Thesis (M.S.)--University of Wisconsin--Madison. / Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 70-71).

No description available.

Une caractéristique remarquable de la convection de Rayleigh-Bénard qui concerne une couche de fluide horizontale chauffée par le bas et refroidie par le haut, est l’établissement spontané de l’ordre spatial et la formation d’une circulation cohérente à grande échelle. Comment les différents facteurs, tels que la géométrie du domaine et les conditions limites, influencent l’écoulement à grande échelle, restent une question ouverte. Malgré sa simplicité apparente, la convection de Rayleigh-Bénard présente une dynamique à grande échelle incroyablement riche et complexe, tels que des modes de torsion et du battement, la rotation du plan et des cessations de la circulation, qui coexistent souvent et se concourent. Une approche couramment utilisée dans l’étude des cessations, consiste à contraindre la circulation à grande échelle à un plan en limitant le domaine fluide à une cellule carrée (2D) ou à une cellule rectangulaire mince (quasi-2D). Cependant, il n’est pas tout à fait clair si les retournements 2-D et (quasi-)2-D correspondent au même phénomène. Le présent document est consacré à l’étude des modes d’écoulement à grande échelle dans la convection turbulente de Rayleigh-Bénard et de l’influence exercée par différents facteurs sur les structures d’écoulement et sur leur évolution temporelle. La caractérisation proposée combine une analyse statistique avec une approche physique s’appuyant sur le moment angulaire ainsi que sur les énergies cinétiques et potentielles pour mettre en évidence les mécanismes physiques sous-jacents. Nous essayons ensuite de relier ces mécanismes à chacun des modes d’écoulement distinctifs observés et à leur évolution. / One outstanding feature of the Rayleigh-Bénard problem which concerns a horizontal fluid layer heated from below and cooled from above, is the spontaneous establishment of spatial ordering and the formation of a coherent large-scale circulation. How different factors, such as the domain geometry and boundary conditions, influence the sizes and shapes of the large-scale flow remains an open question. Despite its apparent simplicity, Rayleigh-Bénard convection exhibits some incredibly rich and complex large-scale dynamics such as torsional modes, rotation, sloshing, and cessations, which often coexist and compete to each other. One approach, commonly used in the study of cessations is to constrain the large scale circulation to a plane by restricting the fluid domain to a (2-D) square cell or to a slim rectangular cell of small aspect ratio in the transversal direction. However, it is not entirely clear whether the 2-D and (quasi-)2-D reversals correspond to the same phenomenon. The present document is dedicated to the study of the large-scale flow patterns in turbulent Rayleigh-Bénard convection, and of the influence exerted by different factors on the flow structures and on their temporal evolution. The proposed characterization combines a statistical analysis with a physical approach relying on the angular momentum as well as the kinetic and potential energies to highlight the underlying physical mechanisms. We subsequently attempt to tie these mechanisms together to each of the distinctive flow patterns observed and to their evolution.

This thesis examines two important physical phenomena that occur when solid fuels are exposed to external radiative heating: (1) the pyrolysis process in reaching ignition conditions and (2) the natural convection around one or more radiatively heated fuel samples. A vegetation fire (bushfire, wildfire, or forest fire) preheating the vegetation which is in its path is a particular example which occurs in nature. However there are many more applications where modelling the pyrolysis process and/or the natural convection is of practical use. For the pyrolysis phenomena, a one-dimensional time dependent pyrolysis model is proposed. The mathematical model is solved numerically and results are used to analyse the influence of the size of a wood-based fuel sample, the heating rate it is exposed to, and its initial moisture content in the process of the sample reaching the conditions where it can produce enough pyrolysate vapour to support a flame (flash point). In many pyrolysis models in the open literature it is assumed that the fuel samples are dry. In the present study it is found that the initial moisture content has a marked effect for a fuel sample reaching its flash point. For the convection phenomena, a two-dimensional steady model, which explores the natural convection around one or more solid fuels, is also presented. The flame front is represented by a radiating panel. This means that the solid fuels receive a non-uniform heating rate depending on their geometry and location in relation to the panel. Changes in temperature and velocity profiles are monitored for varying heating rates and sample sizes (or, equivalently, the Rayleigh number Ra). Additionally, in the case of multiple fuel samples, changes in the distance between the fuels is also taken into account. For multiple fuels in arbitrary locations it is possible that one sample will block some of the radiation from the panel from reaching another sample. This means that the fuel sample will receive a reduced heating rate. This reduction in heating is also incorporated in the natural convection model. Both the pyrolysis and natural convection models are solved numerically using the finite element software package COMSOL Multiphysics. A comparison of COMSOL is performed with benchmark solutions provided by the open literature. A good agreement in the numerical results is observed.

