2024-03-29T09:11:21Z
https://cimec.org.ar/ojs/index.php/cimec-repo/oai
oai:ojs.www.cimec.org.ar:article/3734
2011-09-03T18:06:24Z
cimec-repo:ART
"110903 2011 eng "
dc
El Método Científico
Storti, Mario Alberto
Conferencia dictada a docentes de las Escuelas 15 y 579 de Santo Tomé, Santa Fe, dentro del programa Los Científicos Van a las Escuelas
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3734
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4605
2014-07-30T16:26:48Z
cimec-repo:ART
dc
Report of activities at TU Braunschweig
Rios Rodriguez, Gustavo
http://www.cimec.org.ar
The present report aims to show the research activities that were
carried out during the stay of three months at the Institut für
Konstruktionstechnik (IK) of the Technische Universität at the
city of Braunschweig, from 01/04/14 up to 26/06/14. These
activities were developed within the framework of the
International Research Staff Exchange Scheme (IRSES),
project "NumSim" PIRSES - GA2009_246977 Marie Curie Actions,
funded under the 7 framework program of the European
Commission. The final goal of this work plan is to use High
Performance Computing (HPC) resources to obtain a parallel
multiphysics code by combining the advantages of specialized
codes such as elPaSo [2] and PETSc-FEM [7], in order to deal with
the numerical solution of intrincate Fluid-Structure
Interaction (FSI) problems. The activities comprise the
colaborative work between Dr.Ing.Marco Schauer from the German
Institute with Dr.Ing.Luciano Garelli and Dr.Ing.Gustavo Ríos
Rodriguez, both researchers at CIMEC-CONICET.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4605
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1908
2011-09-03T18:04:11Z
cimec-repo:ART
"110903 2011 eng "
dc
Methodologies for the Numerical Simulation of Fluid Flow in Internal Combustion Engines
López, Ezequiel Javier
The increase in the capability of computing in conjunction with the development of new mathematical models and numerical methods, allow to deal with the resolution of complex problems of importance for both science and engineering. Among these, the CFD (Computational Fluid Dynamics) problems in moving domains, such as Fluid-Structure Interaction (FSI) problems, are a topic of particular interest for researchers because of the difficulty that they present and the large number of applications in which these kind of problems are present. One of such problems is the computation of in-cylinder flows in internal combustion (IC) engines.
The modeling of IC engines is a multidisciplinary subject that involves chemical ther- modynamics, fluid mechanics, turbulence, heat transfer, combustion, and numerical meth- ods. In this thesis, the focus is placed on some aspects of the computational resolution of the fluid dynamics problem. In particular, the topics addressed are the mesh dynamics problem, the resolution of flows at low Mach numbers, and the coupling of 1D/multi-D domains for compressible flows. When an Arbitrary Lagrangian Eulerian (ALE) strategy is applied to solve problems with deformable domains, it is necessary to have a Computational Mesh Dynamics (CMD) technique to resolve the dynamics of the mesh. While the movement of the mesh is an artificial field in a FSI problem, its significance is relevant because it affects considerably the efficiency and accuracy of the computation. For in-cylinder flows in IC engines the movement of the boundary domain is known a priori. In these cases the domain has a very high relative deformation and even changes on its topology. This demands great robustness from the CMD strategy to avoid an excessive deterioration of the grid quality and to reduce the number of remeshing needed in the whole simulation.
The flow inside of an IC engine is characterized by a low Mach number, except in the early moments in which the exhaust valve (or port) is opened. The numerical methods for compressible flow based on the density fail when they are applied to flows with low Mach numbers, which is due to the bad conditioning of the system of equations. For this reason, it is necessary to apply a technique that allows the resolution of compressible flows in all the range of Mach numbers, especially in the low Mach limit.
Then, to perform a simulation in an IC engine is necessary to have a CFD code able to compute compressible turbulent flows with low (and also relatively high) Mach numbers in deformable 3D domains [65]. Given the highly complex geometry of the engines and the physical processes that occur within them, it is at present only possible to solve one part of such machines with a 3D model. In this way, and because of its dynamic behavior, another difficulty that appears is related to the boundary conditions to impose to the model. Usually, these problems are addressed by the simulation of the rest of the engine through 0D/1D models, which is achieved in one hand, modeling the entire machine simultaneously (but the level of detail varies depending on the model) and, on the other hand, providing appropriate conditions to the 3D code. Applying the above approximation, the need to couple appropriately the solutions obtained in the computing domains arises, which can be calculated by different codes.
The large spread in length and time scales of in-cylinder flows in IC engines requires a high degree of refinement in the finite element mesh and, then requires very large computational resources. Thus, a parallel code is needed in order to achieve accurate results in that problems. In addition, due to explicit and semi-implicit schemes have demonstrated to be inefficient when they are applied to IC engines [32], a full implicit scheme might be used. [Slides presented during the PHD dissertation Facultad de Ingeniería y Cs. Hídricas from Universidad Nacional del Litoral para acceder 2009-05-04]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1908
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4032
2011-12-29T14:12:46Z
cimec-repo:ART
dc
Fluid Structure Interaction using an Arbitrary Lagrangian Eulerian Formulation
Garelli, Luciano
Multidisciplinary and Multiphysics coupled problems represent nowadays a challenging field when studying even more complex phenomena that appear in nature and in new technologies (e.g. Magneto-Hydrodynamics, Micro-Electro-Mechanics, Thermo-Mechanics, Fluid-Structure Interaction, etc.). Particularly, when dealing with Fluid-Structure Interaction problems several questions arise, namely the coupling algorithm, the mesh moving strategy, the Galilean Invariance of the scheme, the compliance with the Discrete Geometric Conservation Law (DGCL), etc. Therefore, the aim of this thesis is the development and implementation of a coupling algorithm for existing modules or subsystems, in order to carry out FSI simulations with the focus on distributed memory parallel platforms. Regarding the coupling techniques, some results on the convergence of the strong coupling Gauss-Seidel iteration are presented. Also, the precision of different predictor schemes for the structural system and the influence of the partitioned coupling on stability are discussed. Another key point when solving FSI problems is the use of the ‘‘Arbitrary Lagrangian Eulerian formulation’’ (ALE), which allows the use of moving meshes. As the ALE contributions affect the advective terms, some modifications on the stabilizing and the shock-capturing terms, are needed. Also, the movements of the fluid mesh produces a volume change in time of the elements, which adds to the fluid formulation an extra conservation law to be satisfied. The law is known as the Discrete Geometric Conservation Law (DGCL). In this thesis a new and original methodology for developing DGCL compliant formulations based on an Averaged ALE Jacobians Formulation (AJF) is presented. [Dissertation submitted to the Postgraduate Department of FACULTAD DE INGENIERIA Y CIENCIAS HIDRICAS of the UNIVERSIDAD NACIONAL DEL LITORAL in partial fulfillment of the requirements for the degree of Doctor en Ingeniería - Mención Mecánica Computacional. 2011-12-19]
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4032
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/2
2006-06-24T13:10:27Z
cimec-repo:ART
"060624 2006 eng "
dc
Study of a stabilized mixed finite element with emphasis in its numerical performance for strain localization problems
Sanchez, Pablo Javier
Sonzogni, Victorio Enrique
Huespe, Alfredo
The numerical performance of a stabilized mixed FE formulation based on the
pressure-gradient-projection method (PGP) using equal-order (linear) interpolation
is evaluated by solving solid mechanics problems, such as structural limit load de-
termination and strain localization modeling. All of them present incompressibility
kinematical constraints induced by the constitutive behavior. This work is specially
devised to obtain critical conclusions about the use of PGP model when the mechan-
ical response is governed by strain softening macroscopic mechanisms. An additional
contribution is the numerical comparative analysis of two different strategies, for
solving the complete linear equation system, addressed to a FE parallel code. The
numerical results are compared with the standard Galerkin formulation and with
an alternative stabilized mixed finite element procedure (PSPG).
