			Home About us Contact

Numerical Strategy (numerical + strategy)

Distribution by Scientific Domains

Engineering	60%

Selected Abstracts

Pipelines on heterogeneous systems: models and tools

CONCURRENCY AND COMPUTATION: PRACTICE & EXPERIENCE, Issue 9 2005
F. Almeida
Abstract We study the performance of pipeline algorithms in heterogeneous networks. The concept of heterogeneity is not only restricted to the differences in computational power of the nodes, but also refers to the network capabilities. We develop a skeleton tool that allows us an efficient block-cyclic mapping of pipelines on heterogeneous systems. The tool supports pipelines with a number of stages much larger than the number of physical processors available. We derive an analytical formula that allows us to predict the performance of pipelines in heterogeneous systems. According to the analytical complexity formula, numerical strategies to solve the optimal mapping problem are proposed. The computational results prove the accuracy of the predictions and effectiveness of the approach. Copyright © 2005 John Wiley & Sons, Ltd. [source]

Efficient numerical strategies for spectral stochastic finite element models

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, Issue 10 2005
Doo Bo Chung
Abstract The use of spectral stochastic finite element models results in large systems of equations requiring specialized solution strategies. This paper discusses three different numerical algorithms for solving these large systems of equations. It presents a trade-off of these algorithms in terms of memory usage and computation time. It also shows that the structure of the spectral stochastic stiffness matrix can be exploited to accelerate the solution process, while keeping the memory usage to a minimum. Copyright © 2005 John Wiley & Sons, Ltd. [source]

A robust methodology for RANS simulations of highly underexpanded jets

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS, Issue 12 2008
G. Lehnasch
Abstract This work aims at developing/combining numerical tools adapted to the simulation of the near field of highly underexpanded jets. An overview of the challenging numerical problems related to the complex shock/expansion structure encountered in these flows is given and an efficient and low-cost numerical strategy is proposed to overcome these, even on short computational domains. Based on common upwinding algorithms used on unstructured meshes in a mixed finite-volume/finite-element approach, it relies on an appropriate utilization of zonal anisotropic remeshing algorithms. This methodology is validated for the whole near field of cold air jets issuing from axisymmetric convergent nozzles and yielding various underexpansion ratios. In addition, the most usual corrections of the k,, model used to take into account the compressibility effects on turbulence are precisely assessed. Copyright © 2007 John Wiley & Sons, Ltd. [source]

A finite element strategy for the solution of interface tracking problems

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS, Issue 12 2005
C. Devals
Abstract A finite element-based numerical strategy for interface tracking is developed for the simulation of two-phase flows. The method is based on the solution of an advection equation for the so-called ,pseudo-concentration' of one of the phases. To obtain an accurate description of the interface, a streamline upwind Petrov,Galerkin (SUPG) scheme is combined with an automatic mesh refinement procedure and a filtering technique, making it possible to generate an oscillation-free pseudo-concentration field. The performance of the proposed approach is successfully tested on four classical two-dimensional benchmark problems: the advection skew to the mesh, the transport of a square shape in a constant velocity flow field, the transport of a cut-out cylinder in a rotating flow field and the transport of a disc in a shear flow. Copyright © 2005 John Wiley & Sons, Ltd. [source]

Transport-equilibrium schemes for computing nonclassical shocks.

NUMERICAL METHODS FOR PARTIAL DIFFERENTIAL EQUATIONS, Issue 4 2008
Scalar conservation laws
Abstract This paper presents a new numerical strategy for computing the nonclassical weak solutions of scalar conservation laws which fail to be genuinely nonlinear. We concentrate on the typical situation of concave,convex and convex,concave flux functions. In such situations the so-called nonclassical shocks, violating the classical Oleinik entropy criterion and selected by a prescribed kinetic relation, naturally arise in the resolution of the Riemann problem. Enforcing the kinetic relation from a numerical point of view is known to be a crucial but challenging issue. By means of an algorithm made of two steps, namely an Equilibrium step and a Transport step, we show how to force the validity of the kinetic relation at the discrete level. The proposed strategy is based on the use of a numerical flux function and random numbers. We prove that the resulting scheme enjoys important consistency properties. Numerous numerical evidences illustrate the validity of our approach. © 2007 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2008 [source]