Doctor of Philosophy / Department of Mechanical and Nuclear Engineering / M.H. Hosni / Z.C. Zheng / In order to study the capability of computational methods in investigating the mechanisms associated with disease and contaminants transmission in aircraft cabins, the Computational Fluid Dynamics (CFD) models are used for the simulation of turbulent airflow, tracer gas diffusion, and particle dispersion in a generic aircraft cabin mockup. The CFD models are validated through comparisons of the CFD predictions with the corresponding experimental measurements. It is found that using Large Eddy Simulation (LES) with the Werner-Wengle wall function, one can predict unsteady airflow velocity field with relatively high accuracy. However in the middle region of the cabin mockup, where the recirculation of airflow takes place, the accuracy is not as good as that in other locations. By examining different k-ε models, the current study recommends the use of the RNG k-ε model with the non-equilibrium wall function as a Reynolds Averaged Navier Stokes (RANS) model for predicting the steady-state airflow velocity data. It is also found that changing the cabin air-inlet nozzle height has a significant effect on the flow behavior in the middle and upper part of the cabin, while the flow pattern in the lower part is not affected as much. Through the use of LES and species transport model in simulating tracer gas diffusion, very good agreement between predicted and measured tracer gas concentration data is observed for some monitoring locations, but the agreement level is not uniform for all the sampling point locations. The reasons for the deviations between predictions and measurements for those locations are discussed. The Lagrange-Euler approach is invoked in the particle dispersion simulations. In this approach, the equation of motion for the discrete phase is coupled with the continuous phase governing equations through the calculation of drag and buoyancy forces acting on particles. The continuous phase flow is turbulent and RANS is employed in order to calculate the continuous phase velocity field. A complete study on grid dependence for RANS simulation is performed through a controllable regional mesh refinement scheme. The grid dependence study shows that using unstructured grid with tetrahedral and hybrid elements in the refinement region are more efficient than using structured grid with hexahedral elements. The effect of turbulence on the particle dispersion is taken into account by using a stochastic tracking method (Discrete Random Walk model). One of the significant features of this study is the investigation of the effect of the number of tries on the accuracy of particle concentration predictions when Discrete Random Walk is used to model turbulent distribution of particles. Subsequently, the optimum number of tries to obtain the most accurate predictions is determined. In accordance with the corresponding experimental data, the effect of particle size on particle distribution is also studied and discussed through the simulation of two different sizes of mono-disperse particles in the cabin with straight injection tube, i.e., 3µm and 10µm. Due to the low particle loading, neglecting the effect of particles motion on the continuous phase flow-field seems to be a reasonable, simplifying assumption in running the simulations. However, this assumption is verified through the comparison of the results from 1-way and 2-way coupling simulations. Eventually through the simulations for the particle injection using the cone diffuser, the effects of cabin pressure gradient as well as the particle density on particles dispersion behavior are studied and discussed. In the last part of this dissertation, the turbulent airflow in a full-scale Boeing 767 aircraft cabin mockup with eleven rows of seats and manikins is simulated using steady RANS method. The results of this simulation cannot only be used to study the airflow pattern, but also can be used as the initial condition for running the tracer gas diffusion and particle dispersion simulations in this cabin mockup.

A mesh-less method for solving incompressible, multi-phase flow problems has been developed and is discussed along with the presentation of benchmark results showing good agreement with theoretical and experimental results. Results of a systematic, parametric study of the single phase flow around a 2D circular cylinder at Reynolds numbers up to 1000 are presented and discussed. Simulation results show good agreement with experimental results. Extension of the method to deal with multiphase flow including liquid-to-vapor phase transition along with applications to cavitating flow are discussed. Insight gleaned from numerical experiments of the cavity closure problem are discussed along with recommendations for additional research. Several conclusions regarding the use of the method are made.