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/2
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1767
2011-09-03T18:03:35Z
cimec-repo:ART
"110903 2011 eng "
dc
Refinamiento h-adaptativo de mallas no estructuradas en problemas estacionarios y transitorios de flujos compresibles
Ríos Rodríguez, Gustavo
La presente tesis trata sobre el desarrollo, la implementación y aplicación de una estrategia de refinamiento adaptativo que permite resolver problemas estacionarios y transitorios de flujos compresibles mediante el método de elementos finitos sobre mallas no estructuradas. La estrategia de adaptación se concibe atendiendo a la calidad de la malla. Se llevan a cabo experiencias numéricas que permiten determinar un esquema de particionamiento para tetraedros que resulta conveniente desde el punto de vista de la calidad y del costo de refinamiento. La calidad de los elementos de la malla es evaluada utilizando una métrica de calidad algebraica, mientras que la calidad de la malla es analizada con un enfoque estadístico, asumiendo que la calidad de los elementos responde a una distribución normal. Las mallas que resultan del proceso de adaptación conservan el tipo de elemento y son no conformes. Luego, la presencia de nodos irregulares obliga a imponer restricciones en el refinamiento para que no exista una diferencia arbitraria en el tamaño de elementos vecinos pertenecientes a regiones con distinto grado de refinamiento. Estas restricciones permiten que el algoritmo de refinamiento sea formulado recursivamente. Por otro lado, dado que no se conoce la solución exacta del problema, se debe recurrir a técnicas de estimación del error en la solución aproximada para identificar las regiones del dominio computacional que necesitan ser refinadas. En este sentido, se describen e implementan dos de los métodos más usuales para la estimación a posteriori del error cuando el sistema de ecuaciones que gobierna el problema es hiperbólico. El código de adaptación de mallas se ejecuta secuencialmente en un nodo cualquiera (no necesariamente el servidor) de un cluster Beowulf, mientras que el programa que resuelve las ecuaciones del flujo se ejecuta en paralelo. La solución de las ecuaciones del flujo está basada en la implementación del método de elementos finitos estabilizado SUPG (Streamline-Upwind / Petrov-Galerkin). La formulación utiliza además técnicas de shock-capturing para el tratamiento de los choques. Dicha implementación es parte del solver multifísica PETSc-FEM desarrollado en el CIMEC (Centro Internacional de Métodos Computacionales en Ingeniería). Por otro lado, el algoritmo de adaptación se implementó en dos códigos. El primero de ellos se desarrolló en lenguaje scripting Octave y fue aplicado en la adaptación de mallas bidimensionales y tridimensionales en la etapa de desarrollo conceptual. Dadas las limitaciones de Octave en cuanto al manejo de los recursos computacionales, una implementación más eficiente se implementó en lenguaje C++. Para ello se recurrió al uso de los contenedores y algoritmos facilitados por las librerías de plantillas estándar (STL) y Boost. Ambos códigos se aplican en la resolución de problemas de flujos no viscosos y compresibles bidimensionales, axisimétricos y tridimensionales de gases perfectos sobre mallas no estructuradas de elementos finitos. Se evalúan la precisión en el cálculo de la solución, la calidad de las mallas adaptadas y la performance del código adaptativo. Los resultados numéricos son comparados con soluciones teóricas procedentes de la dinámica de gases y con resultados de ensayos experimentales. Finalmente se elaboran conclusiones sobre la estrategia de adaptación desarrollada y se proponen una serie de actividades a realizar en el futuro, destinadas a mejorar e incorporar nuevas capacidades a dicha estrategia. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 2009-05-14]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1767
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/2960
2011-09-03T18:05:22Z
cimec-repo:ART
"110903 2011 eng "
dc
Coupling domains through absorbing boundary conditions
Storti, Mario Alberto
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/2960
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4039
2012-07-19T16:12:07Z
cimec-repo:ART
dc
Generation of Turbulent Inlet Velocity Conditions for Large Eddy Simulations
Castro, Hugo Guillermo
Paz, Rodrigo
Slides presented at 10th World Congress on Computational Mechanics (WCCM 2012), 33rd Iberian Latin American Congress on Computational Methods In Engineering (CILAMCE 2012). Sao Paulo, Brazil. (July 2012)
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4039
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/546
2006-11-11T00:48:13Z
cimec-repo:ART
dc
Fluid-structure interaction with a staged algorithm
Storti, Mario Alberto
fluid structure interaction, compressible flow, finite elements
A common approach to solving fluid-structure interaction problems is to solve each subproblem in a partitioned procedure where time and space discretization methods could be different. Such a scheme simplifies explicit/implicit integration and it is in favor of the use of different codes specialized on each sub-area. In this work a staggered fluid-structure coupling algorithm is considered. For each time step a "stage-loop" is performed. In the first stage a high order predictor is used for the structure state, then the fluid and the structure systems are advanced in that order. In subsequent stages of the loop each system uses the previously computed state of the other system until convergence. For weakly coupled problems a stable and efficient algorithm is obtained using one stage and an accurate enough predictor. For strongly coupled problems, stability is enhanced by increasing the number of stages in the loop. If the stage loop is iterated until convergence, a monolithic scheme is recovered. In addition, two items that are specially important in fluid structure problems are discussed, namely invariance of the stabilization terms and dynamic absorbing boundary conditions. Finally, numerical examples are presented.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/546
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1381
2011-09-03T17:57:44Z
cimec-repo:ART
"110903 2011 eng "
dc
Computational Methods for Design and Synthesis of Planar Mechanisms
Pucheta, Martín Alejo
The objective of this research work is the study and development of techniques for the design and synthesis of planar-linkage mechanisms. The synthesis of mechanisms consists in finding the suitable mechanism for a given movement. Particularly, this thesis deals with the problem of synthesis of mechanisms starting from the initial specifications or requirements of design, that is to say, starting from zero. Consequently, it is necessary to determine the number and type of components, and the connectivity between them (type synthesis); and then to calculate the dimensions of the components, pivots positions, and the control parameters of the kinematic pairs of the input movement (dimensional synthesis). This thesis deals with the kinematic synthesis of position, whose problem consists in determining the dimensions of a mechanism that satisfies a desired set of displacements and rotations in certain points of a mechanism and for certain instants of simultaneity. This specification is called kinematic task. The allowed space –for the solution mechanism and the development of the task– is a very common requirement that restricts the solutions to obtain. The problem is highly non-linear. Besides, since it includes the selection of the topology to dimension, it constitutes a discrete problem of combinatorial complexity. In order to solve this difficult problem, it is proposed to use a representation of the mechanism based on the Finite Elements Method and Graph Theory, managing to preserve and unify both representations to integrate the synthesis into its subsequent stages of detailed analysis and optimization of the mechanism. The original theoretical aspects presented in this thesis are: The development of a new identifier of isomorphism of mechanisms and its use in the enumeration of kinematic chains and different atlases of mechanisms. The exhaustive enumeration of topologies using sub-graphs search to satisfy structural requirements from the beginning of the design process. The automatic decomposition of the closed-loop topologies into single open chains to solve their dimensional synthesis using analytical equations expressed by complex-numbers. For the dimensional synthesis, all combinations of single open chains (some of them with multiple solutions) are automatically computed. Among them, that solution which minimizes the summation of link sizes subjected to some design restrictions is retained. In the cases in which there are free parameters, a zeroorder optimization technique based on Genetic Algorithms with penalization of restrictions is applied to sweep the design space. The modifications for extending the methodology to the design of flexible mechanisms using Rigid-Body Replacement methods are developed and analyzed. As the final result of the application of this technique, it is obtained a list of alternatives that constitute good initial conditions for subsequent gradient-based optimization already available in commercial software. Throughout the thesis, various test and validation examples are provided, showing the capacity of the inventive tool developed. [Ph.D dissertation Faculta de Ingeniería y Ciencias Hídricas, Universidad Nacional del Litoral]
CIMEC Document Repository
2011-09-03 15:06:50
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1381
CIMEC Document Repository; 2008
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/3735
2011-09-03T18:06:41Z
cimec-repo:ART
"110903 2011 eng "
dc
Resolución de las ecuaciones de Navier-Stokes utilizando CUDA
Costarelli, Santiago
En este trabajo se propone una resolución de las ecuaciones de
Navier-Stokes para el caso de fluidos Newtonianos incompresibles
utilizando la arquitectura CUDA1 provista por NVIDIA2 . Se utiliza
para la discretización espacial un esquema de diferencias finitas en
grillas staggered (para evitar el desacople en la presión), y el
método de Pasos Fraccionados (Fractional-Step) para la integración
temporal. El paso predictor (problema de advección para las ecuaciones
de cantidad de movimiento) se resuelve utilizando el esquema de
Adams-Bashforth de segundo orden estabilizando los términos
convectivos con el método QUICK. Además, el paso de Poisson (para
imponer la incompresibilidad) es resuelto iterativamente mediante el
método de Gradientes Conjugados precondicionando al sistema utilizando
transformadas rápidas de Fourier. En presente el desarrollo se
utilizan librería estándar de CUDA para el manejo de matrices y
vectores, como ser Thrust3 y CUSP4, además de CUFFT5 para las
transformadas rápidas de Fourier. De esta forma, mediante las
herramientas aportadas por las anteriores se confeccionan los kernels
necesarios enfatizando la utilización de memoria shared, accesos
fusionados a la memoria global, reduciendo al mínimo la cantidad de
registros por thread, entre otros. A continuación se presenta una
serie de casos de estudio con el objetivo de validar el desarrollo y
de, posteriormente, comparar las performances obtenidas con
implementaciones en otras arquitecturas (CPU, unicore y multicore).
