Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Em uma grande gama de problemas físicos, governados por equações diferenciais, muitas vezes é de interesse obter-se soluções para o regime transiente e, portanto, deve-se empregar técnicas de integração temporal. Uma primeira possibilidade seria a de aplicar-se métodos explícitos, devido à sua simplicidade e eficiência computacional. Entretanto, esses métodos frequentemente são somente condicionalmente estáveis e estão sujeitos a severas restrições na escolha do passo no tempo. Para problemas advectivos, governados por equações hiperbólicas, esta restrição é conhecida como a condição de Courant-Friedrichs-Lewy (CFL). Quando temse a necessidade de obter soluções numéricas para grandes períodos de tempo, ou quando o custo computacional a cada passo é elevado, esta condição torna-se um empecilho. A fim de contornar esta restrição, métodos implícitos, que são geralmente incondicionalmente estáveis, são utilizados. Neste trabalho, foram aplicadas algumas formulações implícitas para a integração temporal no método Smoothed Particle Hydrodynamics (SPH) de modo a possibilitar o uso de maiores incrementos de tempo e uma forte estabilidade no processo de marcha temporal. Devido ao alto custo computacional exigido pela busca das partículas a cada passo no tempo, esta implementação só será viável se forem aplicados algoritmos eficientes para o tipo de estrutura matricial considerada, tais como os métodos do subespaço de Krylov. Portanto, fez-se um estudo para a escolha apropriada dos métodos que mais se adequavam a este problema, sendo os escolhidos os métodos Bi-Conjugate Gradient (BiCG), o Bi-Conjugate Gradient Stabilized (BiCGSTAB) e o Quasi-Minimal Residual (QMR). Alguns problemas testes foram utilizados a fim de validar as soluções numéricas obtidas com a versão implícita do método SPH. / In a wide range of physical problems governed by differential equations, it is often of interest to obtain solutions for the unsteady state and therefore it must be employed temporal integration techniques. One possibility could be the use of an explicit methods due to its simplicity and computational efficiency. However, these methods are often only conditionally stable and are subject to severe restrictions for the time step choice. For advective problems governed by hyperbolic equations, this restriction is known as the Courant-Friedrichs-Lewy (CFL) condition. When there is the need to obtain numerical solutions for long periods of time, or when the computational cost for each time step is high, this condition becomes a handicap. In order to overcome this restriction implicit methods can be used, which are generally unconditionally stable. In this study, some implicit formulations for time integration are used in the Smoothed Particle Hydrodynamics (SPH) method to enable the use of larger time increments and obtain a strong stability in the time evolution process. Due to the high computational cost required by the particles tracking at each time step, the implementation will be feasible only if efficient algorithms were applied for this type of matrix structure such as Krylov subspace methods. Therefore, we carried out a study for the appropriate choice of methods best suited to this problem, and the methods chosen were the Bi-Conjugate Gradient (BiCG), the Bi-Conjugate Gradient Stabilized (BiCGSTAB) and the Quasi-Minimal Residual(QMR). Some test problems were used to validate the numerical solutions obtained with the implicit version of the SPH method.

