Begell House
International Journal for Multiscale Computational Engineering
International Journal for Multiscale Computational Engineering
1543-1649
4
2
2006
Preface
Valeria
Krzhizhanovskaya
University of Amsterdam, The Netherlands
Bastien
Chopard
Computer Science Department, University of Geneva, 1211 Geneva 4, Switzerland
Yuriy E.
Gorbachev
Institute of High-Performance Computations and Databases of the St. Petersburg State Polytechnic University, Russian Federation
207
A Hybrid Lattice Boltzmann Finite Difference Scheme for the Diffusion Equation
We show how a lattice Boltzmann (LB) can be spatially coupled with a finite difference (FD) scheme, each method running on a separate region, to solve a given problem. The typical situation we consider is a computational domain that is partitioned into two regions. The same spatiotemporal physical process extends over the full domain, but a different numerical method is used over each region. At the interface of the subdomains, the LB and FD must be connected so as to ensure a perfect continuity of the physical quantities. We derive the theoretical concepts, which allow us to link both methods in the case of a diffusion process, and validate them with numerical simulations on a two-dimensional domain. We also consider the case of different size grids for which the coupling has to be complemented with an interpolation procedure.
Davide
Alemani
CABE, University of Geneva, 1211 Geneva 4, Switzerland
Bastien
Chopard
Computer Science Department, University of Geneva, 1211 Geneva 4, Switzerland
Paul
Albuquerque
Computer Science Department, University of Geneva, 1211 Geneva 4, Switzerland; and LII, Ecole d'Ingénieurs de Genéve, HES-SO, 1202 Geneva, Switzerland
Pierre
Leone
Computer Science Department, University of Geneva, 1211 Geneva 4, Switzerland
209-219
Soot Particle Deposition within Porous Structures using a Method-of-Moments-Lattice-Boltzmann Approach
This paper deals with the combination of two computational methods to simulate the flow of particle-laden fluids through porous structures: the lattice Boltzmann method (LBM), which is a method to solve the Navier-Stokes equation in complex geometries and the method of moments (MoM), which describes the time evolution of nonhomogeneous particle distributions. The combination of these methods makes it possible to take phenomena into account that depend on particle size and size distribution of the transported material. It is also possible to simulate changes in the size distribution. The applicability of the method is demonstrated by simulating the deposition of diesel soot on the porous structure of a filter. The geometry of the filter material has been reconstructed from CT scans of the filter material.
Bernhard F.W.
Gschaider
Christian-Doppler-Laboratory for Applied Computational Thermofluiddynamics, Mining University Leoben, Franz-Josef-Strasse 18, 8700 Leoben, Austria
Claudia C.
Honeger
Christian-Doppler-Laboratory for Applied Computational Thermofluiddynamics, Mining University Leoben, Franz-Josef-Strasse 18, 8700 Leoben, Austria
Christian E. P.
Redl
Christian-Doppler-Laboratory for Applied Computational Thermofluiddynamics, Mining University Leoben, Franz-Josef-Strasse 18, 8700 Leoben, Austria
Johannes
Leixnering
Christian-Doppler-Laboratory for Applied Computational Thermofluiddynamics, Mining University Leoben, Franz-Josef-Strasse 18, 8700 Leoben, Austria
221-232
Self-consistent Particle Model of Discharge Plasmas in Hydrogen
We describe a mixed particle/continuum model for discharge plasmas in molecular gases developed by our group, which couples a particle description of the plasma phase with the diffusion/reaction kinetics of atoms and molecules in the gas phase. In particular, we focus on hydrogen plasma, in view of the interesting negative ion kinetics. The model includes an improved treatment of ion kinetics, which incorporates original solutions for specific problems of multiple time scale kinetics. Results obtained for a parallel plate, radio frequency discharge considered as a test case are discussed, also in comparison with fluid code results.
Savino
Longo
Dipartimento di Chimica dell'Universitá di Bari, and IMIP/CNR, Via Orabona 4, 70126 Bari, Italy
Mario
Capitelli
Dipartimento di Chimica, Universita di Bari, via Orabona n. 4, 70126 Bari, Italy; IMIP-CNR, sect Bari, via Orabona n. 4, 70126 Ban, Italy
P.
Diomede
Dipartimento di Chimica dell'Universitá di Bari, Via Orabona 4, 70126 Bari, Italy
233-242
Multiscale Simulation of Hall Discharge
Non-Maxwellian behavior and plasma-wall interaction are key processes in the physics of Hall thrusters. For this purpose, a 2D{r,z}-3V axisymmetric fully kinetic particle-in-cell/Monte Carlo collision (PIC-MCC) model of the acceleration channel including the process of secondary electron emission (SEE) from the dielectric walls has been developed. In order to make the simulation possible with regard to the computational time due to the different electron and neutral time scale, a reduction of the thruster dimension was done. This was derived from a new physics-based scaling law. This model has demonstrated its outstanding capability in improving the physics insight into the processes in a stationary plasma thruster (SPT) and in reproducing experimental data accurately.