Convective heat transfer in a rotary kiln was studied as a function of operating parameters. The experiments were carried out in a steel kiln of 0.19 m in diameter and 2.44 m in length. The operating parameters covered included gas flow rate, solid throughput, rotational speed, degree of solid holdup, inclination angle, particle size and temperature. To minimize radiation effects, air was used as the heating medium and maximum inlet air temperatures were limited to 650 K. Ottawa sand was used in all the runs except in the study of the effect of particle size where limestone was employed. The experiments were conducted under conditions where the bed height along the kiln was maintained constant and the bed was in the rolling mode. Both the heat transfer coefficients from the gas to the solids bed and the gas to the rotating wall were found to be significantly influenced by gas flow rate. Increasing rotation al speed increases the gas to bed heat transfer, but decreases the gas to wall heat transfer. The former effect is relatively small. The effect of degree of fill was slightly negative in the gas to solids bed heat transfer, and insignificant in the heat transfer from the gas to wall. The effects of inclination angle, solid throughput, particle size and temperature were found negligible over the range tested. One of the major findings in this study is that contrary to suggestions in the literature, the coefficients for gas to bed heat transfer are about an order of magnitude higher than those for gas to wall. The higher coefficients for gas to solids bed are attributed to two factors, the underestimation of the true area by basing coefficients on the plane chord area and the effect on the gas film resistance of the rapid particle velocity on the bed surface. The experimental data were correlated in a form suitable for design purposes, and the results compared with meager data available in the literature. A mathematical model was developed for convective heat transfer from the gas to a rolling solids bed. The model requires the knowledge of the gas to particle heat transfer coefficient and the rolling velocity of the aerated particles. The model gives a reasonable prediction of the gas to bed coefficient in a rotary kiln using values of the gas to particle coefficient taken from the literature. The required data on the surface velocity of particles was obtained in a lucite kiln of the same size. Residence time distribution of particles was also studied briefly to verify that solids were nearly in axial plug flow. A simple mathematical model of a rotary kiln heat exchanger is presented. This model predicts gas, solids and wall-temperatures in a kiln as a function of the kiln design and operating parameters using the heat transfer correlations developed in this work. / Applied Science, Faculty of / Chemical and Biological Engineering, Department of / Graduate

Thesis. 1978. M.S.--Massachusetts Institute of Technology. Dept. of Meteorology. / Microfiche copy available in Archives and Science. / Bibliography: leaves 133-137. / M.S.

During the April 25 – 28, 2011 severe weather outbreak, 350 tornadoes were confirmed across 21 states, making the event the largest 3-day outbreak in U.S. History. Of the 350 tornadoes, 13 were of EF4 or EF5 strength. Due to complex terrain and vegetation in northeastern Alabama, northwestern Georgia, and eastern Tennessee, only four tornadoes were analyzed in this study. Abrupt changes in vegetation and the related sensible and latent heat fluxes have been shown to enhance convective activity along and near the resulting land surface discontinuities. This study analyzed heightened convective activity (analyzed by looking at cloud-to-ground lightning data) along each tornado track on days of weak synoptic forcing. Post- tornado months showed no signs of enhanced convective activity along any of the tornado tracks analyzed in this study, which could be attributed to several factors including study period, duration of intensity, tornado track length and width, and land cover.

No description available.

No description available.

A safety analysis for the McMaster Nuclear Reactor has been carried out for postulated scenarios of loss or termination of forced flow in the reactor core in a state of shutdown, with loss of pool inventory of different magnitudes including core uncovery. Models were developed to evaluate the natural convection flow through the core assemblies for the different conditions within the aforementioned envelope. The flow rate was used to get the temperature or enthalpy rise along the heated channel in order to estimate the corresponding clad temperatures in the given scenarios. The models were constructed from first principles using the one-dimensional momentum conservation law, incorporating the Boussinesq approximation for the single-phase case and the Homogeneous Equilibrium Model assumptions when a two-phase mixture was present. In order to obtain the flow rate and enthalpy rise along the channel, knowledge of the assembly power and inlet temperature is required. The power was calculated using a well known decay power correlation. The pool temperature which was used as the assembly inlet temperature was calculated via a lumped parameter model using a simple energy balance between the core output (again by using the decay-heat profile) and the pool heatup. Heat losses from the pool were neglected and the model allowed for reaching saturation temperature in the pool. In this case, water vaporization was calculated using the latent heat to assess pool inventory loss rate. For all scenarios before core uncovery, the models predict that clad and fuel temperatures remained well below limits associated with clad blistering or melting. Consequently, it is asserted natural convection and acceptable temperatures will be sustained in the McMaster Nuclear Reactor while the core remains covered. In the most severe draining before uncovery, in which the pool drains to just before exposing the core, it takes approximately a week (180 hours) after shutdown for boiling to start in the core’s hottest channel. For core uncovery, the models predict that the clad remains below the blistering temperature for pool height at 9.4% of the heated channel’s height (corresponding to exposing about 61.7 cm of the assembly), and below melting temperature for pool height at 8.1% of the heated channel’s height (corresponding to exposing about 62.5 cm of the assembly). Both heights are below the height of the bottom of the lowest beam tube, at which the worst draining case will end. / Thesis / Master of Applied Science (MASc)