In this work, the mesh-free smoothed particle hydrodynamics (SPH) method is applied in the modelling of the direct laser interference patterning (DLIP) of metal surfaces. The DLIP technique allows the fabrication of periodic microstructures on technical surfaces using nanosecond laser pulses. Here, the interference of two coherent partial beams with a sinusoidal energy density distribution of the interference pattern is concerned, which is employed to generate line-like surface structures. However, the mechanisms effective during nanosecond pulsed DLIP of metals are not yet fully understood. The physical phenomena occurring due to the interaction of laser radiation with metallic materials are first considered and the governing differential equations are stated. The fundamentals of the SPH method and the approaches to the numerical treatment of the conservation equations are presented. Physical processes relevant to the modelling of laser material processing are solved by suitable SPH techniques, i.e. the approximations are verified with respect to test problems with analytical or known numerical solutions. Consequently, the SPH method is used to devise a thermal model of the DLIP process, considering the absorption of the laser radiation, the heat conduction into the workpiece and the latent heat of involved phase changes. This model is extended to compute the melt pool convection during DLIP, which is driven by surface tension gradients due to temperature gradients. For this purpose, an incompressible SPH (ISPH) method is used, representing a novel approach to the modelling of the laser-induced melt pool flow. The numerical model is employed to perform simulations of DLIP on metal substrates. Firstly, the thermal simulation of the single pulse patterning of stainless steel is in good agreement with experimental results. The application of DLIP to stainless steel and aluminium is then simulated by the comprehensive model including the melt pool flow. Moreover, this model is further extended to consider the non-linear temperature dependence of surface tension, as in liquid steel in the presence of a surface active element. The simulation results reveal a distinct behaviour of stainless steel and aluminium substrates. A markedly deeper melt pool and considerable velocity magnitudes of the thermocapillary convection at the melt surface are computed for DLIP of aluminium. In contrast, the melt pool flow is less pronounced during DLIP of stainless steel, whereas higher surface temperatures are predicted. Hence the Marangoni convection is a conceivable effective mechanism during the structuring of aluminium at moderate energy density. The different character of the melt pool convection during DLIP of stainless steel and aluminium is corroborated by experimental observations. Furthermore, the simulations for stainless steel with different sulphur content indicate distinct melt pool flow patterns and support the explanation of the microstructures found after DLIP experiments. The role of vapourisation and the induced recoil pressure in the microstructure evolution due to DLIP on metal substrates at elevated fluences could be prospectively investigated. In this regard, the consideration of the melt pool surface deformation in the ISPH algorithm, and particularly a suitable pressure boundary condition, is required.:I The research problem 1 Motivation 2 Modelling of laser material processing 2.1 Interaction of laser radiation with materials 2.1.1 Absorption of laser radiation 2.1.2 Heat conduction and phase change 2.1.3 Molten pool convection 2.1.4 Vapourisation regime 2.2 Mathematical modelling of laser material interaction 2.2.1 Conservation equations in Lagrangian formulation 2.2.2 Influence of surface tension 3 State of the art in laser microprocessing and the SPH method 3.1 Laser microprocessing 3.2 Simulation of direct laser interference patterning 3.3 The mesh-free smoothed particle hydrodynamics method 3.3.1 Fundamental approximations and kernel function 3.3.2 Particle distribution and interaction length 3.3.3 Approximation of derivatives 3.3.4 Treatment of boundaries 3.3.5 Neighbourhood search 3.4 Numerical modelling of laser material processing by SPH II SPH model development for direct laser interference patterning 4 SPH modelling of heat transfer and fluid flow 4.1 Solution of the heat diffusion equation 4.2 Formulation of equations governing fluid flow 4.2.1 Equation of continuity 4.2.2 Approximation of pressure gradient term 4.2.3 Treatment of viscosity 4.3 Weakly compressible SPH method for solving fluid flow 4.3.1 Particle motion 4.3.2 Time integration 4.3.3 Time step criteria 4.4 Incompressible