Computational fluid dynamics (CFD) plays an important role in many scientific applications, ranging from designing more effective boat engines or aircraft wings to predicting tomorrow's weather, but at the cost of requiring huge amounts of computing time. Also, traditional algorithms suffer from scalability limitations, making them hard to parallelize massively. As a relatively new and promising method for computational fluid dynamics, the Lattice Boltzmann algorithm tries to solve the scalability problems of conventional, but well-tested algorithms in computational fluid dynamics. Through its inherently local structure, it is well suited for parallel processing, and has been implemented on many different kinds of parallel platforms. Adapteva's Epiphany platform is a modern, low-power manycore architecture, which is designed to scale up to thousands of cores, and has even more ambitious plans for the future. Hardware support for floating-point calculations makes it a possible choice in scientific settings. The goal of this thesis is to analyze the performance of the Lattice Boltzmann algorithm on the Epiphany platform. This is done by implementing and testing the lid cavity test case in two and three dimensions. In real applications, high performance on large lattices with millions of nodes is very important. Although the tested Epiphany implementation scales very good, the hardware does not provide adequate amounts of local memory and external memory bandwidth, currently preventing widespread use in computational fluid dynamics.

No description available.

No description available.

Heat exchanger fouling is the deposition of material onto the heat transfer surface causing a reduction in thermal efficiency. A study using Computational Fluid Dynamics (CFD) was conducted to increase understanding of key aspects of fouling in desalination processes. Fouling is a complex phenomenon and therefore this numerical model was developed in stages. Each stage required a critical assessment of each fouling process in order to design physical models to describe the process???s intricate kinetic and thermodynamic behaviour. The completed physical models were incorporated into the simulations through employing extra transport equations, and coding additional subroutines depicting the behaviour of the aqueous phase involved in the fouling phenomena prominent in crystalline streams. The research objectives of creating a CFD model to predict fouling behaviour and assess the influence of key operating parameters were achieved. The completed model of the key crystallisation fouling processes monitors the temporal variation of the fouling resistance. The fouling rates predicted from these results revealed that the numerical model satisfactorily reproduced the phenomenon observed experimentally. Inspection of the CFD results at a local level indicated that the interface temperature was the most influential operating parameter. The research also examined the likelihood that the crystallisation and particulate fouling mechanisms coexist. It was found that the distribution of velocity increased the likelihood of the particulate phase forming within the boundary layer, thus emphasizing the importance of differentiating between behaviour within the bulk and the boundary layer. These numerical results also implied that the probability of this composite fouling was greater in turbulent flow. Finally, supersaturation was confirmed as the key parameter when precipitation occurred within the bulk/boundary layer. This investigation demonstrated the advantages of using CFD to assess heat exchanger fouling. It produced additional physical models which when incorporated into the CFD code adequately modeled key aspects of the crystallisation and particulate fouling mechanisms. These innovative modelling ideas should encourage extensive use of CFD in future fouling investigations. It is recommended that further work include detailed experimental data to assist in defining the key kinetic and thermodynamic parameters to extend the scope of the required physical models.

The design of supersonic flow paths for scramjet engines and high Mach number shock tunnel nozzles is complicated by high temperature flow effects and multidimensional inviscid/ viscous flow interactions. Due to these complications, design in the past has been enabled by making flow modelling simplifications that detract from the accuracy of the flow analysis. A relatively new approach to designing aerodynamic bodies, which automates design and does not require as many simplifying assumptions to be effective, is the coupling of a computational flow solver to an optimization algorithm. In this study, a new three-dimensional space-marching computational flow solver is developed and coupled to a gradient-search optimization algorithm. This new design tool is then used for the design optimization of an axisymmetric scramjet flow path and two high Mach number shock tunnel nozzles. The flow solver used in the design tool is an explicit, upwind, space-marching, finite-volume solver for integrating the three-dimensional parabolized Navier-Stokes equations. It is developed with an emphasis on simplicity and efficiency. Cross-stream fluxes are calculated using Toro's efficient upwind, linearized, approximate Riemann solver in flow regions of slowly varying data, and an Osher type solver in the remainder of the flow. Vigneron's technique of splitting the streamwise pressure gradient in subsonic regions is used to stabilise the flux calculations. A three-dimensional implementation of an algebraic turbulence model, a finite-rate chemistry model and a thermodynamic equilibrium model are also implemented within the solver. A range of test cases is performed to (1) validate and verify the phenomenological models implemented within the solver, thereby ensuring the simulation results used for design are credible, and (2) demonstrate the speed of the solver. The first application of the new computational design tool is the design of a scramjet flow path, which is optimized for maximum axial thrust at a flight Mach number of 12. The optimization of a scramjet flow path has been examined previously, however, this study differs to others published in that the flow is modelled using a turbulence model and a finite-rate chemical reaction model which add to the fidelity of the simulations. The external shape of the scramjet vehicle is constrained early on in the design process, therefore, the design of the scramjet is restricted to the internal flow path. Because of this constraint, and the large internal surface area of the combustor and the high skin friction iv within the combustor, the net calculated force exerted on the scramjet for both the initial and optimized design is a drag force. The drag force of the initial design, however, is reduced by 60% through optimization. The second application of the design tool is the wall contour of an axisymmetric Mach 7 shock tunnel nozzle, which is computationally optimized for minimum test core flow variation to a level of +/- 0.019 degrees for the flow angularity and +/- 0.26% for the Pitot pressure. The design is verified by constructing a nozzle with the optimized wall contour and conducting experimental Pitot surveys of the nozzle exit flow. The measured standard deviation in core flow Pitot pressure is 1.6%. However, because there is a large amount of experimental noise, it is expected that the actual core flow uniformity may be better than indicated by the raw experimental data. The last application of the computational design tool is a contoured Mach 7 square cross-section shock tunnel nozzle. This is a three-dimensional optimization problem that demonstrates the versatility of the design tool, since the effort required to implement the optimization algorithm is independent of the complexity of the flow-field and flow solver. Optimization results show that the variation in the test core flow properties could only be reduced to a Mach number variation of +/- 7% and flow angle variation of +/- 1.2 degrees ,for a short nozzle suitable for a shock tunnel. The magnitudes of the optimized nozzle exit flow deviations for the short nozzle and two other longer nozzles indicate that generating uniform flow becomes increasingly difficult as the length of square cross-section nozzles is reduced. Overall, the current research shows that coupling a flow solver to an optimization algorithm is an effective and insightful way of designing scramjets and shock tunnel nozzles.

Dissertation (Ph.D.) -- Worcester Polytechnic Institute. / Keywords: steam reforming; reactor modeling; packed bed; fixed bed; CFD. Includes bibliographical references (p.).

Thesis (Ph. D.)--Brigham Young University. Dept. of Chemical Engineering, 2007. / Includes bibliographical references (p. 163-169).

Thesis (M.S.)--California Polytechnic State University, 2008. / Mode of access: Internet. Title from PDF title page; viewed on March 17, 2009. Major professor: David D. Marshall. "Presented to the faculty of California Polytechnic State University, San Luis Obispo." "In partial fulfillment of the requirements for the degree of Master of Science in Aerospace Engineering." "June 2008." Includes bibliographical references (p. 126-127). Also available on microfiche.

Thesis (Ph. D.)--Oregon State University, 2009. / Printout. Includes bibliographical references (leaves 220-225). Also available on the World Wide Web.

Thesis (Ph. D.)--University of Texas at Austin, 2008. / Vita. Includes bibliographical references.