Este trabalho apresenta um ambiente de simulação de escoamentos com simetria radial e superfícies livres, baseado no sistema Freeflow. O sistema é formado por três módulos: um modelador de moldes, um simulador e um visualizador de escoamentos. O simulador implementa o método GENSMAC para a solução das equações de Navier-Stokes em coordenadas cilíndricas, utilizando diferenças finitas em uma malha diferenciada. São introduzidos os efeitos da tensão superficial e do ângulo de contato nas simulações com simetria radial. Alguns resultados de simulações utilizando este sistema e uma validação do código são apresentados, comparando simulações com soluções analíticas e experimentais, e estudando a convergência do método. / This work presents an environment for the simulation of axisymmetric free surface flows, based on the Freeflow system. The system contains three modules: a geometric model modeller, a simulator and a visualizator. The simulator implements the GENSMAC method for the solution of the Navier-Stokes equations in cylindrical coordinates, using finite differences in a staggered grid. The effects of surface tension and contact angle are introduced in the axisymmetric simulations. Some results from simulations using this system and a validation of the code are presented, comparing the simulations with analytical and experimental solutions, and studying the convergence of the method.

The present work is primarily concerned with studying the influence of an upwind spatial discretization on the capability of representing turbulent flows on aerospace applications, in the context of a flow simulation code that is fairly close to a production code. Therefore, the work addresses the issues of implementing and validating an advanced turbulence model for high Reynolds number aerospace applications in the context of an existing flux-vector splitting simulation tool, which incorporates several advances in current CFD practice, including parallel processing. The flow simulation tool used in the present work was originally developed for high speed, high altitude, hypersonic applications. Hence, the code did not include any provisions for turbulence modeling, since most flows at these conditions can be adequately treated as laminar flows. Moreover, due to the presence of strong shock waves, which are typical of hypersonic applications, a very dissipative spatial discretization scheme, based on the upwind flux vector splitting concept, was employed in the construction of the inviscid numerical fluxes. Therefore, the use of such a tool for the simulation of turbulent aerospace flows requires the implementation of a turbulence closure, as well as an adequate treatment of the excessive artificial dissipation automatically generated by the original spatial discretization scheme. In the present case, the flows of interest are simulated using the three-dimensional Reynolds-averaged Navier-Stokes equations. The turbulence closure considered is the one-equation, eddy viscosity, Spalart-Allmaras model. The work discusses in detail the theoretical and numerical formulation of the selected model, as well as the validation studies. The work also demonstrates how the spatial discretization scheme is selectively modified such that the flow simulation tool remains robust for high speed applications at the same time that it can accurately compute turbulent boundary layers. Furthermore, the work also addresses the parallelization and other high performance computational issues, demonstrating that the resultant flow simulation code can achieve adequate performance on current multi-CPU, multi-core computational clusters. Finally, the work discusses issues that could be considered for the continuation of the research effort here undertaken.

O presente trabalho objetiva o desenvolvimento, implementação e análise de métodos numéricos baseados em técnicas de malhas de blocos múltiplos. As malhas de blocos múltiplos se dividem em malhas de blocos múltiplos justapostos e malhas de blocos múltiplos sobrepostos. Dentre as classes existentes destes métodos, o trabalho vai se concentrar em técnicas de malhas sobrepostos. Dentre as classes existentes destes métodos, o trabalho vai se concentrar em técnicas de malhas sobrepostas, porque estas oferecem uma maior flexibilidade para o procedimento de geração de malhas estruturadas. Estas técnicas são particularmente adequadas para configurações realisticamente complexas. O presente trabalho analisa a técnica Chimera, explicando os processos a serem feitos para a criação de uma malha adequada para aplicação desta técnica. O trabalho compreeende também o desenvolvimento e validação de códigos de simulação baseados em uma formulação bidimensional e tridimensional. O algoritmo numérico para a solução de equações Euler/Navier-Stokes usa um esquema centrado com dissipação artificial no contexto de diferenças finitas. As aplicações realizadas com esta técnica foram feitas sobre geometrias de interesse em aerodinâmica.

In the context of gradient-based numerical optimisation, the adjoint method is an e cient way of computing the gradient of the cost function at a computational cost independent of the number of design parameters, which makes it a captivating option for industrial CFD applications involving costly primal solves. The method is however a ected by instabilities, some of which are inherited from the primal solver, notably if the latter does not fully converge. The present work is an attempt at curbing primal solver limitations with the goal of indirectly alleviating adjoint robustness issues. To that end, a novel discretisation scheme for the steady-state incompressible Navier- Stokes problem is proposed: Mixed Hybrid Finite Volumes (MHFV). The scheme draws inspiration from the family of Mimetic Finite Di erences and Mixed Virtual Elements strategies, rid of some limitations and numerical artefacts typical of classical Finite Volumes which may hinder convergence properties. Derivation of MHFV operators is illustrated and each scheme is validated via manufactured solutions: rst for pure anisotropic di usion problems, then convection-di usion-reaction and nally Navier-Stokes. Traditional and novel Navier-Stokes solution algorithms are also investigated, adapted to MHFV and compared in terms of performance. The attention is then turned to the discrete adjoint Navier-Stokes system, which is assembled in an automated way following the principles of Equational Di erentiation, i.e. the di erentiation of the primal discrete equations themselves rather than the algorithm used to solve them. Practical/computational aspects of the assembly are discussed, then the adjoint gradient is validated and a few solution algorithms for the MHFV adjoint Navier-Stokes are proposed and tested. Finally, two examples of full shape optimisation procedures on internal ow test cases (S-bend and U-bend) are reported.