SPH method for solving fluid flow 4.4.1 Time integration 4.4.2 Discrete incompressible SPH algorithm 4.4.3 Time step criteria 4.5 Simulation of thermal fluid flow using ISPH 4.5.1 Semi-implicit time integration 4.5.2 Solution of the pressure Poisson equation 5 Verification of the SPH implementation 5.1 Transient heat conduction in laser-irradiated plate 5.1.1 Problem description 5.1.2 Dimensionless formulation 5.1.3 Numerical solution and results 5.2 Viscous flow 5.2.1 Couette flow 5.2.2 Poiseuille flow 5.3 Thermal convection 5.3.1 Natural convection in a square cavity 5.3.2 Rayleigh--Marangoni--Bénard convection in liquid aluminium 6 SPH model of direct laser interference patterning 6.1 Characteristics of the process 6.2 Thermal model 6.2.1 Non-dimensionalisation 6.2.2 Numerical solution of governing equation 6.2.3 Verification of the computation 6.2.4 Numerical test 6.3 Thermofluiddynamic model 6.3.1 Non-dimensionalisation 6.3.2 Numerical solution of governing equations 6.3.3 Discretisation 6.3.4 Resolution independence study 7 SPH simulation of direct laser interference patterning 7.1 Thermal model 7.1.1 DLIP experiments on stainless steel substrates 7.1.2 Thermal simulation of DLIP on steel substrate 7.2 Thermofluiddynamic model 7.2.1 Material properties and simulation parameters 7.2.2 Numerical results for steel substrate 7.2.3 Numerical results for aluminium substrate 7.2.4 Discussion and comparison with experiments 7.3 Extended thermofluiddynamic model 7.3.1 Model parameters 7.3.2 Influence of sulphur content on DLIP of stainless steel 8 Conclusions and outlook Bibliography / In dieser Arbeit wird die direkte Laserinterferenzstrukturierung (Direct Laser Interference Patterning, DLIP) von Metallen mit der netzfreien Smoothed Particle Hydrodynamics (SPH) Methode modelliert. Das DLIP-Verfahren ermöglicht die Fertigung periodischer Mikrostrukturen auf technischen Oberflächen mit Nanosekunden-Laserpulsen. Hier wird die Zweistrahlinterferenz mit einer sinusförmigen Energiedichteverteilung des Interferenzmusters behandelt, die linienförmige Oberflächenstrukturen erzeugt. Die bei der direkten Interferenzstrukturierung von Metallen mit Nanosekunden-Laserpuls wirksamen Mechanismen sind jedoch noch nicht verstanden. Die aufgrund der Wechselwirkung von Laserstrahlung mit metallischen Werkstoffen auftretenden physikalischen Phänomene werden zuerst betrachtet und die sie bestimmenden Differentialgleichungen angegeben. Die Grundlagen der SPH-Methode sowie deren Herangehensweisen an die numerische Behandlung der Erhaltungsgleichungen werden vorgestellt. Für die Modellierung der Lasermaterialbearbeitung relevante physikalische Vorgänge werden mittels geeigneter SPH-Ansätze gelöst, d. h. anhand von Testproblemen mit bekannter Lösung verifiziert. Das mit SPH zunächst erstellte thermische Modell des DLIP-Prozesses berücksichtigt die Absorption der Laserstrahlung, die Wärmeleitung im Werkstück und die Enthalpien der Phasenübergänge. Das Modell wird zur Berechnung der Schmelzbadströmung bei der DLIP-Anwendung, angetrieben von Oberflächenspannungsgradienten verursacht durch Temperaturgradienten, erweitert. Hierbei wird eine inkompressible SPH (ISPH) Methode eingesetzt, in der Simulation laserinduzierter Schmelzbäder ein neuartiger Ansatz. Mit dem numerischen Modell werden Simulationen des DLIP-Verfahrens für metallische Substrate durchgeführt. Die thermische Simulation der Strukturierung von Edelstahl stimmt gut mit einem Experiment überein. Weiterhin wird die Anwendung von DLIP auf Edelstahl und Aluminium mit dem thermofluiddynamischen Modell simuliert. Außerdem wird das Modell um eine nichtlinear temperaturabhängige Oberflächenspannung, wie sie für Stahlschmelze in Anwesenheit eines oberflächenaktiven Elements vorliegt, ergänzt. Die Simulationen zeigen ein verschiedenes Verhalten von Edelstahl und Aluminium. Bei der Strukturierung von Aluminium treten ein deutlich tieferes Schmelzbad und erhebliche Geschwindigkeitsbeträge der thermokapillaren Konvektion an der Schmelzeoberfläche auf. Hingegen ist die Strömung bei der DLIP-Anwendung auf Edelstahl schwächer ausgeprägt und höhere Oberflächentemperaturen werden erreicht. Die Marangoni-Konvektion ist daher ein wirksamer Schmelzeverdrängungsmechanismus bei der Strukturierung von Aluminium mit moderater Energiedichte. Die unterschiedliche Schmelzbadströmung für die beiden Werkstoffe wird durch experimentelle Beobachtungen bestätigt. In Abhängigkeit des Schwefelgehalts von Edelstahl zeigen Simulationen verschiedene Strömungsmuster im Schmelzbad und unterstützen die Erklärung experimentell festgestellter Mikrostrukturen. Die Untersuchung der Wirkung der Verdampfung und des