Robust preconditioners for incompressible MHD models

Abstract

ABSTRACT. In this paper, we develop two classes of robust preconditioners for the structurepreserving discretization of the incompressible magnetohydrodynamics (MHD) system. By studying the well-posedness of the discrete system, we design block preconditioners for them and carry out rigorous analysis on their performance. We prove that such preconditioners are robust with respect to most physical and discretization parameters. In our proof, we improve the existing estimates of the block triangular preconditioners for saddle point problems by removing the scaling parameters, which are usually difficult to choose in practice. This new technique is not only applicable to the MHD system, but also to other problems. Moreover, we prove that Krylov iterative methods with our preconditioners preserve the divergence-free condition exactly, which complements the structure-preserving discretization. Another feature is that we can directly generalize this technique to other discretizations of the MHD system. We also present preliminary numerical results to support the theoretical results and demonstrate the robustness of the proposed preconditioners.

A robust solver for the finite element approximation of stationary incompressible MHD equations in 3D

“…e.g. [13,14,16,18,21,24,[26][27][28]31] and the references therein). In [18], Gunzburger et al studied well-posedness and the finite element method for the stationary incompressible MHD equations.…”

A face‐based monolithic approach for the incompressible magnetohydrodynamics equations

“…In recent years, the motion of an electrically conducting fluid within a three‐dimensional lid‐driven cavity has been extensively investigated in the literature . In here, the calculations are carried out for a lid‐driven cubic cavity with a horizontally applied magnetic field, B =(1,0,0) ⊤ corresponding to the work of Li and Zheng due to its relatively high Hartmann number.…”

“…It may be noted that the ratio between the off‐diagonal entries in the B 12 and B 21 blocks is proportional to and its large values adversely effect the system condition number. In the literature, the several block preconditioning techniques are proposed . In here, to remove the zero block in the original system, an upper triangular right preconditioner is used as follows: which leads to …”

A coupling of hybrid mixed and continuous Galerkin finite element methods for poroelasticity

“…Next theorem shows that (30) and A E are FOV-equivalent and, therefore, B E L provides a preconditoner for general minimal residual (GMRES) method as suggested in [32,40,41,42] Theorem 5.3. Assuming a shape regular mesh and the discretization described above, there exists constants Σ L and Υ L , independent of discretization and physical parameters, such that, for any x = (u, p, β) ⊤ = 0,…”

“…In this section, we use the well-posedness to develop block preconditioners for the linear system A E (20). Following the general framework developed in [32,33], we first consider block diagonal preconditioners (norm-equivalent preconditioners), then we discuss block triangular preconditioners following the framework developed in [32,40,41,42] for Field-of-Value (FOV) equivalent preconditioners. We theoretically show that their performance is robust with respect to the discretization and physical parameters.…”