Volume 29, Issue 3
Algebraic Multigrid Block Preconditioning for Multi-Group Radiation Diffusion Equations

Xiaoqiang Yue, Shulei Zhang, Xiaowen Xu, Shi Shu & Weidong Shi

Commun. Comput. Phys., 29 (2021), pp. 831-852.

Published online: 2021-01

Preview Purchase PDF 20 3290
Export citation
  • Abstract

The paper focuses on developing and studying efficient block preconditioners based on classical algebraic multigrid (AMG) for the large-scale sparse linear systems arising from the fully coupled and implicitly cell-centered finite volume discretization of multi-group radiation diffusion equations, whose coefficient matrices can be rearranged into the $(G+2)×(G+2)$ block form, where $G$ is the number of energy groups. The preconditioning techniques are the monolithic classical AMG method, physical-variable based coarsening two-level algorithm and two types of block Schur complement preconditioners. The classical AMG method is applied to solve the subsystems which originate in the last three block preconditioners. The coupling strength and diagonal dominance are further explored to improve performance. We take advantage of representative one- and twenty-group linear systems from capsule implosion simulations to test the robustness, efficiency, strong and weak parallel scaling properties of the proposed methods. Numerical results demonstrate that block preconditioners lead to mesh- and problem-independent convergence, outperform the frequently-used AMG preconditioner and scale well both algorithmically and in parallel.

  • Keywords

Radiation diffusion equations, algebraic multigrid, block preconditioning, Schur complement, parallel computing.

  • AMS Subject Headings

65F10, 65N55, 65Z05

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CiCP-29-831, author = {Yue , Xiaoqiang and Zhang , Shulei and Xu , Xiaowen and Shu , Shi and Shi , Weidong}, title = {Algebraic Multigrid Block Preconditioning for Multi-Group Radiation Diffusion Equations}, journal = {Communications in Computational Physics}, year = {2021}, volume = {29}, number = {3}, pages = {831--852}, abstract = {

The paper focuses on developing and studying efficient block preconditioners based on classical algebraic multigrid (AMG) for the large-scale sparse linear systems arising from the fully coupled and implicitly cell-centered finite volume discretization of multi-group radiation diffusion equations, whose coefficient matrices can be rearranged into the $(G+2)×(G+2)$ block form, where $G$ is the number of energy groups. The preconditioning techniques are the monolithic classical AMG method, physical-variable based coarsening two-level algorithm and two types of block Schur complement preconditioners. The classical AMG method is applied to solve the subsystems which originate in the last three block preconditioners. The coupling strength and diagonal dominance are further explored to improve performance. We take advantage of representative one- and twenty-group linear systems from capsule implosion simulations to test the robustness, efficiency, strong and weak parallel scaling properties of the proposed methods. Numerical results demonstrate that block preconditioners lead to mesh- and problem-independent convergence, outperform the frequently-used AMG preconditioner and scale well both algorithmically and in parallel.

}, issn = {1991-7120}, doi = {https://doi.org/10.4208/cicp.OA-2020-0030}, url = {http://global-sci.org/intro/article_detail/cicp/18568.html} }
TY - JOUR T1 - Algebraic Multigrid Block Preconditioning for Multi-Group Radiation Diffusion Equations AU - Yue , Xiaoqiang AU - Zhang , Shulei AU - Xu , Xiaowen AU - Shu , Shi AU - Shi , Weidong JO - Communications in Computational Physics VL - 3 SP - 831 EP - 852 PY - 2021 DA - 2021/01 SN - 29 DO - http://doi.org/10.4208/cicp.OA-2020-0030 UR - https://global-sci.org/intro/article_detail/cicp/18568.html KW - Radiation diffusion equations, algebraic multigrid, block preconditioning, Schur complement, parallel computing. AB -

The paper focuses on developing and studying efficient block preconditioners based on classical algebraic multigrid (AMG) for the large-scale sparse linear systems arising from the fully coupled and implicitly cell-centered finite volume discretization of multi-group radiation diffusion equations, whose coefficient matrices can be rearranged into the $(G+2)×(G+2)$ block form, where $G$ is the number of energy groups. The preconditioning techniques are the monolithic classical AMG method, physical-variable based coarsening two-level algorithm and two types of block Schur complement preconditioners. The classical AMG method is applied to solve the subsystems which originate in the last three block preconditioners. The coupling strength and diagonal dominance are further explored to improve performance. We take advantage of representative one- and twenty-group linear systems from capsule implosion simulations to test the robustness, efficiency, strong and weak parallel scaling properties of the proposed methods. Numerical results demonstrate that block preconditioners lead to mesh- and problem-independent convergence, outperform the frequently-used AMG preconditioner and scale well both algorithmically and in parallel.

Xiaoqiang Yue, Shulei Zhang, Xiaowen Xu, Shi Shu & Weidong Shi. (2021). Algebraic Multigrid Block Preconditioning for Multi-Group Radiation Diffusion Equations. Communications in Computational Physics. 29 (3). 831-852. doi:10.4208/cicp.OA-2020-0030
Copy to clipboard
The citation has been copied to your clipboard