In case of fire, the smoke plum will reach the roof creating a ceiling jet leading to the edge and starting to descend. The smoke descending rate depends heavily on the smoke production rate and the way it was produced. Recent research has been concentrated on the smoke produced in an axi-symmetric plume model, but rarely on wall or corner plume models. In addition, the smoke production rate was estimated based on different fire sizes, which neglected the effect introduced by the make-up air, and is the main theme of this study. During the earlier stage of this study, the calculation models to predict the smoke entrainment rate was analyzed to identify its effectiveness in calculating the smoke production rate under cornered fires. In addition, the effect of the make-up air on the mechanical and natural smoke exhaust system was analyzed to calculate the corresponding smoke descending rate. Finally, the optimal calculation model to be utilized for the performance-based fire engineering designs of smoke management systems will be proposed, so that the make-up air effect can also be coupled during the design process.

The design of smoke management systems for buildings such as atria, covered shopping malls and sports arenas require appropriate calculation methods to predict the volume of smoky gases produced in the event of a fire. The volume of smoke must be calculated in order to determine the required fan capacity or ventilator area for a smoke management system. In design, consideration is often given to entrainment of air into a smoke flow from a compartment opening that subsequently spills and rises into an adjacent atrium void. This type of plume is commonly known as a thermal spill plume. There has been much controversy over the validity of various entrainment calculation methods for the spill plume and there are considerable differences in the calculated smoke production rates using these methods. There are also scenarios involving the spill plume where design guidance is very limited. Whilst over-sizing of the required smoke exhaust can be uneconomical, under-sizing can compromise the design objectives. This work attempts to rigorously characterises thermal spill plume entrainment using new data obtained from an extensive series of 1/10th physical scale modelling experiments, supported by numerical modelling using Computational Fluid Dynamics. Spill plume behaviour and subsequent entrainment appears to be specifically dependent on the characteristics of the layer flow below spill edge, particularly in terms of the width and the depth of the flow. Plumes generated from narrow, deep layer flows entrain air at a greater rate with respect to height compared to plumes generated from wide, shallow layers. The findings of this work go some way to explain and reconcile differences in entrainment reported between previous studies. New guidance has been developed for the thermal spill plume in smoke management design, in the form of a range of new simplified design formulae,improvements to analytical calculation methods and an initial assessment of the use of numerical modelling using Computational Fluid Dynamics.

abstract: This thesis examines the modeling, analysis, and control system design issues for scramjet powered hypersonic vehicles. A nonlinear three degrees of freedom longitudinal model which includes aero-propulsion-elasticity effects was used for all analysis. This model is based upon classical compressible flow and Euler-Bernouli structural concepts. Higher fidelity computational fluid dynamics and finite elementmethods are needed formore precise intermediate and final evaluations. The methods presented within this thesis were shown to be useful for guiding initial control relevant design. The model was used to examine the vehicles static and dynamic characteristics over the vehicles trimmable region. The vehicle has significant longitudinal coupling between the fuel equivalency ratio (FER) and the flight path angle (FPA). For control system design, a two-input two-output plant (FER - elevator to speed-FPA) with 11 states (including 3 flexible modes) was used. Velocity, FPA, and pitch were assumed to be available for feedback. Propulsion system design issues were given special consideration. The impact of engine characteristics (design) and plume model on control system design were addressed.Various engine designs were considered for comparison purpose. With accurate plume modeling, effective coupling from the FER to the FPA was increased, which made the peak frequency-dependent (singular value) conditioning of the two-input two-output plant (FER-elevator to speed-FPA) worse. This forced the control designer to trade off desirable (performance/robustness) properties between the plant input and output. For the vehicle under consideration (with a very aggressive engine and significant coupling), it has been observed that a large FPA settling time is needed in order to obtain reasonable (performance/ robustness) properties at the plant input. Ideas for alleviating this fundamental tradeoff were presented. Plume modeling was also found to be particularly significant. Controllers based on plants with insufficient plume fidelity did not work well with the higher fidelity plants. Given the above, the thesismakes significant contributions to control relevant hypersonic vehicle modeling, analysis, and design. / Dissertation/Thesis / M.S. Electrical Engineering 2010

No description available.

No description available.

Iceland has long been thought to be underlain by a thermal upwelling, or plume, rising from deep within the mantle. This study tests this hypothesis, by a) seeking evidence for a plume in the lower mantle in azimuth anomalies at the NORSAR array and b) mapping the three-dimensional structure of the mantle beneath Iceland using teleseismic tomography and data from an Iceland-wide broadband seismometer network. A temporary network of 30 digital broadband, three-component seismographs was deployed 1996-1998 to complement the existing, permanent seismic network on Iceland. This created a dense, well-distributed network. 3159 P-wave and 1338 S-wave arrival times were measured and inverted for velocity structure using the ACH method of teleseismic tomography. The preferred models are well-resolved down to -400 km, and reveal a low-velocity body with anomaly up to -2.9% in V(_p)) and -4.9% in V(_s) beneath central Iceland. This persists throughout the entire model depth range. The amplitudes of the anomalies imply an excess temperature of 200-300 K relative to the surrounding mantle. The morphology of the anomaly changes from cylindrical to tabular at 250-300 km depth, a feature that resolution tests suggest is real. This is consistent with the predictions of some convection models and suggests that the plume is restricted to the upper mantle. Anomalies in v(_p) and v(_s) provide evidence for lateral flow of material beneath the Reykjanes Ridge to the southeast in the depth range 50-200 km. Similar anomalies are present beneath the Kolbeinsey Ridge to the north only beneath 160 km. This shows that flow outwards beneath the Kolbeinsey Ridge is blocked by the Tjörnes Fracture Zone above 160 km. Azimuthal anomalies detected on the NORSAR array for rays travelling beneath Iceland at 1,500 km depth are consistent with a plume beneath Iceland at this lower-mantle depth with a Gaussian radius of 125 km and a strength of 1.5%. The observations do not serve as proof for such an anomaly because the solution is not unique. V(_p)/V(_s) ratios are 1% high throughout most of the plume, and up to 3.2% high at depths of 100-300 km beneath central and east-central Iceland. This suggests that up to a few percent of melt pervades the entire plume.