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3735
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4947
2015-06-12T19:05:49Z
cimec-repo:ART
"150612 2015 eng "
dc
Assessment of Nuclear Power Reactor Using Computational Fluid Dynamics
Corzo, Santiago F.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4947
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/2667
2011-09-03T18:04:26Z
cimec-repo:ART
"110903 2011 eng "
dc
Coeficientes de Presión en Cubiertas Abovedadas Aisladas
Balbastro, Gustavo
En la presente Tesis se determinan de los coeficientes de presión debidos a la acción del viento sobre construcciones de cubiertas abovedadas sin cerramientos laterales llevada a cabo mediante simulación computacional, así como también la validación experimental del método empleado mediante ensayos en túnel de viento. Pese a que las construcciones de este tipo son muy frecuentes, al menos en la República Argentina, los reglamentos disponibles no aportan información para la valoración de las cargas aerodinámicas sobre estas. Se describe también un modelo experimental nuevo, desarrollado a los fines de esta investigación. Con los resultados alcanzados se proponen valores para el uso práctico en la cuantificación de las cargas para el diseño de tales estructuras y se exponen también las conclusiones obtenidas de la observación de colapsos de estructuras del tipo estudiado y otras similares a estas. [Tesis propuesta para el doctorado en Ingeniería Civil. Universidad Tecnológica Nacional. Facultad Regional Santa Fe]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/2667
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4034
2012-02-01T17:21:11Z
cimec-repo:ART
dc
Fluid Structure Interaction using an Arbitrary Lagrangian Eulerian Formulation
Garelli, Luciano
Multidisciplinary and Multiphysics coupled problems represent nowadays a challenging field when studying even more complex phenomena that appear in nature and in new technologies (e.g. Magneto-Hydrodynamics, Micro-Electro-Mechanics, Thermo-Mechanics, Fluid-Structure Interaction, etc.). Particularly, when dealing with Fluid-Structure Interaction problems several questions arise, namely the coupling algorithm, the mesh moving strategy, the Galilean Invariance of the scheme, the compliance with the Discrete Geometric Conservation Law (DGCL), etc. Therefore, the aim of this thesis is the development and implementation of a coupling algorithm for existing modules or subsystems, in order to carry out FSI simulations with the focus on distributed memory parallel platforms. Regarding the coupling techniques, some results on the convergence of the strong coupling Gauss-Seidel iteration are presented. Also, the precision of different predictor schemes for the structural system and the influence of the partitioned coupling on stability are discussed. Another key point when solving FSI problems is the use of the ‘‘Arbitrary Lagrangian Eulerian formulation’’ (ALE), which allows the use of moving meshes. As the ALE contributions affect the advective terms, some modifications on the stabilizing and the shock-capturing terms, are needed. Also, the movements of the fluid mesh produces a volume change in time of the elements, which adds to the fluid formulation an extra conservation law to be satisfied. The law is known as the Discrete Geometric Conservation Law (DGCL). In this thesis a new and original methodology for developing DGCL compliant formulations based on an Averaged ALE Jacobians Formulation (AJF) is presented. [Slides for the PhD dissertation submitted to the Postgraduate Department of FACULTAD DE INGENIERIA Y CIENCIAS HIDRICAS of the UNIVERSIDAD NACIONAL DEL LITORAL in partial fulfillment of the requirements for the degree of Doctor en Ingeniería - Mención Mecánica Computacional. 2011-12-19]
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4034
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1
2006-06-24T13:10:27Z
cimec-repo:ART
"060624 2006 eng "
dc
Dynamic Boundary Conditions in CFD
Storti, Mario Alberto
Nigro, Norberto Marcelo
Paz, Rodrigo Rafael
Dalcin, Lisandro Daniel
The number and type of boundary conditions to be used in the
numerical modeling of fluid mechanics problems is normally chosen
according to a simplified analysis of the characteristics, and also
from the experience of the modeler. The problem is harder at
input/output boundaries which are, in most cases, artificial
boundaries, so that a bad decision about the boundary conditions to
be imposed may affect the precision and stability of the whole
computation. For inviscid flows, the analysis of the sense of
propagation in the normal direction to the boundaries gives the
number of conditions to be imposed and, in addition, the conditions
that are absorbing for the waves impinging normal to the
boundary. In practice, it amounts to counting the number of positive
and negative eigenvalues of the advective flux Jacobian projected
onto the normal. The problem is still harder when the number of
incoming characteristics varies during the computation, and to
correctly treat these cases poses both mathematical and practical
problems. One example considered here is compressible flow where the
flow regime at a certain part of an inlet/outlet boundary can change
from subsonic to supersonic and the flow can revert. In this work
the technique for dynamically imposing the correct number of
boundary conditions along the computation, using Lagrange
multipliers and penalization is discussed, and several numerical
examples are presented.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1765
2011-09-03T18:03:16Z
cimec-repo:ART
"110903 2011 eng "
dc
Elementos Finitos Estabilizados para Flujos con Superficie Libre: Seguimiento y Captura de Interfase
Battaglia, Laura
En esta Tesis se considera la resolución de problemas transientes de flujo con superficie libre de fluidos incompresibles y viscosos, con comportamiento newtoniano, en régimen de flujo laminar e isotérmico y sin influencia de la tensión superficial, tanto para dos como para tres dimensiones.Para ello, se proponen dos estrategias complementarias resueltas mediante elementos finitos estabilizados en el espacio y por diferencias finitas en el tiempo. La primer metodología es una técnica de seguimiento de interfase enmarcada en una formulación lagrangiana-euleriana arbitraria en la cual se resuelven en cada paso de tiempo las ecuaciones de Navier--Stokes en una única fase líquida, para luego determinar el desplazamiento de la superficie libre y, finalmente, readaptar la malla de elementos finitos a la nueva conformación geométrica del dominio espacial. En este paradigma, la atención está puesta en evitar inestabilidades numéricas espurias durante el movimiento de la superficie libre, ya sea mediante un operador de suavizado, o bien a través de una ecuación de transporte de la elevación nodal de la superficie libre estabilizada numéricamente. Este método es aplicable a problemas con pequeños desplazamientos de la superficie libre, o eventualmente deformaciones de mayor amplitud que no produzcan pérdida de unicidad en la interfase.La segunda estrategia consiste en un método de captura de interfase basado en una función de nivel para la cual se simula el flujo de dos fluidos, cada uno de ellos indicado con un valor positivo o negativo de dicha función, tal que el valor cero constituye naturalmente la interfase. En este caso, se consideran tres etapas acopladas: la primera provee la resolución de las ecuaciones de Navier--Stokes para el caso de fluido no homogéneo, esto es, considerando que las propiedades físicas de cada fase dependen del valor de la función de nivel en el elemento a evaluar; la segunda etapa resuelve el transporte de la función de nivel mediante un paso de advección convencional; la tercer etapa consiste en una reinicialización de la función de nivel tal que se verifiquen ciertas propiedades en la transición entre los fluidos, a través de un operador diferencial con dos variantes: continua y discontinua. En la primera de estas variantes, la reinicialización se realiza por medio de un operador continuo en la transición líquido/gas y exhibe un comportamiento superior con respecto a la restante, discontinua en la frontera. Esta segunda metodología es aplicable a desplazamientos medianos y grandes de la superficie libre, pues admite su rotura o que se pliegue sobre sí misma, sin que la pérdida de unicidad de la interfase invalide su aplicación.Las dos propuestas presentadas son abordadas numéricamente a través de la alternancia de etapas, cada una de las cuales es resuelta mediante programación distribuida, lo cual inscribe ambos métodos en un paradigma multiparalelo. Los resultados obtenidos para diversos problemas de solución conocida permiten validar ambas estrategias, incluyendo la convergencia en malla. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 2009-04-23]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1765
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/3731
2011-09-03T18:05:41Z
cimec-repo:ART
"110903 2011 eng "
dc
Dynamic Absorbing Boundary Conditions for Advective-Difusive Systems with Unknown Riemann Invariants
Storti, Mario Alberto
Slides o Dynamic Boundary Conditions
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3731
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4040
2012-10-05T13:53:22Z
cimec-repo:ART
dc
Absorbing Boundary Conditions for Compressible Flow at Low Mach Number and Free Surface Flows with Level-Set
Storti, Mario Alberto
This report compiles several issues related with absorbing boundary conditios for several problems in computational fluid mechanics
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4040
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/547
2006-12-23T16:08:50Z
cimec-repo:ART
dc
Fluid-structure interaction with a staged algorithm. Aerospace applications.
