Comparisons of Partitioned Solution Procedures for Transient Coupled Problems in Sequential and Parallel Processing

Matteazzi, Renato; Schrefler, Bernhard A.; Vitaliani, Renato

doi:10.4203/ccp.38.8.2

Cited by 7 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The accuracy criterion was ( t 2 ) min = 0.075 sm −2 > 0.048 sm −2 , which verifies the previous condition. Our objective is to compare the LATIN approach with the Iterative Standard Parallel Procedure (ISPP [63]), which is one of the standard partitioning schemes. In order to compare the results in terms of computation cost, let us introduce the following notations: as before, n I is the number of time steps in the interval I being studied; n S and n F are the numbers of global uncoupled resolutions (i.e.…”

Section: Illustration Of the Latin Multiscale Strategymentioning

confidence: 99%

Proper Generalized Decomposition for Multiscale and Multiphysics Problems

Néron

Ladevèze

2010

Arch Computat Methods Eng

View full text Add to dashboard Cite

This paper is a review of the developments of the Proper Generalized Decomposition (PGD) method for the resolution, using the multiscale/multiphysics LATIN method, of the nonlinear, time-dependent problems ((visco)plasticity, damage, . . . ) encountered in computational mechanics. PGD leads to considerable savings in terms of computing time and storage, and makes engineering problems which would otherwise be completely out of range of industrial codes accessible.

show abstract

Section: Illustration Of the Latin Multiscale Strategymentioning

confidence: 99%

Proper Generalized Decomposition for Multiscale and Multiphysics Problems

Néron

Ladevèze

2010

Arch Computat Methods Eng

View full text Add to dashboard Cite

show abstract

“…In this paper, we compare the LATIN approach to the Iterated Standard Staggered Procedure (ISPP [MSV96]), which is one of these standard partitioning schemes. This comparison is made on 1D and 2D problems and is an extension of the first comparison published earlier along with the principles of the proposed multiphysics strategy in [DLS03].…”

Section: Introductionmentioning

confidence: 99%

A Multi-Time-Scale Strategy for Multiphysics Problems: Application to Poroelasticity

Dureisseix

Ladevèze

Néron

et al. 2003

Int J Mult Comp Eng

View full text Add to dashboard Cite

“…It is possible to stabilize several schemes with dedicated techniques, such as in References [21,22], or by condensing one type of unknown onto the others; (for further description see, for instance, References [19,23]). These latter techniques su er from high computational cost because of the condensation step which destroys the sparsity of the operators.…”

Section: Iterated Standard Parallel Proceduresmentioning

confidence: 99%

“…In order to compare the proposed approach with the Iterated Standard Parallel Procedure (ISPP) described in Reference [19], a reference solution was required. To produce that reference solution, we used the so-called direct monolithic scheme, which is brie y described below.…”

Section: Problem Description and Discretizationmentioning

confidence: 99%

“…E ciency of the representations: The space ÿelds are determined by solving a small number of global systems (19), (22) (within the ÿxed-point subiterations), while the time functions are calculated by integrating a small number of di erential equations (moreover, in the linear case treated here, these equations are explicit (18), (21)). If the convergence rate is similar to that of the case without representation, signiÿcant performance improvement can be expected.…”

Section: Solid Partmentioning

confidence: 99%

See 1 more Smart Citation

A LATIN computational strategy for multiphysics problems: application to poroelasticity

Dureisseix

Ladevèze

Schrefler

2003

Numerical Meth Engineering

View full text Add to dashboard Cite

Multiphysics phenomena and coupled-eld problems usually lead to analyses which are computationally intensive. Strategies to keep the cost of these problems aordable are of special interest. For coupled uid–structure problems, for instance, partitioned procedures and staggered algorithms are often preferred to direct analysis. In this paper, we describe a new strategy for solving coupled multiphysics problems which is built upon the LArge Time INcrement (LATIN) method. The proposed application concerns the consolidation of saturated porous soil, which is a strongly coupled uid–solid problem. The goal of this paper is to discuss the eciency of the proposed approach, especially when using an appropriate time-space approximation of the unknowns for the iterative resolution of the uncoupled global problem. The use of a set of radial loads as an adaptive approximation of the solution during iterations will be validated and a strategy for limiting the number of global resolutions will be tested on multiphysics problems

show abstract

Comparisons of Partitioned Solution Procedures for Transient Coupled Problems in Sequential and Parallel Processing

Cited by 7 publications

References 0 publications

Proper Generalized Decomposition for Multiscale and Multiphysics Problems

Proper Generalized Decomposition for Multiscale and Multiphysics Problems

A Multi-Time-Scale Strategy for Multiphysics Problems: Application to Poroelasticity

A LATIN computational strategy for multiphysics problems: application to poroelasticity

Contact Info

Product

Resources

About