induzierten Rückstoßdruckes auf die Strukturausbildung bei höheren Fluenzen erfordert die Berücksichtigung der Oberflächendeformation sowie eine geeignete Druckrandbedingung im ISPH-Algorithmus.:I The research problem 1 Motivation 2 Modelling of laser material processing 2.1 Interaction of laser radiation with materials 2.1.1 Absorption of laser radiation 2.1.2 Heat conduction and phase change 2.1.3 Molten pool convection 2.1.4 Vapourisation regime 2.2 Mathematical modelling of laser material interaction 2.2.1 Conservation equations in Lagrangian formulation 2.2.2 Influence of surface tension 3 State of the art in laser microprocessing and the SPH method 3.1 Laser microprocessing 3.2 Simulation of direct laser interference patterning 3.3 The mesh-free smoothed particle hydrodynamics method 3.3.1 Fundamental approximations and kernel function 3.3.2 Particle distribution and interaction length 3.3.3 Approximation of derivatives 3.3.4 Treatment of boundaries 3.3.5 Neighbourhood search 3.4 Numerical modelling of laser material processing by SPH II SPH model development for direct laser interference patterning 4 SPH modelling of heat transfer and fluid flow 4.1 Solution of the heat diffusion equation 4.2 Formulation of equations governing fluid flow 4.2.1 Equation of continuity 4.2.2 Approximation of pressure gradient term 4.2.3 Treatment of viscosity 4.3 Weakly compressible SPH method for solving fluid flow 4.3.1 Particle motion 4.3.2 Time integration 4.3.3 Time step criteria 4.4 Incompressible SPH method for solving fluid flow 4.4.1 Time integration 4.4.2 Discrete incompressible SPH algorithm 4.4.3 Time step criteria 4.5 Simulation of thermal fluid flow using ISPH 4.5.1 Semi-implicit time integration 4.5.2 Solution of the pressure Poisson equation 5 Verification of the SPH implementation 5.1 Transient heat conduction in laser-irradiated plate 5.1.1 Problem description 5.1.2 Dimensionless formulation 5.1.3 Numerical solution and results 5.2 Viscous flow 5.2.1 Couette flow 5.2.2 Poiseuille flow 5.3 Thermal convection 5.3.1 Natural convection in a square cavity 5.3.2 Rayleigh--Marangoni--Bénard convection in liquid aluminium 6 SPH model of direct laser interference patterning 6.1 Characteristics of the process 6.2 Thermal model 6.2.1 Non-dimensionalisation 6.2.2 Numerical solution of governing equation 6.2.3 Verification of the computation 6.2.4 Numerical test 6.3 Thermofluiddynamic model 6.3.1 Non-dimensionalisation 6.3.2 Numerical solution of governing equations 6.3.3 Discretisation 6.3.4 Resolution independence study 7 SPH simulation of direct laser interference patterning 7.1 Thermal model 7.1.1 DLIP experiments on stainless steel substrates 7.1.2 Thermal simulation of DLIP on steel substrate 7.2 Thermofluiddynamic model 7.2.1 Material properties and simulation parameters 7.2.2 Numerical results for steel substrate 7.2.3 Numerical results for aluminium substrate 7.2.4 Discussion and comparison with experiments 7.3 Extended thermofluiddynamic model 7.3.1 Model parameters 7.3.2 Influence of sulphur content on DLIP of stainless steel 8 Conclusions and outlook Bibliography

In meshfree methods, partial differential equations are solved on an unstructured cloud of points distributed throughout the computational domain. In collocated meshfree methods, the differential operators are directly approximated at each grid point based on a local cloud of neighboring points. The set of neighboring nodes used to construct the local approximation is determined using a variable search radius. The variable search radius establishes an implicit nodal connectivity and hence a mesh is not required. As a result, meshfree methods have the potential flexibility to handle problem sets where the computational grid may undergo large deformations as well as where the grid may need to undergo adaptive refinement. In this work we develop the sharp interface formulation of the immersed boundary method for collocated meshfree approximations. We use the framework to implement three meshfree methods: General Finite Differences (GFD), Smoothed Particle Hydrodynamics (SPH), and Moving Least Squares (MLS). We evaluate the numerical accuracy and convergence rate of these methods by solving the 2D Poisson equation. We demonstrate that GFD is computationally more efficient than MLS and show that its accuracy is superior to a popular corrected form of SPH and comparable to MLS. We then use GFD to solve several canonic steady state fluid flow problems on meshfree grids generated using uniform and variable radii Poisson disk algorithm.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

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.