Storti, Mario Alberto
Nigro, Norberto Marcelo
Paz, Rodrigo Rafael
Dalcín, Lisandro Daniel
Ríos Rodríguez, Gustavo
López, Ezequiel
A common approach to solving fluid-structure interaction problems is to solve each subproblem in a partitioned procedure where time and space discretization methods could be different. Such a scheme simplifies explicit/implicit integration and it is in favor of the use of different codes specialized on each sub-area. In this work a staggered fluid-structure coupling algorithm is considered. For each time step a "stage-loop" is performed. In the first stage a high order predictor is used for the structure state, then the fluid and the structure systems are advanced in that order. In subsequent stages of the loop each system uses the previously computed state of the other system until convergence. For weakly coupled problems a stable and efficient algorithm is obtained using one stage and an accurate enough predictor. For strongly coupled problems, stability is enhanced by increasing the number of stages in the loop. If the stage loop is iterated until convergence, a strongly coupled scheme is recovered. In addition, two items that are specially important in fluid structure problems are discussed, namely invariance of the stabilization terms and dynamic absorbing boundary conditions. Finally, numerical examples in aerospace applications are presented.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/547
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1093
2011-09-03T18:08:04Z
cimec-repo:ART
"110903 2011 eng "
dc
Particle Trace Visualization Technique
Storti, Mario Alberto
Ríos Rodríguez, Gustavo A.
Nigro, Norberto
Dalcín, Lisandro
Paz, Rodrigo
Particle tracing is a visualization technique that gives to the observer the simultaneous perception of direction and velocity of the flow. An efficient algorithm is presented allowing the computation of particle trajectories for unsteady flows in moving meshes, including also the computation of skin-friction particles, i.e. particles bound to the body skin. Visualization with the OpenDX package, as well as the parallel implementation in SMP architectures is also described. Discussion of visualization parameters, as particle density, and frame rate in order to avoid coherence effect is discussed. [Slides presented at Enief 2007]
CIMEC Document Repository
2007-02-11 23:59:33
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1093
CIMEC Document Repository; 2007
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/851
2011-09-03T18:07:33Z
cimec-repo:ART
"110903 2011 eng "
dc
Métodos iterativos para la solución de problemas lineales y no-lineales
Storti, Mario Alberto
Paz, Rodrigo R.
Notas del curso
CIMEC Document Repository
2007-02-11 23:59:33
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/851
CIMEC Document Repository; 2007
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1320
2008-08-01T17:26:21Z
cimec-repo:ART
dc
Desarrollos en Mecánica Computacional de Falla Material. Aplicación a Modelos de Plasticidad y Daño
Sánchez, Pablo Javier
En este trabajo de investigación se estudian y desarrollan formulaciones de elemen- tos finitos para la simulación numérica del fenómeno de localización de deformaciones inducido por el comportamiento constitutivo. Se analizan y plantean estrategias capaces de capturar las características más importantes del problema, incluyendo en este mo- delado el desarrollo de discontinuidades en el sólido (fisuras, fracturas, bandas de corte o de deslizamiento, según el contexto) las cuales se propagan en un medio inicialmente continuo, como consecuencia última del proceso macroscópico de degradación material. Los conceptos teóricos aquí desarrollados se fundamentan en las siguientes hipótesis básicas: i) régimen quasi estático isotérmico, ii) pequeñas deformaciones, desplazamien- tos y rotaciones, iii) material homogéneo e isótropo dotado de ecuaciones constitutivas susceptibles de presentar inestabilidad material por la presencia de ablandamiento por deformación. El problema de falla se encara desde diferentes puntos de vista. En primera instancia, se estudian estrategias no locales formuladas, alternativamente, en términos de un modelo de gradientes implícito (anexo A). Luego de reconocer algunas limitaciones en estos esquemas de regularización, al menos desde el punto de vista práctico ingenieril, en el contenido de la tesis se introduce la aproximación al problema mediante los denominados modelos cohesivos basados en elementos finitos con discontinuidades embebidas. En particular adoptamos como estrategia principal en este trabajo la aproximación por discontinuidades fuertes del continuo (Continuum Strong Discontinuity Approach CSDA), ver capítulo 3. Se presenta además, un extenso y comparativo análisis sobre las dos familias de ele- mentos finitos con discontinuidades fuertes embebidas potencialmente más eficaces para simular fractura frágil, tomando como base un modelo de daño isótropo. Tras realizar una eficiente implementación de ambas tecnologías de elementos, a partir de este estudio (capítulo 4) se obtienen conclusiones rigurosas con respecto a tópicos de fundamental im- portancia en el análisis de falla y que en la actualidad no están suficientemente claras, a saber: robustez, velocidad de convergencia, exactitud y costo computacional. Otro aporte novedoso, en el contexto teórico de la CSDA, es la formulación matemáti- ca e implementación de un nuevo elemento finito para el modelado de bandas de corte en plasticidad isocórica (materiales gobernados por una ley de plasticidad de Von Mis- es), véase capítulo 6. El elemento propuesto se basa en acoplar consistentemente una formulación mixta estabilizada de base (para evitar el bloqueo volumétrico en la etapa pre-bifurcación, típico en modelos J2 ) con la regularización constitutiva y enriquecimiento cinemático que aporta la aproximación por discontinuidades fuertes utilizada. Previamente, en el capítulo 5, se realiza un riguroso estudio sobre el desempeño numéri- co del esquema de estabilización utilizado, aquí denominado PGP, especialmente direc- cionado a evaluar su desempeño para la simulación del fenómeno de localización de de- formaciones. El mismo está formulado bajo el concepto de sub-escalas ortogonales, véase apéndice B. Como aporte adicional, en el apéndice C, mostramos además la performance numérica de su implementación en un entorno de cálculo distribuido. Finalmente, y a manera de validación, se muestran diversos ejemplos numéricos que representan verdaderos desafíos desde el punto de vista de la simulación computacional, incluyendo casos 3D, mostrando la performance de las estrategias utilizadas y desarrolla- das en esta tesis. [Tesis presentada para obtener el título de Doctor en Ingeniería, FICH-UNL, defensa 2008-03-06]
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1320
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4036
2012-04-11T18:56:02Z
cimec-repo:ART
dc
A FFT Preconditioning Technique for the Solution of Incompressible Flow on GPU's
Storti, Mario Alberto
The resolution of Computational Fluid Dynamics (CFD) problems on Graphic Processing Units (GPU's) requires of specialized algorithms due to the particular hardware architecture of these devices. Algorithms that fall in the category of cellular automata (CA) are the best fitted, for instance explicit Finite Volume or Finite Element methods. But in the case of incompressible flow it is not possible to develop a pure explicit algorithm, due to the essentially non-local character of the incompressibility condition. In this case the algorithms that are closer to an explicit approach, are segregated algorithms, like the Fractional Step Method. In these algorithms the more time consuming stage is (asymptotically for large problems) the solution of the Poisson's equation for pressure. A common choice for it's solution is the IOP (Iterated Orthogonal Projection) method, which requires a series of solutions on the complete mesh. In this work a variant of the IOP, called Accelerated Global Preconditioning (AGP), is proposed. It is based on using a Preconditioned Conjugate Gradient (which is an accelerated iterative method, in contrast with the stationary scheme used in IOP) for the pressure on the fluid, and preconditioning with the solution on the global domain (fluid and solid). [Slides of talk given at CIMNE (Barcelona, Spain), 2012-04-11]
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4036
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/452
2006-09-13T02:04:03Z
cimec-repo:ART
dc
A Pseudo-Spectral Approach for the Incompressible Boundary Layer Equations with Automatic Normal Scaling
Storti, Mario Alberto
Partial differential equations, boundary value problems, boundary layer
A pseudo-spectral numerical method for the solution of the incompressible 2D boundary layer equations is presented. The method is based on Fourier expansion in the normal transformed coordinate, similarly to the transformation that leads to the polynomial Tchebichev expansion in finite intervals, but appropriated to semi-infinite intervals, so that no extra parameter is needed for the outer boundary of the layer. A scaling is applied to the normal coordinate but with the innovation that it is based on the computed boundary layer thickness, i.e. not assuming a priori a variation for it. Spectral decay of the expansion coefficients is shown for the similar solution to the family of wedge flows. Also, spectral convergence of the error is shown for the case of a convergent channel (one of the similar "wedge flows"), for which an analytical solution is available. The method pretends to have a good performance also when using very few parameters, so that results with four terms in the Fourier series (it amounts to two independent parameters) are compared with the well known method from von Karman and Pohlhausen.
