K. Choukairy¹, R. Bennacer²

FDMP-Fluid Dynamics & Materials Processing, Vol.8, No.2, pp. 155-172, 2012, DOI:10.3970/fdmp.2012.008.155

Abstract This study carries the natural thermosolutal convection induced in heterogeneous porous media. The configuration considered is cartesian. The horizontal and vertical walls are submitted to different mass and heat transfer. The equations which govern this type of flow are solved numerically by using the finite volume method. The flow is considered two-dimensional and laminar. The model of Darcy and the approximation of the Boussinesq are taken into account. The parameters which control the problem are the thermal Darcy-Rayleigh number, Rt, the buoyancy ratio, N, the Lewis number, Le, the aspect ratio of the enclosure, A… More >

Ji Wang, Yangyang Chen, Guijia Chen, Jianke Du

The International Conference on Computational & Experimental Engineering and Sciences, Vol.17, No.3, pp. 89-90, 2011, DOI:10.3970/icces.2011.017.089

Abstract The finite element analysis of quartz crystal resonators is increasingly important due to the precise design requirements in frequency and electrical properties with the consideration of crystal blank, processing, mounting, and packaging. To reduce the computational cost, one proven approach is to use the Mindlin plate theory for the thickness-shear vibrations of crystal plates with electrodes and other complications. This approach has been implemented in parallel finite element method with the sophisticated software components for the solutions of linear systems in terms such as eigenvalues, mode shapes, and amplitudes, which in turn can be used… More >

G.A. Gravvanis¹, K.M. Giannoutakis²

CMES-Computer Modeling in Engineering & Sciences, Vol.71, No.4, pp. 305-330, 2011, DOI:10.3970/cmes.2011.071.305

Abstract In this paper we present parallel explicit approximate inverse matrix techniques for solving sparse linear systems on shared memory systems, which are derived using the finite element method for biharmonic equations in three space variables. Our approach for solving such equations is by considering the biharmonic equation as a coupled equation approach (pair of Poisson equation), using a FE approximation scheme, yielding an inner-outer iteration method. Additionally, parallel approximate inverse matrix algorithms are introduced for the efficient solution of sparse linear systems, based on an anti-diagonal computational approach that eliminates the data dependencies. Parallel explicit More >

Wenqing Wang¹, Thomas Schnicke², Olaf Kolditz³

CMC-Computers, Materials & Continua, Vol.21, No.3, pp. 217-236, 2011, DOI:10.3970/cmc.2011.021.217

Abstract This work focuses on parallel finite element simulation of thermal hydraulic and mechanical (THM) coupled processes in porous media, which is a common phenomenon in geological applications such as nuclear waste repository and CO2 storage facilities. The Galerkin finite element method is applied to solve the derived partial differential equations. To deal with the coupling terms among the equations, the momentum equation is solved individually in a monolithic manner, and moreover their solving processes are incorporated into the solving processes of nonisothermal hydraulic equation and heat transport equation in a staggered manner. The computation task… More >

S.H. Ju¹

CMES-Computer Modeling in Engineering & Sciences, Vol.64, No.1, pp. 91-109, 2010, DOI:10.3970/cmes.2010.064.091

Abstract This study develops an OpenMP scheme to parallel the preconditioned conjugate gradient methods (PCG) in shared memory computers. The proposed method is simple and systematic, so a minor change in traditional PCG methods may produce effective parallelism. At first, the global stiffness matrix is re-numbered in order to produce a parallel three-line form matrix, and a subroutine only needs to be called once in the finite element analysis. Several basic OpenMP commands are then added into the traditional incomplete Cholesky factorization (ILU) and symmetric successive over-relaxation (SSOR) codes to make the procedures of matrix multiplication, More >

Yantao Zhang¹, Xiong Zhang^1,2, Yan Liu¹

CMES-Computer Modeling in Engineering & Sciences, Vol.69, No.2, pp. 143-166, 2010, DOI:10.3970/cmes.2010.069.143

Abstract Material point method(MPM) is a promising method in solving problems involving large deformations, especially explosion and penetration. In MPM, particles can move around the computing domain dynamically, which can result in load imbalance easily. In parallelizing MPM using OpenMP, data races will occur in the stage of grid node updating if we use loop-level parallelism for these loops. Huang et al. proposed a domain decomposition method to overcome data races [Huang, Zhang, Ma and Wang (2008)]. However, significant modifications of the original serial code are required. In this paper, we proposed a new alternated grid More >

Jong Keun Moon¹, Seung Jo Kim²

CMES-Computer Modeling in Engineering & Sciences, Vol.67, No.3, pp. 239-264, 2010, DOI:10.3970/cmes.2010.067.239

Abstract Large structural problems with high precision and complexity require a high-performance computation using the efficient parallel algorithm. The purpose of this paper is to present the parallel performance of thermal-structural coupled analysis tested on a parallel cluster system. In the coupled analysis, the heat transfer analysis is carried out, and then the structural analysis is performed based on temperature distribution. For the automatic and efficient connection of two parallel analysis modules, the several communication patterns were studied. The parallel performance was demonstrated for the sample and the real application problems, such as a laminated composite More >

Ji Joong Moon¹, Seung Jo Kim¹, Minhyung Lee²

CMES-Computer Modeling in Engineering & Sciences, Vol.59, No.3, pp. 275-300, 2010, DOI:10.3970/cmes.2010.059.275

Abstract This paper describes the parallelization of contact/impact simulation for fracture modeling of brittle materials using a node separation scheme (NSS). We successfully demonstrated the fracture modeling of brittle materials using a cohesive fracture model. Since a NSS continuously generates new free surfaces as the computation progresses, the methodology requires increased computational time. To perform a simulation within a reasonable time period, a parallelization study is conducted. Particular methods for effective parallelization, especially for brittle materials, are described in detail. The crucial and most difficult strategy is the management of the data structure and communication needed More >

O. Allix¹, P. Kerfriden¹, P. Gosselet¹

CMES-Computer Modeling in Engineering & Sciences, Vol.55, No.3, pp. 271-292, 2010, DOI:10.3970/cmes.2010.055.271

Abstract We present numerical enhancements of a multiscale domain decomposition strategy based on a LaTIn solver and dedicated to the computation of the debounding in laminated composites. We show that the classical scale separation is irrelevant in the process zones, which results in a drop in the convergence rate of the strategy. We show that performing nonlinear subresolutions in the vicinity of the front of the crack at each prediction stage of the iterative solver permits to restore the effectiveness of the method. More >

Ying Liu¹, Yufeng Nie², Weiwei Zhang², Lei Wang²

CMES-Computer Modeling in Engineering & Sciences, Vol.55, No.1, pp. 89-110, 2010, DOI:10.3970/cmes.2010.055.089

Abstract In the light of the ideas and treatment technologies about molecular dynamics simulation and bubble meshing, a new approach of node placement for the meshless method called node placement method by bubble simulation (NPBS method), is proposed. Nodes are seen as the centers of the bubbles which can be moved by their interacting forces. Through dynamic simulation, bubbles are placed into a near-optimal configuration, and the centers of bubbles will form a good-quality node distribution in the domain. This process doesn't need updating the mesh connection constantly, i.e., is totally meshfree. Some example results show… More >