In this thesis, a new approach is developed, avoiding the correspondence problem caused by the difference in embodiment between imitator and demonstrator in imitation learning. In our work, the imitator is a fluidic system of dynamics totally different than the imitatee, which is a human performing hand gestures and human body postures. The fluidic system is composed of fluid particles, which are used for the discretization of the problem domain. In this work, we demonstrate the fluidics formation control so as to imitate by observation initially given human body poses and hand gestures. Our fluidic formation control is based on setting suitable parameters of Smoothed Particle Hydrodynamics (SPH), which is a particle based Lagrangian method, according to imitation learning. In the controller part, we developed three approaches: In the first one, we used Artificial Neural Networks (ANN) for training of the input-output pairs on the fluidic imitation system. We extracted shape based feature vectors for human hand gestures as inputs of the system and for output we took the fluid dynamics parameters. In the second approach, we employed the Principal Component Analysis (PCA) method for human hand gesture and human body pose classification and imitation. Lastly, we developed a region based controller which assigns the fluid parameters according to the human body poses and hand gestures. In this controller, our algorithm determines the best fitting ellipses on human body regions and human hand finger positions and maps ellipse parameters to the fluid parameters. The fluid parameters adjusted by the fluidics imitation controller are body force (f), density, stiffness coefficient and velocity of particles (V) so as to lead formations of fluidic swarms to human body poses and hand gestures.

This thesis proposes a framework for controlling mobile robot networks based on a fluid dynamics paradigm. The approach is inspired by natural behaviors of fluids demonstrating desirable characteristics for collective robots. The underlying mathematical formalism is developed through establishing analogies between fluid bodies and multi-robot systems such that robots are modeled as fluid elements that constitute a fluid body. The governing equations of fluid dynamics are adapted to multi-robot systems and applied on control of robots. The model governs flow of a robot based on its local interactions with neighboring robots and surrounding environment. Therefore, it provides a layer of decentralized reactive control on low level behaviors, such as obstacle avoidance, deployment, and flow. These behaviors are inherent to the nature of fluids and provide emergent coordination among robots. The framework also introduces a high-level control layer that can be designed according to requirements of the particular task. Emergence of cooperation and collective behavior can be controlled in this layer via a set of parameters obtained from the mathematical description of the system in the lower layer. Validity and potential of the approach have been experimented through simulations primarily on two common collective robotic tasks / deployment and navigation. It is shown that gas-like mobile sensor networks can provide effective coverage in unknown, unstructured, and dynamically changing environments through self-spreading. On the other hand, robots can also demonstrate directional flow in navigation or path following tasks, showing that a wide range of multi-robot applications can potentially be developed using the framework.

High strain rate plastic deformation of metals results in the formation of localized zones of severe shear strain known as adiabatic shear bands (ASBs), which are a precursor to shear failure. The formation of ASBs in a high-strength low alloy steel, namely AISI 4340, was examined based on prior heat treatments (using different austenization and tempering temperatures), testing temperatures, and impact strain rates in order to map out grain size and grain deformation behaviour during the formation of ASBs. In the current experimental investigation, ASB formation was shown to be a microstructural phenomenon which depends on microstructural properties such as grain size, shape, orientation, and distribution of phases and hard particles—all controlled by the heat treatment process. Each grain is unique and its material properties are heterogeneous (based on its size, shape, and the complexity of the microstructure within the grain). Using measurements of grain size at various heat treatments as well as dynamic stress-strain data, a finite element model was developed using Matlab and explicit dynamic software LSDYNA to simulate the microstructural deformation of grains during the formation of ASBs. The model simulates the geometrical grain microstructure of steel in 2D using the Voronoi Tessellation algorithm and takes into account grain size, shape, orientation, and microstructural material property inhomogeneity between the grains and grain boundaries. The model takes advantage of the Smooth Particle Hydrodynamics (SPH) meshless method to simulate highly localized deformation as well as the Johnson-Cook Plasticity material model for defining the behavior of the steel at various heat treatments under high strain rate deformation.The Grain Model provides a superior representation of the kinematics of ASB formation on the microstructural level, based on grain size, shape and orientation. It is able to simulate the microstructural mechanism of ASB formation and grain refinement in AISI 4340 steel, more accurately and realistically than traditional macroscopic models, for a wide range of heat treatment and testing conditions.