For 3D problems, the boundary layer equations are solved in a completely general mesh and coordinate system on the surface using the tensorial form of the equations. In order to advance the solution in the streamwise coordinate a mesh-less approximation is used in the coordinates on the surface. This feature allows the treatment of very general geometries. 3D numerical examples include the yawed cylinder and flat plate, and 3D axisymmetric flows like the cone, the sphere and a rotating sphere.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/452
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1764
2011-09-03T17:59:28Z
cimec-repo:ART
"110903 2011 eng "
dc
Métodos Numéricos para el Problema de la Resistencia de Ola en Barcos
D'Elía, Jorge
Se presenta una condición de frontera absorbente Discreta No-Local
(DNL) para el problema de la resistencia de ola en barcos. Esta
condición es completamemte absorbente en el sentido de que la solución
es independiente de la posición del borde corriente abajo y es no
local en el sentido de que representa matrices llenas que conectan
todas las incógnitas en dos capas consecutivas en los planos de
entrada y de salida. La misma es implementada en dos formas
principales: por un modelo de elementos finitos, y por un modelo de
paneles (o de elementos de borde). La implementación por elementos
finitos es obtenida a partir de un análisis directo de las ecuaciones
en diferencias resultantes, asumiendo que la malla es
unidimensionalmente estructurada (en la dirección longitudinal),
mientras que la implementación por paneles está acoplada con una
descomposición finita de Fourier, sobre un borde artificial corriente
abajo, dando un problema equivalente que es resuelto en un dominio
acotado. En este caso el arrastre es computado mediante la clásica
integración de la presión sobre el casco mojado en condiciones
hidrostáticas, y las alturas de ola en la superficie libre corriente
abajo del borde artificial son obtenidas como un procedimiento de
pos-procesamiento. La condición de borde absorbente DNL muestra tres
aspectos. Primero, en contraste con los métodos cuasi-Dawson evita el
empleo de viscosidades numéricas en la discretización, de modo que un
esquema de segundo orden centrado puede emplearse en el operador de
superficie libre. Segundo, asimismo permite considerar regiones más
reducidas para la superficie libre, con el consecuente ahorro en los
recursos computacionales. Tercero, el uso de un esquema centrado para
el operador de superficie libre permite una discretización completa
por elementos finitos, donde el arrastre es luego computado por un
balance de flujo, el cual es más exacto y garantiza resistencias
positivas. Los resultados numéricos incluyen la estela de un ferry a
quince esloras. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 1997-12-22]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1764
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/3732
2011-09-03T18:05:53Z
cimec-repo:ART
"110903 2011 eng "
dc
An FFT Preconditioning Technique for the Solution of Incompressible Flow with Fractional Step Methods on GPGPU’s
Storti, Mario Alberto
The resolution of Computational Fluid Dynamics (CFD) problems on Graphic Processing Units (GPU’s) requires of specialized algorithms due to the particular hardware architecture of these devices. Algorithms that fall in the category of cellular automata (CA) are the best fitted, for instance explicit Finite Volume or Finite Element methods. But in the case of incompressible flow it is not possible to develop a pure explicit algorithm, due to the essentially non-local character of the incompressibility condition. In this case the algorithms that are closer to an explicit approach, are segregated algorithms, like the Fractional Step Method. In these algorithms the more time consuming stage is (asymptotically for large problems) the solution of the Poisson’s equation for pressure. A common choice for it’s solution is the IOP (Iterated Orthogonal Projection) method, which requires a series of solutions on the complete mesh. In this work a variant of the IOP, called Accelerated Global Preconditioning (AGP), is proposed. It is based on using a Preconditioned Conjugate Gradient (which is an accelerated iterative method, in contrast with the stationary scheme used in IOP ) for the pressure on the fluid, and preconditioning with the solution on the global domain (fluid and solid). Of course, solving the problem on the global domain represents more computational work than solving the problem only in the fluid, but this can be faster in a structured mesh if a fast solvers as Multigrid or Fast Fourier Transform (FFT) is used. The main advantage of AGP over IOP is that it is an accelerated solver, whereas the IOP is stationary. In addition AGP iterates only on pressure, whereas IOP iterates on both pressure and velocity.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3732
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4331
2012-11-18T19:45:49Z
cimec-repo:ART
dc
A FFT Preconditioning Technique for the Solution of Incompressible Flow
Storti, Mario Alberto
Costarelli, Santiago
Paz, Rodrigo Rafael
Dalcin, Lisandro Daniel
Idelsohn, Sergio Rodolfo
computational mechanics, numerical methods, computational fluid dynamics, computational structural dynamics
The Accelerated Global Preconditioning (AGP) algorithm for the solution of
the Poisson equation specially oriented to the solution of Navier-Stokes
equations on GPU hardware was presented. It shares some features with the
well known IOP iteration scheme.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4331
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/550
2007-05-11T17:29:03Z
cimec-repo:ART
dc
Métodos de Descomposición de Dominio en Mecánica de Fluidos Computacional (Ph.D talk)
Paz, Rodrigo Rafael
[Slides, oral defense]
The main purpose of the present thesis work is the efficient solution
of large scale problems arising in Computational Fluid Dynamics
challenge problems, the proposition of new ideas in preconditioning
techniques, the implementation of such ideas in a parallel
multiphysics C++ code using the message passing paradigm via MPI/PETSc
libraries and its evaluation on a Beowulf class cluster. These topics
are presented in the first part of this work. The second part is
devoted to the application of the algorithm proposed in the first part
to the solution of more general/complex problems like the wave
absorption on fictitious boundaries and the resolution of
fluid-structure problems in the supersonic regime of a compressible
fluid flow.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/550
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/3736
2011-09-03T22:14:51Z
cimec-repo:ART
dc
Generación de Mallas Tridimensionales por Métodos Duales
Calvo, Néstor
En esta tesis se presentan algunos desarrollos implementados para construir mallas poliédricas por métodos duales. La dualidad entre una malla construida mediante la Triangulación de Delaunay y la estructura geométrica provista por el Diagrama de Voronoï de sus nodos se explota y se generaliza en dos direcciones distintas. Una de ellas conduce a una teselación de excelente calidad con Poliedros Esféricos, éstos son poliedros que tienen sus nodos en las cercanías de una superficie esférica y cuya unión es una malla que se aproxima al dual geométrico y topológico del Diagrama de Voronoï. Para utilizar estas poliedrizaciones, tanto en el Método de Elementos Finitos como en Métodos de Partículas, se implementaron y se explican aquí las Coordenadas Laplacianas Elementales utilizadas como funciones de forma para la interpolación. Para problemas que incluyen una superficie libre se desarrollo una variante del método de Alpha-Shapes, modificado para reconocer la superficie libre de una distribución de puntos con densidad variable, tal como la requieren estos métodos de cálculo. La otra dirección hacia la cual se extendió el concepto de dualidad consiste en un análisis de la dualidad entre las mallas de hexaedros y un conjunto de superficies mutuamente interceptadas. Dicho análisis se aprovechó para desarrollar algoritmos útiles para la generación de mallas de hexaedros. Para completar el desarrollo se implementaron y describen aquí, algunos métodos y rutinas de tratamiento de las mallas de frontera y los mecanismos de optimización de la calidad de dichas mallas. [PhD Thesis dissertation presented ad Universidad Nacional del Litoral (2005)]
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3736
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1087
2011-09-03T18:07:44Z
cimec-repo:ART
"110903 2011 eng "
dc
Desarrollo de un código de interacción viscosa-invíscida orientado a turbomaquinaria
Prado, Ricardo A.