This research has utilised the Massachusetts Institute of Technology gen- eral circulation model (MITgcm) along with observations taken as part of the River In uences on Shelf Ecosystems (RISE) study to investigate the dynamic processes associated with the Columbia River plume at different temporal and spatial scales. Firstly, a high resolution ( x= y=25 m) investigation of the near-field plume was undertaken using the fully non-hydrostatic mode of the MITgcm. This resulted in the reproduction of a detailed inner plume as well as a series of radiated internal waves. In addition to first mode internal waves, second order waves were radiated from the plume boundary when propagation ve- locity becomes sub-critical. Third mode internal waves were also observed, trapped at the plume head. The fine plume structure produced revealed sec- ondary fronts within the plume that also generated internal waves. These features increase the mixing occurring inside the plume, resulting in greater entrainment of underlying waters into the plume. The use of Lagrangian drifters within the model produced detailed results of the recirculation tak- ing place within the emerging plume and how this recirculation changes with depth. This has implications for the near-field recirculation of biologically important solutes present in the plume waters. A second coarser resolution horizontal grid ( x= y=500 m) was imple- mented to investigate the processes of the large-scale plume with the addi- tion of wind forcing. Experiments with both simplified and realistic wind scenarios were carried out and comparisons with in-situ data were made. This revealed the dominance of wind effects on the outer plume and tidal effects on the inner plume. In the simplified wind cases, the classical the- ory of plume propagation under the action of upwelling and downwelling favourable winds was recreated. For the case of realistic winds, there was some success in reproducing a hindcast of the plume location. Tracer fields were used to represent nutrient concentrations based on observed data. Whilst these results showed variations from observations, they did allow a spatially and temporally complete view to be taken of nutrient distribu- tion in the region.

This work presents an approach which provides the real-time estimation of the gas concentration in a plume using an unmanned aerial vehicle (UAV) equipped with concentration sensors. The plume is assumed to be generated by a moving aerial or ground source with unknown strength and location, and is modeled by the unsteady advection-diffusion equation with ambient winds and eddy diffusivities. The UAV dynamics is described using the point-mass model of a fixed-wing aircraft resulting in a sixth-order nonlinear dynamical system. The state (gas concentration) estimator takes the form of a Luenberger observer based on the advection-diffusion equation. The UAV in the approach is guided towards the region with the larger state-estimation error via an appropriate choice of a Lyapunov function thus coupling the UAV guidance with the performance of the gas concentration estimator. This coupled controls-CFD guidance scheme provides the desired Cartesian velocities for the UAV and based on these velocities a lower-level controller processes the control signals that are transmitted to the UAV. The finite-volume discretization of the estimator incorporates a second-order total variation diminishing (TVD) scheme for the advection term. For computational efficiency needed in real-time applications, a dynamic grid adaptation for the estimator with local grid-refinement centered at the UAV location is proposed. The approach is tested numerically for several source trajectories using existing specifications for the UAV considered. The estimated plumes are compared with simulated concentration data. The estimator performance is analyzed by the behavior of the RMS error of the concentration and the distance between the sensor and the source.

This dissertation investigates the dynamics of the tidally modulated outflow from the Columbia River mouth using high resolution measurements of velocity, density and turbulent microstructure. At high tide, flow through the river mouth reverses from flood (onshore) to ebb (offshore). During ebb, buoyant fluid issues from the river mouth and spreads offshore across the ocean surface. This is the Columbia River tidal plume. The fluid velocity of the tidal plume is super-critical (greater than the wavespeed of coastal stratification), which creates a zone of sharp surface velocity convergence at its leading edge, causing a front to form. From early ebb to peak ebb, constant front propagation speed and plume expansion rate are controlled by a linearly increasing volume-flux through the river mouth. Within the plume, turbulence at the plume base is strongly related to the difference between the shear-squared, S², and four times the buoyancy frequency squared, 4N². A parameterization based on the excess shear-squared, S² - 4N², represents Reynolds stress well, indicating that it is driven by Kelvin-Helmholtz instability. During peak ebb of large tides, high volume-flux through the mouth drives high S² - 4N², causing high plume-base stress, which forces significant deceleration of the plume. During smaller tides the volume-flux is smaller, S² - 4N² lower, and the stress too weak to significantly decelerate the plume. During mid-ebb of both small and large ebbs, increasing buoyancy flux from the river mouth raises plume stratification, which suppresses S² - 4N² and stress. As ebb ends, decreasing volume flux and deflection by the Coriolis effect limit plume expansion. This weakens surface velocity convergence, causing the front to diffuse. On longer timescales, plume N² is modulated by changes in river flow; higher river flow causes higher N². During peak ebb of large tides this increase in N² supports higher S², resulting in higher S² - 4N², which causes larger internal stress. These results describe the primary dynamics of the Columbia River tidal plume from front formation to late-ebb, and relate variability in those dynamics to tidal and river-flow forcing. / Graduation date: 2011 / Access restricted to the OSU Community at author's request from Jan. 26, 2011 - Jan. 26, 2013

No description available.