This thesis presents a two-phase liquid-solid numerical model using Smoothed Particle Hydrodynamics (SPH). The scheme is developed for multi-phase flows in industrial tanks containing sediment used in the nuclear industry for decommissioning. These two-phase liquid-sediments flows feature a changing interfacial profile, large deformations and fragmentation of the interface with internal jets generating resuspension of the solid phase. SPH is a meshless Lagrangian discretization scheme whose major advantage is the absence of a mesh making the method ideal for interfacial and highly non-linear flows with fragmentation and resuspension. Emphasis has been given to the yield profile and rheological characteristics of the sediment solid phase using a yielding, shear and suspension layer which is needed to predict accurately the erosion phenomena. The numerical SPH scheme is based on the explicit treatment of both phases using Newtonian and non-Newtonian Bingham-type constitutive models. This is supplemented by a yield criterion to predict the onset of yielding of the sediment surface and a suspension model at low volumetric concentrations of sediment solid. The multi-phase model has been compared with experimental and 2-D reference numerical models for scour following a dry-bed dam break yielding satisfactory results and improvements over well-known SPH multi-phase models. A 3-D case using more than 4 million particles, that is to the author’s best knowledge one of the largest liquid-sediment SPH simulations, is presented for the first time. The numerical model is accelerated with the use of Graphic Processing Units (GPUs), with massively parallel capabilities. With the adoption of a multi-phase model the computational requirements increase due to extra arithmetic operations required to resolve both phases and the additional memory requirements for storing a second phase in the device memory. The open source weakly compressible SPH solver DualSPHysics was chosen as the platform for both CPU and GPU implementations. The implementation and optimisation of the multi-phase GPU code achieved a speed up of over 50 compared to a single thread serial code. Prior to this thesis, large resolution liquid-solid simulations were prohibitive and 3-D simulations with millions of particles were unfeasible unless variable particle resolution was employed. Finally, the thesis addresses the challenging problem of enforcing wall boundary conditions in SPH with a novel extension of an existing Modified Virtual Boundary Particle (MVBP) technique. In contrast to the MVBP method, the extended MVBP (eMVBP) boundary condition guarantees that arbitrarily complex domains can be readily discretized ensuring approximate zeroth and first order consistency for all particles whose smoothing kernel support overlaps the boundary. The 2-D eMVBP method has also been extended to 3-D using boundary surfaces discretized into sets of triangular planes to represent the solid wall. Boundary particles are then obtained by translating a full uniform stencil according to the fluid particle position and applying an efficient ray casting algorithm to select particles inside the fluid domain. No special treatment for corners and low computational cost make the method ideal for GPU parallelization. The models are validated for a number of 2-D and 3-D cases, where significantly improved behaviour is obtained in comparison with the conventional boundary techniques. Finally the capability of the numerical scheme to simulate a dam break simulation is also shown in 2-D and 3-D.

Simulating fluid behavior has proven to be a demanding challenge which requires complex computational models and highly efficient data structures. Smoothed Particle Hydrodynamics (SPH) is a particle based computational model used to simulate fluid behavior that has been found capable of producing convincing results. However, the SPH algorithm is computational heavy which makes it cumbersome to work with. This master thesis describes how the SPH algorithm can be accelerated by utilizing the GPU’s computational resources. It describes a model for how to distribute the work load on the GPU and presents a suitable data structure. In addition, it proposes a method to represent and handle moving objects in the fluids surroundings. Finally, the performance gain due to the GPU is evaluated by comparing processing times with an identical implementation running solely on the CPU.