×

Efficient solution of mechanical and biomechanical problems by domain decomposition. (English) Zbl 1010.74069

This paper shows the ability of domain decomposition method with a Neumann-Neumann preconditioner to solve industrial and biomechanical problems. After introducing the main features of decomposition and preconditioner technique, the authors present a large scale engineering problem stemming from an anisotropic non-homogeneous three-dimensional slender structure. Next, a three-dimensional model of human liver is in the focus of attention. First, a linear elasticity model is considered and solved by Neumann-Neumann approach, by a direct method, and by the preconditioned conjugate gradient method with different preconditioners. Then a nonlinear hyperelastic model is treated by the Neumann-Neumann domain decomposition algorithm. In all instances, the proposed method appears to be more reliable than standard iterative methods and more efficient than direct methods. The algorithm has been implemented on different parallel computers in the framework of the general purpose finite element library Modulef.

MSC:

74S05 Finite element methods applied to problems in solid mechanics
74L15 Biomechanical solid mechanics
65N55 Multigrid methods; domain decomposition for boundary value problems involving PDEs
65Y05 Parallel numerical computation
74B05 Classical linear elasticity
74B20 Nonlinear elasticity

Software:

Modulef
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Amestoy, Calculateurs Paralléles Réseaux et Systèmes Répartis 10 pp 509– (1998)
[2] and . Direct solvers in the PARASOL project. In Greenwich Conference, 1998.
[3] Modèles anatomiques déformables en temps-réel: application à la simulation de chirurgie avec retour d’effort. Docteur en informatique, Université de Nice Sophia Antipolis, INRIA Sophia Antipolis, 19 Novembre 1997.
[4] Cotin, Lecture Notes in Computer Science 1131 pp 535– (1996) · doi:10.1007/BFb0046995
[5] Delingette, Proceedings of the IEEE 86 pp 512– (1998) · doi:10.1109/5.662876
[6] and . Towards a unified theory of domain decomposition algorithm for elliptic problems. In editor, Proceedings of the Third International Symposium on Domain Decomposition Methods for Partial Differential Equations, Houston, Texas, March 1989. SIAM, Philadelphia, 1990.
[7] and . Additive Schwarz methods for elliptic finite element problems in three-dimensions. In , and , editors, Proceedings of the Fifth International Symposium on Domain Decomposition Methods for Partial Differential Equations, Norfolk, May 1991. SIAM, Philadelphia, 1992.
[8] Frey, Comput. Visual Sci. 1 pp 113– (1998) · Zbl 0911.68202 · doi:10.1007/s007910050011
[9] George, Comput. Methods Appl. Mech. Eng. 92 pp 269– (1991) · Zbl 0756.65133 · doi:10.1016/0045-7825(91)90017-Z
[10] Numerical Methods for Nonlinear Three-dimensional Elasticity, and , editors, volume 3, pages 465-622, 1994. · Zbl 0875.73234
[11] Le Tallec, SIAM J. Numer. Anal. 35 pp 836– (1998) · Zbl 0917.73071 · doi:10.1137/S0036142995291019
[12] and . Méthodes de décomposition de domaines en calcul des structures. In Colloque National en Calcul des Structures, pages 33-49, Giens, May 1993. Csma, Gamni, Inria, Hermès.
[13] and . Solving large scale structural problems on parallel computers using domain decomposition. In Solving Large Scale Problems in Mechanics, Chapter 2, pages 49-82. John Wiley & Sons, 1997.
[14] and . Generalized Neumann-Neumann preconditioners for iterative substructuring. In and , editors, Proceedings of the Ninth International Symposium on Domain Decomposition Methods for Partial Differential Equations, Bergen, June 1996. 1998.
[15] Mandel, Commun. Numer. methods eng. 9 pp 233– (1993) · Zbl 0796.65126 · doi:10.1002/cnm.1640090307
[16] Przemieniecki, Am. Inst. Aero. Astro. J. 1 pp 138– (1963) · doi:10.2514/3.1483
[17] Incomplete gaussian elimination as a preconditioning for generalized conjugate gradient acceleration. In Proceedings of the Seventh Symposium on Numerical Simulation of Reservoir Performance of the Society of Petroleum Engineers, 1983.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.