Francesco
Taccogna
Dipartimento di Chimica dell'Universitá di Bari, via Orabona 4, 70126 Bari, Italy
Savino
Longo
Dipartimento di Chimica dell'Universitá di Bari, and IMIP/CNR, Via Orabona 4, 70126 Bari, Italy
Mario
Capitelli
Dipartimento di Chimica, Universita di Bari, via Orabona n. 4, 70126 Bari, Italy; IMIP-CNR, sect Bari, via Orabona n. 4, 70126 Ban, Italy
Ralf
Schneider
Max Planck Institute für Plasmaphysik, Wendelsteinstr. 1, D-17491 Greifswald, Germany
243-254
Implicit and Explicit Higher-Order Time Integration Schemes for Fluid-Structure Interaction Computations
In this paper, higher-order time integration schemes are applied to fluid-structure interaction (FSI) simulations. For a given accuracy, we investigate the efficiency of higher-order time integration schemes compared to lower-order methods. In the partitioned FSI simulations on a one-dimensional piston problem, a mixed implicit/explicit (IMEX) time integration scheme is employed: the implicit scheme is used to integrate the fluid and structural dynamics, whereas an explicit Runge-Kutta scheme integrates the coupling terms. The resulting IMEX scheme retains the order of the implicit and explicit schemes. In the IMEX scheme considered, the implicit scheme consists of an explicit first stage, singly diagonally implicit Runge-Kutta scheme, which is a multistage, L-stable scheme.
Alexander
van Zuijlen
Delft University of Technology, Faculty of Aerospace Engineering, P.O. Box 5058,2600GB, The Netherlands
Hester
Bijl
TU Delft, Kluyverweg 1 (10.18), 2629 HS Delft, The Netherlands
255-263
Modeling Ionic Continua Under Multifield Conditions
Recent advances in the development of electroactive polymer materials and composites along with the need for new multifunctional exploitation of these materials have underlined the need for the multi-field modeling of their behavior. Behavioral modeling of these materials is essential for design, material qualification, and material certification for sensing, actuation, and energy harvesting applications. The present paper proposes and applies a methodology for modeling the behavior of ionic continua under multi-field influence at the macro length scale. The computational implementation of this methodology addresses generation and solution of both the constitutive and the field evolution equations by appropriate use of continuum mechanics, irreversible thermodynamics, and electrodynamics. An application of this methodology for the case of electric multi-component anisotropic hygrothermoelasticity generates a constitutive model for a large class of materials capable of actuation, sensing, and energy harvesting applications. A specialization of this theory for isotropic and bi-component chemo-thermo-electro-elastic materials is provided along with the corresponding field equations. To demonstrate the capabilities of this approach for realistic applications, a system of nonlinear governing partial differential equations is derived to describe the state evolution of large deflection plates made from such material systems. These equations represent the electro-hygro-thermal generalization of the well-known von-Karman equations for large deflection plates and capture the actuating behavior of these plates. Finally, numerical solutions of these equations for two sets of boundary conditions are presented to demonstrate solution feasibility and realism of modeling in the context of actuation-based applications.
John
Michopoulos
Computational Multiphysics Systems Laboratory, Code 6394, Center for Computational Material Science, Naval Research Laboratory, USA
265-279
Formation of Dwarf Galaxies in Reionized Universe with Heterogeneous Multicomputer System
HMCS (heterogeneous multicomputer system) is a very powerful and ideal computational environment for large-scale computational astrophysics simulations including multiple physical phenomena, such as radiation transfer, radiative cooling, chemical reaction network, hydrodynamics, and gravity. In this system, general purpose and special purpose parallel processing systems are involved to realize very high-performance computation. We have constructed a system with MPP and PC-cluster as general purpose side and GRAPE-6 gravity engine as special purpose side. We perform 3D radiation smoothed-particle-hydrodynamics simulations on the formation and the photoevaporation of sub galactic objects (M ∼ 108−109 MΘ). We confirm the suppression of the formation of small galaxies after the reionization. We also find that the galaxies that undergo violent photoevaporation process still retain certain amount of stars, which are formed at small scale high-density peaks. These cooled components merge with each other when the dark matter halo of the whole system is formed. It is also found these low mass galaxies should have large mass-to-light ratios, and these systems could be the progenitor of dwarf spheroidal galaxies in local group.
Taisuke
Boku
Institute of Information Sciences and Electronics, and Center for Computational Physics, University of Tsukuba, Tennodai, Tsukuba-shi, Ibaraki, Japan
Hajime
Susa
Department of Physics, Faculty of Science, Rikkyo University
Kenji
Onuma
Doctoral Program of Systems and Information Engineering, Graduate School, University of Tsukuba, Tennodai, Tsukuba-shi, Ibaraki, Japan
Masayuki
Umemura
Institute of Physics, Center for Computational Physics, University of Tsukuba, Tennodai, Tsukuba-shi, Ibaraki, Japan
Mitsuhisa
Sato
Institute of Information Sciences and Electronics, and Center for Computational Physics, University of Tsukuba, Tennodai, Tsukuba-shi, Ibaraki, Japan
Daisuke
Takahashi
Institute of Information Sciences and Electronics, and Center for Computational Physics, University of Tsukuba, Tennodai, Tsukuba-shi, Ibaraki, Japan
281-289
Two-Dimensional Analysis of Shape Memory Alloys under Small Loadings
A mathematical model is constructed for modeling the behavior of shape memory alloy (SMA) patches in two dimensions. The effect of phase transformations is included into the model via a specific choice of the free energy functional valid for square-to-rectangular transformations. It is shown that the classical one-dimensional Falk dynamic model, applied to shape memory alloy rods, could be regarded as a special case of the formulated two-dimensional model. The model is analyzed numerically with the method of lines and a series of computational experiments for SMA materials is discussed, in detail, in order to compare the results of simulations to those obtained with conventional one-dimensional models.
L. X.
Wang
MCI, Faculty of Science and Engineering, University of Southern Denmark, Sonderborg, DK-6400, Denmark
Roderick V. N.
Melnik
Center for Coupled Dynamics & Complex Systems, Wilfrid Laurier University, 75 University Avenue West, Waterloo, ON, N2L 3C5, Canada
291-304