Os dispositivos oscilantes por translação de ondas correspondem a um dos sistemas de maior potencial de geração de energia elétrica a partir das ondas do oceano, razão pela qual diversos engenheiros e pesquisadores dedicam-se ao desenvolvimento e aperfeiçoamento desta tecnologia. O princípio de funcionamento destes conversores, cujos principais exemplos são o Oyster, o Waveroller e o Langlee, baseia-se no movimento oscilatório de um sistema placapistão, que promove a geração de eletricidade por meio de uma turbina. Os trabalhos existentes acerca destes dispositivos ainda não reúnem informações suficientes que possibilitem a estimativa do comportamento hidrodinâmico e da captação de energia dos mesmos em situações reais de funcionamento. Desta forma, o presente trabalho busca preencher estas lacunas por meio de modelagem numérica bidimensional, baseada em médias de Reynolds (RANS), de ondas regulares atuando sobre os respectivos dispositivos. As simulações são realizadas pelo código computacional livre e aberto OpenFOAM v 4.1 e sua extensão OLAFOAM, ambos fundamentados em volumes finitos e que utilizam a metodologia VOF para o tratamento da superfície livre, solucionando as equações de Navier-Stokes e da continuidade. A representação da dinâmica de corpo rígido dos dispositivos é tratada por uma metodologia que permite a deformação dos elementos de malha e o movimento do fundo do domínio, que oscila juntamente com as placas dos conversores no intuito de assegurar a convergência numérica. Tal metodologia é validada com base na comparação dos resultados numéricos com valores experimentais presentes na literatura. As simulações realizadas permitiram concluir que o método de fundo deformável se mostra adequado no estudo da hidrodinâmica e captação de energia destes conversores, sendo uma alternativa a outras metodologias existentes. Um aumento na largura das placas destes conversores relaciona-se a uma intensificação das velocidades angulares e numa maior captação de energia, sendo a largura ideal de placa aquela cuja dimensão é próxima a duas vezes a sua altura. Um aumento na espessura das placas ocasiona uma diminuição na oscilação e na captação energética dos dispositivos, sendo a espessura ideal com dimensão entre 10 % e 20 % da altura da placa. A captação de energia tende a diminuir com o aumento da profundidade, sendo que, para lâminas de água superiores a duas vezes a altura da placa do conversor, a influência da profundidade torna-se pouco significativa. O aumento do período de onda é responsável por intensificar a ação do campo de velocidades nas regiões próximas ao conversor, sendo o período de máxima captação de energia em torno de 11 s em escala real. Maiores alturas de onda são responsáveis por intensificar o movimento e a captação de energia dos conversores. / The oscillating wave surge converters represents one of the most effective systems to transform the energy of ocean waves into electric power, reason why many researchers and engineers are dedicated to the development and improvement of this technology. The main examples of those kind of converters are Oyster, Waveroller and Langlee, which energy production is related to a oscillating plate-piston mechanism, that drives a secondary hydraulic system, moving a turbine and generating electricity. Recent papers about this technology does not bring enough information to predict the hydrodynamic and the power capture in real operation cases. In order to fulfill those gaps, the present study is focused on bidimensional numerical modelling of regular ocean waves interaction over the converters, based on Reynolds Average Navier Stokes equations (RANS). The numerical simulations are performed by the opensource code OpenFOAM v 4.1 and its extension OLAFOAM, both based on finite volume discretization and VOF method for free surface representation, which solves the Navier-Stokes and continuity equations. The rigid body dynamics is represented by a methodology that uses the mesh morphing method combined with an oscillating bottom, which moves according to the plate movement, in order to ensure numerical convergence. The present method is validated by the comparison of the numerical model results with previous experimental studies. The numerical results led to the conclusion that the oscillating bottom method seems to be adequate to perform a study based on the hydrodynamic and power capture of the oscillating wave surge converters, representing an alternative method to modelling the dynamic of those devices. An increase in plate width are related to an increase on the device oscillation and its power capture. The ideal width is approximately twice the plate height. Thicker plates causes lower angular velocities, as a result, lower electric generation. The more efficient thickness is around 10 % to 20 % of plate height. The power capture decreases according to an increase on water depth, and the depths with values around twice of the plate height does not seem to modify the device hydrodynamics. Higher wave periods causes an intensification of the velocity field near the converter. The pitching period, which is related to the maximum power captured, is near to 11 s in real scale. The highest wave heights are responsible to increase the converter oscillation and electricity generation.

