MSTuE27
Decoupling methods for multiphysics and multiscale problems  Part I of VII
For Part II, see MSWeD27
For Part III, see MSWeE27
For Part IV, see MSThBC27
For Part V, see MSThD27
For Part VI, see MSThE27
For Part VII, see MSFrD27
Date: August 11
Time: 16:0018:00
Room: 407
(Note: Click title to show the abstract.)
Organizer:
He, Xiaoming (Missouri Univ. of Sci. & Tech.)
Xu, Xuejun (Inst. of Computational Mathematics, AMSS, CAS)
Abstract: The inherent multiphysics and multiscale features of many real world problems accentuate the importance to develop efficient and stable numerical methods for the relevant PDEs, especially the decoupling methods. Although great efforts have been made for solving these problems, many practical and analytical challenges remain to be solved. This minisymposium intends to create a forum for junior and senior researchers from different fields to discuss recent advances on the decoupling methods for multiphysics and multiscale problems with their applications.
MSTuE271
16:0016:30
On a New Robintype Nonoverlapping Domain Decomposition Preconditioner
Xu, Xuejun (Inst. of Computational Mathematics, AMSS, CAS)
Abstract: In this talk, we shall present a new Robintype nonoverlapping domain decomposition preconditioner. The unknown variables to be solved in this preconditioned algebraic system are the Robin transmission condition on the interface, which are different from the wellknown DD methods like substructuring nonoverlapping DD method and FETI method. By choosing suitable parameter on each subdomain boundary and using the tool of energy estimate, for the secondorder elliptic problem, we prove that the condition number of the preconditioned system is C(1+log(H/h ))^2, where H is the coarse mesh size and h is the fine mesh size. Numerical results shall be given to illustrate the efficiency of our DD preconditioner. This talk is based on a joint work with Yongxiang Liu.
MSTuE272
16:3017:00
Direct numerical simulation of charged particles in complex flows
Ruede, Ulrich (FA Univ. Erlangen)
Abstract: Massive parallelism enables the fully resolved simulation
of flows with large ensembles of suspended rigid particles
that are represented as individual geometric objects. Our approach uses
a Lagrangian approach based on rigid multi body dynamics and
an Eulerian description of the flow with the Lattice Boltzmann
method. Additionally we model electrostatic forces on the particles
using a finite volume discretization for the electric field.
All effects are coupled and result in a sixway interaction.
MSTuE273
17:0017:30
Efficient Multistage Preconditioners for Highly Heterogeneous Reservoir Simulations on Parallel Distributed Systems
Chen, Zhangxin (Xi'an Jiaotong Univ. & Univ. of Calgary)
Abstract: Largescale reservoir simulation has been a big challenge due to the difficulty of solving linear systems resulted from nonlinear Newton iterations. For black oil simulation, for example, more than 90% of running time is spent on the solution of linear systems. The problem is getting worse when developing parallel reservoir simulators using parallel distributed systems with tens of hundreds of CPUs. Efficient linear solvers and preconditioners are critical to the development of parallel reservoir simulators.
This presentation will address our recent work on developing parallel physicsbased preconditioners for highly heterogeneous reservoir simulations. A family of new Constrained Pressure Residual (CPR)like preconditioners and advanced matrix preprocessing methods are developed, including two new threestage preconditioners and three fourstage preconditioners. A pressure system is solved by an algebraic multigrid method, a saturation system is solved by a restricted additive Schwarz method (domain decomposition method), and the entire linear system is also solved by the restricted additive Schwarz method. To overcome a convective issue in reservoir simulation, a parallel potentialbased matrix reordering method is employed to stabilize our preconditioners. Matrix decoupling methods, such as the alternative block factorization (ABF) strategy and the quasiIMPES (implicit pressure explicit saturation) strategy, are also applied. With the restricted additive Schwarz and algebraic multigrid methods, our preconditioners have good scalability for parallel computers.
Our preconditioners have been applied to oilwater and black oil benchmark simulations. For the SPE 10 project, which is a big challenge for a linear solver because of highly heterogeneous permeability and porosity, our preconditioners with Krylov subspace solvers are stable and efficient. When using 128 CPUs, the number of iterations of our linear solvers is less than 20, and the SPE 10 project is finished in 4.5 minutes. When applying our method to a refined SPE 1 project from black oil simulation with over 80 million of grid cells, the number of iterations of our linear solvers is fewer than 3 using 1,024 CPU cores. Benchmarks with 4,096 CPU cores on IBM Blue Gene/Q are also performed and linear scalability is obtained. Our numerical experiments show that our preconditioners and linear solvers are stable with a large number of CPU cores and are efficient for highly heterogeneous simulations.
MSTuE274
17:3018:00
Strength failure models in an Eulerian context
Grove, John (Los Alamos National Laboratory)
Abstract: We will describe an anisotropic failure model for ductile material failure. The model is based on the polar decomposition of the stress tensor. When a stress component exceeds a specified failure tolerance the stress tensor is modified to remove this component in the direction of the corresponding eigenvector. We will discuss the implementation of this method in an Eulerian hydrodynamics code and thermodynamic issues associated with the model.
Return
Footnote: Code: TypeDateTimeRoom No.
Type : IL=Invited Lecture, SL=Special Lectures, MS=Minisymposia, IM=Industrial Minisymposia, CP=Contributed Papers, PP=Posters
Date: Mo=Monday, Tu=Tuesday, We=Wednesday, Th=Thursday, Fr=Friday
Time : A=8:309:30, B=10:0011:00, C=11:1012:10, BC=10:0012:10, D=13:3015:30, E=16:0018:00, F=19:0020:00, G=12:1013:30, H=15:3016:00
Room No.: TBA