boundary layer; wind turbine
Se describe la metodología utilizada para analizar el comportamiento fluidodinámico de un tipo particular de turbomáquina, la turbina eólica de eje horizontal. Se considera que las palas de la turbina, que presentan torsión y curvatura, trabajan bajo condiciones de rotación y viento incidente uniformes. El análisis se realiza mediante un proceso de interacción entre la zona del campo fluidodinámico en donde los efectos viscosos pueden despreciarse y la región próxima a las superficies de la pala, donde los efectos viscosos son preponderantes. La zona externa, caracterizada por un flujo invíscido, es resuelta mediante la aplicación del método de los paneles, técnica que representa tanto la pala como su estela helicoidal mediante un entramado de hilos vorticosos, cada uno de ellos con una circulación asociada. La capa viscosa que se desarrolla sobre las superficies es resuelta por medio de la discretización de las ecuaciones de la capa límite tridimensional de Prandtl, bajo condiciones estacionarias, en un sistema coordenado curvilíneo no ortogonal, implementando la técnica de las diferencias finitas. Ambas soluciones numéricas están acopladas pues el flujo invíscido actúa de condición de contorno para la solución viscosa. El análisis del flujo en la capa viscosa se limita a una condición de régimen laminar, razón por la cual el presente estudio se restringe a turbinas eólicas de pequeño porte. Entre las contribuciones de la presente tesis se destacan la modificación del modelo de Dyment para la representación de turbinas eólicas mediante la teoría de los discos actuadores y el desarrollo de las ecuaciones de la capa límite estacionaria tridimensional en componentes físicas contravariantes en un sistema coordenado curvilíneo generalizado (no- ortonormal ni homogéneo) solidario a una pala rotante.
CIMEC Document Repository
2007-02-11 23:59:33
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1087
CIMEC Document Repository; 2007
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/2958
2011-09-03T18:04:49Z
cimec-repo:ART
"110903 2011 eng "
dc
The FastMat2 Matrix Library. Description and Parallel Implementation
Storti, Mario Alberto
Castro, Hugo Guillermo
Paz, Rodrigo Rafael
Dalcín, Lisandro Daniel
Finite element codes usually have two levels of programming. At the outer level a series of nested loops for time stepping, or Newton iteration, leads to a series of linear problems that require residual and matrix assemblies which are then solved with iterative solvers like CG (for Conjugate Gradient) or GMRES (Generalized Minimal Residual Method). At the inner level, these assemblies are performed with a loop over all the elements in the mesh. The residual and matrix are computed for each element assembled in the global vector/matrix. The FastMat2 class is the main matrix library used in the PETSc-FEM code (http://www.cimec.org.ar/petscfem) for computations at the element level. It is full multi-index and performs the standard matrix functions like addition, tensor contractions (in the matrix sense and element-by-element), eigenvalue decomposition, inversion. A key point in its design is that many logical operations (row or column selections for instance) are the same for each element. So this kind of operations are cached in a Direct Acyclic Graph structure (DAG). Its implementation, in particular in a shared memory with OpenMP (for Open Multi-Processing) environment is discussed.
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/2958
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4037
2012-07-18T22:14:18Z
cimec-repo:ART
dc
Absorbing Boundary Conditions for Free-Surface Flows in Open Domains
Storti, Mario Alberto
The presented scheme is a first order AL for the NS equations with free surface. Good absorption is observed with absorbing layer of 1 wavelength or less. Extension to higher order AL's or ABC's is an ongoing work. The AL has been shown for a tracking surface ALE formulation, but can be extended to Level-Set or VOF schemes as well.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4037
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/454
2006-09-13T02:21:30Z
cimec-repo:ART
dc
A Pseudo-Spectral Approach for the Incompressible Boundary Layer Equations with Automatic Normal Scaling
Storti, Mario Alberto
A pseudo-spectral numerical method for the solution of the incompressible 2D boundary layer equations is presented. The method is based on Fourier expansion in the normal transformed coordinate, similarly to the transformation that leads to the polynomial Tchebichev expansion in finite intervals, but appropriated to semi-infinite intervals, so that no extra parameter is needed for the outer boundary of the layer. A scaling is applied to the normal coordinate but with the innovation that it is based on the computed boundary layer thickness, i.e. not assuming a priori a variation for it. Spectral decay of the expansion coefficients is shown for the similar solution to the family of wedge flows. Also, spectral convergence of the error is shown for the case of a convergent channel (one of the similar "wedge flows"), for which an analytical solution is available. The method pretends to have a good performance also when using very few parameters, so that results with four terms in the Fourier series (it amounts to two independent parameters) are compared with the well known method from von Karman and Pohlhausen.
For 3D problems, the boundary layer equations are solved in a completely general mesh and coordinate system on the surface using the tensorial form of the equations. In order to advance the solution in the streamwise coordinate a mesh-less approximation is used in the coordinates on the surface. This feature allows the treatment of very general geometries. 3D numerical examples include the yawed cylinder and flat plate, and 3D axisymmetric flows like the cone, the sphere and a rotating sphere.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/454
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1388
2011-09-03T17:59:10Z
cimec-repo:ART
"110903 2011 eng "
dc
Techniques for High-Performance Distributed Computing in Computational Fluid Mechanics
Dalcín, Lisandro Daniel
Although a lot of progress has been made in theory as well as practice, the true costs of accessing parallel environments are still largely dominated by software. The number of end-user parallelized applications is still very small, as well as the number of people affected to their development. Engineers and scientists not specialized in programming or numerical computing, and even small and medium size software companies, hardly ever considered developing their own parallelized code. High performance computing is traditionally associated with software development using compiled languages. However, in typical applica- tions programs, only a small part of the code is time-critical enough to require the efficiency of compiled languages. The rest of the code is generally related to memory management, error handling, input/output, and user interaction, and those are usually the most error-prone and time-consuming lines of code to write and debug in the whole development process. Interpreted high-level languages can be really advantageous for these kind of tasks. This thesis reports the attempts to facilitate the access to high-performance parallel computing resources within a Python programming environment. The target audience are all members of the scientific and engineering community using Python on a regular basis as the supporting environment for develop- ing applications and performing numerical simulations. The target computing platforms range from multiple-processor and/or multiple-core desktop comput- ers, clusters of workstations or dedicated computing nodes either with stan- dard or special network interconnects, to high-performance shared memory machines. The net result of this effort are two open source and public domain packages, MPI for Python (known in short as mpi4py) and PETSc for Python (known in short as petsc4py).
