Solving large scale linear systems efficiently plays an important role in a petroleum reservoir simulator, and the key part is how to choose an effective parallel preconditioner. Properly choosing a good preconditioner has been beyond the pure algebraic field. An integrated preconditioner should include such components as physical background, characteristics of PDE mathematical model, nonlinear solving method, linear solving algorithm, domain decomposition and parallel computation. We first discuss some parallel preconditioning techniques, and then construct an integrated preconditioner, which is based on large scale distributed parallel processing, and reservoir simulation-oriented. The infrastructure of this preconditioner contains such famous preconditioning construction techniques as coarse grid correction, constraint residual correction and subspace projection correction. We essentially use multi-step means to integrate totally eight types of preconditioning components in order to give out the final preconditioner. Million-grid cell scale industrial reservoir data were tested on native high performance computers. Numerical statistics and analyses show that this preconditioner achieves satisfying parallel efficiency and acceleration effect.
[1]
H. V. D. Vorst,et al.
The effect of incomplete decomposition preconditioning on the convergence of conjugate gradients
,
1992
.
[2]
Jinchao Xu,et al.
Iterative Methods by Space Decomposition and Subspace Correction
,
1992,
SIAM Rev..
[3]
Wei Liu,et al.
Parallel Reservoir Simulation on Shared and Distributed Memory System
,
2000
.
[4]
Peng Wang,et al.
Parallel Compositional Reservoir Simulation on Clusters of PCs
,
2001,
Int. J. High Perform. Comput. Appl..
[5]
Cao Jian.
THE IMPLEMENTATION OF A PARALLEL SOFTWARE OF PETROLEUM RESERVOIR SIMULATION AND ITS APPLICATION ON NATIVE HIGH PERFORMANCE COMPUTERS
,
2002
.
[6]
Jia-changSun.
MULTIVARIATE FOURIER SERIES OVER A CLASS OF NON TENSOR-PRODUCT PARTITION DOMAINS
,
2003
.
[7]
Yousef Saad,et al.
Iterative methods for sparse linear systems
,
2003
.