This project is concerned with the development and implementation of a novel preconditioning method for the iterative solution of linear systems that arise in the finite element discretisation of the incompressible Navier-Stokes equations with weakly imposed boundary conditions. In this context we studied an augmented approach where the Schur complement associated with the momentum block of the Navier-Stokes equations has special sparse structure. We follow the standard inf-sup stable method of discretising the Navier-Stokes equations by the Taylor-Hood elements with the Lagrange multiplier constraints discretised using the same order approximation on matching grids. The resulting system of nonlinear equations is solved iteratively by Newton's method. The spectrum of the linearised Oseen's problem, preconditioned by the exact augmentation preconditioner was analysed. Then we developed inexact versions of the preconditioner aimed at achieving optimal scaling of the algorithm in terms of computational resources and wall-clock times. The experimental evaluation of the methodology involve a number of benchmark problems in two and three spatial dimensions. The obtained results demonstrate efficiency, robustness and almost optimal scaling of the solution algorithm with respect to the discrete problem size. We used OOMPH-LIB as a testbed for our experiments. The preconditioning strategies were implemented using OOMPH-LIB's Parallel Block Preconditioning Framework. The initial version of the software was significantly upgraded during the course of this project with newly implemented functionalities to facilitate the rapid development of sophisticated hierarchical design of parallel block preconditioners. Parallel performance analysis of the newly introduced functionalities demonstrate negligible overhead in terms of wall-clock time and the framework demonstrate good weak and strong parallel scaling.

Este trabalho apresenta um ambiente de simulação de escoamentos com simetria radial e superfícies livres, baseado no sistema Freeflow. O sistema é formado por três módulos: um modelador de moldes, um simulador e um visualizador de escoamentos. O simulador implementa o método GENSMAC para a solução das equações de Navier-Stokes em coordenadas cilíndricas, utilizando diferenças finitas em uma malha diferenciada. São introduzidos os efeitos da tensão superficial e do ângulo de contato nas simulações com simetria radial. Alguns resultados de simulações utilizando este sistema e uma validação do código são apresentados, comparando simulações com soluções analíticas e experimentais, e estudando a convergência do método. / This work presents an environment for the simulation of axisymmetric free surface flows, based on the Freeflow system. The system contains three modules: a geometric model modeller, a simulator and a visualizator. The simulator implements the GENSMAC method for the solution of the Navier-Stokes equations in cylindrical coordinates, using finite differences in a staggered grid. The effects of surface tension and contact angle are introduced in the axisymmetric simulations. Some results from simulations using this system and a validation of the code are presented, comparing the simulations with analytical and experimental solutions, and studying the convergence of the method.

The Seattle District wants to better manage the Ballard Locks and structures along the Lake Washington Ship Canal (LWSC) in a way that will maintain the environmental sustainability and biodiversity in the area. Due to strict salt water intrusion regulations in the LWSC, the Seattle District is working on upgrading their management practices such that they will resolve two inter-related problems. First, to improve the fish passage conditions for migrating salmon; and second, to learn how to better manage the salt wedge that forms and intrudes upstream. Based on the hydrodynamic and water quality results that are produced by this research, the Engineer Research and Development Center (ERDC) Portland Office will use their Eulerian-Lagrangian-Agent-Model (ELAM) to analyze fish patterns, looking for the most beneficial management schemes that assist salmon in migrating upstream. This research implemented CFD engineering techniques to help better understand the effectiveness of the hydraulic structures in the area, as well as come up with management practices that both mitigate the salt water intrusion from Puget Sound, and improve the migrating passages for salmon.