MPI for Python [8, 9, 10], is an open-source, public-domain software project that provides bindings of the Message Passing Interface (MPI) standard for the Python programming language. MPI for Python is a general-purpose and full-featured package targeting the development of parallel application codes in Python. Its facilities allow parallel Python programs to easily exploit multiple processors. MPI for Python employs a back-end MPI implementation, thus being immediately available on any parallel environment providing access to any MPI library. PETSc for Python [11] is an open-source, public-domain software project that provides access to the Portable, Extensible Toolkit for Scientific Computation (PETSc) libraries within the Python programming language. PETSc for Python is a general-purpose and full-featured package. Its facilities allow sequential and parallel Python applications to exploit state of the art algorithms and data structures readily available in PETSc. MPI for Python and PETSc for Python packages are fully integrated to PETSc-FEM [12], an MPI and PETSc based parallel, multiphysics, finite el- ements code. Within a parallel Python programming environment, this soft- ware infrastructure supported research activities related to the simulation of electrophoretic processes in microfluidic chips. This work is part of a mul- tidisciplinary effort oriented to design and develop these devices in order to improve current techniques in clinical analysis and early diagnosis of cancer. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 2008-06-26]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1388
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/3733
2011-09-03T18:06:11Z
cimec-repo:ART
"110903 2011 eng "
dc
Supercomputadoras en Carrera
Storti, Mario Alberto
Café Científico, Chopería Santa Fe, Santa Fe, 2010-04-29
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/3733
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4340
2014-07-30T16:27:39Z
cimec-repo:ART
dc
Inlet Turbulence Generation for Large Eddy Simulation: Applications to Fluid-Structure Interaction Problems
Castro, Hugo Guillermo
The Large Eddy Simulation (LES) method has become practically a stan-
dard approach for the resolution of the Navier-Stokes (NS) equations
when the simulation of a turbulent flow is required. This is not only
due to the high accuracy and feasibility of LES but also to the
growing computational power and affordability that have taken place in
recent years. However, there are a number of issues that are still
being intensively stud- ied: mesh generation strategies, inlet and
boundary conditions, subgrid- scale (SGS) models, among others, in
order to extend the set of problems that can be solved by LES or to
reduce the computational cost involved. This scenario is further
complicated if the fluid-structure interaction (FSI) problem is added:
coupling algorithm and mesh moving strategy must be defined for every
problem. A particular field where the Computational Fluid Dynamics
(CFD) and FSI meet is in the road vehicle aerodynamics study. Until
recently (and even at present days) it has been preferred the
experimental rather than numerical simulation of such problems due to
the reliability gained by wind tunnels. Nevertheless, this
experimental tool has also shortcomings that can be corrected and even
strengths that can be improved by the use of numerical simulation. It
is the aim of this thesis to study the applicability of LES on road
vehicle aerodynamics including inlet turbulence generation and
fluid-structure in- teraction. In order to do that, a turbulent flow
over a simplified car model known as Ahmed's body is simulated. The
results obtained in this work lead to the conclusion that it is
possible to analyze the aerodynamic prop- erties of road vehicle
models in a complementary way with experimental and computational
tools.
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4340
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1907
2011-09-03T18:03:55Z
cimec-repo:ART
"110903 2011 eng "
dc
Methodologies for the Numerical Simulation of Fluid Flow in Internal Combustion Engines
López, Ezequiel Javier
The increase in the capability of computing in conjunction with the development of new mathematical models and numerical methods, allow to deal with the resolution of complex problems of importance for both science and engineering. Among these, the CFD (Computational Fluid Dynamics) problems in moving domains, such as Fluid-Structure Interaction (FSI) problems, are a topic of particular interest for researchers because of the difficulty that they present and the large number of applications in which these kind of problems are present. One of such problems is the computation of in-cylinder flows in internal combustion (IC) engines.
The modeling of IC engines is a multidisciplinary subject that involves chemical ther- modynamics, fluid mechanics, turbulence, heat transfer, combustion, and numerical meth- ods. In this thesis, the focus is placed on some aspects of the computational resolution of the fluid dynamics problem. In particular, the topics addressed are the mesh dynamics problem, the resolution of flows at low Mach numbers, and the coupling of 1D/multi-D domains for compressible flows. When an Arbitrary Lagrangian Eulerian (ALE) strategy is applied to solve problems with deformable domains, it is necessary to have a Computational Mesh Dynamics (CMD) technique to resolve the dynamics of the mesh. While the movement of the mesh is an artificial field in a FSI problem, its significance is relevant because it affects considerably the efficiency and accuracy of the computation. For in-cylinder flows in IC engines the movement of the boundary domain is known a priori. In these cases the domain has a very high relative deformation and even changes on its topology. This demands great robustness from the CMD strategy to avoid an excessive deterioration of the grid quality and to reduce the number of remeshing needed in the whole simulation.
The flow inside of an IC engine is characterized by a low Mach number, except in the early moments in which the exhaust valve (or port) is opened. The numerical methods for compressible flow based on the density fail when they are applied to flows with low Mach numbers, which is due to the bad conditioning of the system of equations. For this reason, it is necessary to apply a technique that allows the resolution of compressible flows in all the range of Mach numbers, especially in the low Mach limit.
Then, to perform a simulation in an IC engine is necessary to have a CFD code able to compute compressible turbulent flows with low (and also relatively high) Mach numbers in deformable 3D domains [65]. Given the highly complex geometry of the engines and the physical processes that occur within them, it is at present only possible to solve one part of such machines with a 3D model. In this way, and because of its dynamic behavior, another difficulty that appears is related to the boundary conditions to impose to the model. Usually, these problems are addressed by the simulation of the rest of the engine through 0D/1D models, which is achieved in one hand, modeling the entire machine simultaneously (but the level of detail varies depending on the model) and, on the other hand, providing appropriate conditions to the 3D code. Applying the above approximation, the need to couple appropriately the solutions obtained in the computing domains arises, which can be calculated by different codes.
The large spread in length and time scales of in-cylinder flows in IC engines requires a high degree of refinement in the finite element mesh and, then requires very large computational resources. Thus, a parallel code is needed in order to achieve accurate results in that problems. In addition, due to explicit and semi-implicit schemes have demonstrated to be inefficient when they are applied to IC engines [32], a full implicit scheme might be used. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 2009-05-04]
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1907
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4031
2011-11-05T19:28:28Z
cimec-repo:ART
dc
Automatic high order absorption layers for advective-diffusive systems of equations
Storti, Mario Alberto
Reflection of waves at artificial boundaries degrade the quality of numerical solutions and can prevent convergence to the infinite domain solution. Absorbing boundary conditions (ABC's) eliminate this reflections ensuring convergence to the infinite domain solution. High order ABC's reduce further the computational cost by allowing shorter domains for the same level of precision. The PML technique (Perfectly Matched Layer) has been applied in order to obtain very efficient high order ABC's, but however, they tend to be unstable. In addition, the development of the PML absorbing term is specific to each physical problem. In this paper we propose a technique that develops an ABC for any advective-diffusive problem to any order and has the practical advantage that can be implemented only in terms of the computation of the advective fluxes. The technique is based on the standard expansion of the absorbing term in terms of lateral wave number and computing the matrix coefficients for the higher order terms by polynomial fitting. Several numerical examples are presented. [Slides presented at ENIEF 2011, 2011/11/1-4, Rosario, Argentina]
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4031
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1088
2011-09-03T18:07:53Z
cimec-repo:ART
"110903 2011 eng "
dc
Modelado del flujo multifase en la producción de acero por colada continua
Zanotti, Angel Luis
