acid contact time. By extensive numerical experiments with the model, we develop correlations of fracture conductivity and average fracture width at zero closure stress as a function of statistical parameters of permeability and mineralogy distributions. With the model, we analyzed the relationships among fracture surface etching patterns, conductivities, and the distributions of permeability and mineralogy. From result analysis, we found that a fracture with channels extending from the inlet to the outlet of the fracture has a high conductivity because fluid flow in deep channels needs a very small pressure drop. Such long and highly conductive channels can be created by acids if the formation has heterogeneities in either permeability or mineralogy, or both, with high correlation length in the direction of the fracture, which is the case in laminated formations.