En la presente Tesis se desarrollan e implementan los modelos que permiten simular el flujo multifásico que tiene lugar en el proceso de fabricación de acero por el método de colada continua. Se analizan los fenómenos involucrados y las condiciones operativas que posibilitan mejorar la calidad del acero producido. En primera instancia se propone un modelo de dos fluidos para un flujo de dos fases en régimen disperso, cuya configuración consiste en burbujas de gas en un líquido (Capí- tulo 2). La estrategia numérica se encuentra fundada en una formulación estabilizada de igual orden del método de elementos finitos (SUPG-PSPG), donde el solver es dividido de acuerdo a las diferentes fases involucradas en el cálculo, tratando a cada fase monolíti- camente (cantidad de movimiento lineal y continuidad acoplada). Este esquema evita las oscilaciones debidas a la restricción de incomprensibilidad y advección que dominan el flu- jo, como así también, las patologías numéricas relacionadas con las fuerzas de interacción entre las fases. El modelo numérico se implementa en el código computacional PETSc- FEM, elaborado en el CIMEC. Se resuelven, a manera de validación, ejemplos numéricos que demuestran la aplicabilidad del código en condiciones de régimen diluido (fracción de vacío no superior al 1 %). Por otra parte, al resolver el problema bifásico en un modelo agua-aire para la con- figuración de la cuchara siderúrgica, se ha encontrado problemas de convergencia. Estas dificultades numéricas se ven considerablemente acentuadas para fracciones de vacío eleva- das (aproximadamente 100 %), altas velocidades relativas y diferencia de densidad grande entre las fases, siendo estas condiciones las que existen en las regiones cercanas al tapón poroso en la cuchara siderúrgica. Posiblemente esta sea la razón por la cual las aplicaciones del modelo multifluido se orientan generalmente a la resolución de flujos muy diluidos. Considerando que el modelo diferencial de dos fluidos para el flujo transitorio tiene raíces características complejas y por lo tanto no está bien puesto como un problema de valores iniciales, salvo para el caso trivial de igual velocidad de fases, en el Capítulo 3 se propone una novedosa estrategia matemática que evita los inconvenientes causados por el comportamiento no-hiperbólico del modelo original. La misma está basada en el precondicionamiento de la matriz de masa y permite resolver problemas en condiciones aún más exigentes a las que existen en la cuchara siderúrgica. Se desarrolló un análisis comparativo de los resultados, para problemas de solución análitica conocida (1D), con y sin matriz de precondicionamiento obteniéndose sólo para el primer caso y en todas las situaciones la solución en el estacionario. Luego de haber implementado un modelo de dos fluidos que tiene escasa robustez e importante sensibilidad de los resultados con los coeficientes de ajuste de los modelos de los términos de interacción, en el Capítulo 4 se introduce al modelo ASMM (Algebraic Slip Mixture Model) como una alternativa valiosa para la resolución del flujo multifásico en la cuchara siderúrgica. El mismo cuenta con la robustez necesaria para las condiciones de flujo presentes en el problema. En el Capítulo 5 con la finalidad de validar el código se efectúa un estudio comparativo de los resultados numéricos obtenidos con el modelo ASMM, el modelo dos fluidos (calcu- lados por Milleli) y datos experimentales relevados por Anagbo y Brimacombe. El sistema físico resuelto consiste en un modelo agua-aire para la configuracion de la cuchara siderúr- gica. Habiéndose obtenido resultados satisfactorios en esta etapa se procedió a simular la cuchara siderúrgica. Indicadores tales como el área de apertura del ojo, la velocidad críti- ca de atrape y el número de Weber se utilizaron para predecir el atrape de escoria para diferentes caudales de inyección de argón. Finalmente en el Capítulo 6 se propone un modelo híbrido ASMM-Dos Fluidos, donde la mezcla acero-argón es resuelta a través de un modelo algebraico y la interacción entre la mezcla y la escoria es tratada haciendo uso de un modelo de dos fluidos. Este novedoso tratamiento del flujo multifásico se implementa en el código comercial CFX y tiene como finalidad obtener una herramienta que permita aplicar el modelo MUSIG - Multiple Size Group Model, que se encuentra implementado en el mismo código. Este último permite realizar el cálculo con varios grupos de partículas, gotas o burbujas de distinto tamaño a un costo computacional relativamente bajo y posibilitaría cuantificar y caracterizar a las partículas de escoria atrapadas en el acero. Con el modelo híbrido se resuelven los problemas de validación y la cuchara siderúrgica en distintas condiciones de operación, para configuraciones axisimétricas y 3D, para problemas con y sin superficie libre. De esta manera se ha logrado obtener valiosas herramientas que permiten simular las condiciones tan extremas que se encuentran en una cuchara siderúrgica y que exigen estrategias computacionales robustas y de permanente desarrollo. [Ph.D dissertation Faculta de Ingeniería y Ciencias Hídricas, Universidad Nacional del Litoral]
CIMEC Document Repository
2007-02-11 23:59:33
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1088
CIMEC Document Repository; 2007
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/2959
2011-09-03T18:05:05Z
cimec-repo:ART
"110903 2011 eng "
dc
Numerical simulation of solid particle deposition in ducts
Storti, Mario Alberto
Nigro, Norberto Marcelo
Paz, Rodrigo Rafael
Dalcín, Lisandro Daniel
Solid-gas multiphase flows are present in many industrial processes as, for instance, the ventilation system of sinterization plants. One of the most important results from the practical point of view for the transport of particles in ventilation ducts is the prediction of deposition rates. Large
particles in ventilation ducts tend to fall quickly and to accumulate in a saltation layer near the lower surface of the duct. When the particle flux in the saltation layer exceeds a given threshold which depends on the shear stress, the particles begin to accumulate and a deposited layer growths. If this layer growths too much its weight may cause the collapse of the supporting structure. For small particle diameter the Eulerian/Eulerian approach is useful because the characteristic times in which the particle velocity reaches the equilibrium velocity is small. In this case a system of PDE’s is solved for each phase. When the particle diameter is very small the Algebraic Slip Model is appropriate, whereas for a moderate diameters the momentum equation of the solid phase must be solved and the Two-Fluid Model is more
appropriate. When the particles are large, and consequently the characteristic time is large, they behave almost independently of the fluid, and terms like particle inertia, virtual mass, lift, and Magnus effect are important. This effects are difficult to cast as terms in the momentum equation as PDE’s through an average process and then the Lagrangian approach where each particle trajectory is solved as an ODE is more appropriate. Moreover, the resulting system of equations may be ill posed. In addition, some
effects related with the interaction of particles with solid boundaries, like bouncing, lift-off and sliding can be treated almost uniquely in the Lagrangian formulation. In this article the implementation of an Eulerian/Lagrangian formulation for the transport of moderate to large solid particles is described.
The implementation includes an efficient tracking of particles through unstructured moving meshes, the Schiller-Naumann drag model, and partially elastic collision with solid surfaces. Deposition rates can be computed in terms of the particle flux in the saltation layer and the shear stress at the wall. Several
examples in ventilation ducts for industrial sinterization plants are presented.
CIMEC Document Repository
2011-09-03 15:07:00
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/2959
CIMEC Document Repository; 2009
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/4038
2012-07-19T16:13:31Z
cimec-repo:ART
dc
An Enrichment Scheme for Solidification Problems
Cosimo, Alejandro
Fachinotti, Victori
Cardona, Alberto
Slides presented at 10th World Congress on Computational Mechanics (WCCM 2012), 33rd Iberian Latin American Congress on Computational Methods In Engineering (CILAMCE 2012). Sao Paulo, Brazil. (July 2012)
CIMEC Document Repository
2011-09-03 15:07:07
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/4038
CIMEC Document Repository; 2011
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/461
2006-09-23T22:10:00Z
cimec-repo:ART
dc
Métodos Computacionales en Capa Límite
Storti, Mario Alberto
Nigro, Norberto M.
Dalcín, Lisandro Daniel
Paz, Rodrigo Rafael
D'Elía, Jorge
Prado, Ricardo
A pseudo-spectral numerical method for the solution of the incompressible 2D boundary layer equations is presented. A scaling is applied to the normal coordinate but with the innovation that it is based on the computed boundary layer thickness, i.e. not assuming a priori a variation for it. Spectral decay of the expansion coefficients is shown for the similar solution to wedge flows. Also, spectral convergence of the error is shown for the case of a convergent channel (one of the similar "wedge flows"), for which an analytical solution is available. The method pretends to have a good performance also even with very few parameters, so that results with four terms in the Fourier series (it amounts to two independent parameters) are compared with the well known method from von K\`arm\`an and Pohlhausen.
CIMEC Document Repository
2006-06-24 10:06:06
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/461
CIMEC Document Repository; 2006
eng
Copyright (c)
oai:ojs.www.cimec.org.ar:article/1384
2011-09-03T17:58:51Z
cimec-repo:ART
"110903 2011 eng "
dc
Simulación computacional de los procesos fluidodinámicos en el interior de motores de combustión interna
Ramajo, Damián Enrique
El trabajo desarrollado tiene como objetivo general aportar conocimientos acerca del comportamiento del flujo en el interior de motores de 4 válvulas por cilindro, haciendo especial hincapié en aquellas estructuras turbulentas que caracterizan macroscópicamente dicho flujo. El interés puesto en este trabajo se sustenta en una diversidad de estudios previos, llevados a cabo por investigadores de todo el mundo, que buscan establecer relaciones entre la turbulencia, generada durante la admisión y la compresión, y el proceso de combustión que toma lugar posteriormente. [Tesis presentada como parte de los requisitos de la Facultad de Ingeniería y Cs. Hídricas de la Universidad Nacional del Litoral para acceder al grado de Doctor en Ingeniería Mención en Mecánica Computacional. Defendida 2008-06-07]
CIMEC Document Repository
2011-09-03 15:06:50
Peer-reviewed Article
application/pdf
https://cimec.org.ar/ojs/index.php/cimec-repo/article/view/1384
CIMEC Document Repository; 2008
eng
Copyright (c)