EI、Scopus 收录
中文核心期刊

一种不可压缩理想磁流体的解耦保结构算法

A DECOUPLED STRUCTURE-PRESERVING SCHEME FOR THE INCOMPRESSIBLE IDEAL MHD SYSTEM

  • 摘要: 不可压缩理想磁流体(MHD)方程组描述了流体速度场、磁场以及电场之间的紧密耦合, 是研究高温等离子体、天体物理以及核聚变等复杂物理过程的基本模型. 然而, 由于其多物理场强耦合及非线性特性, 求解该系统在数值方法设计上面临巨大挑战. 特别是传统数值方法在离散层面难以严格保持系统所固有的本征性质如质量守恒、电荷守恒等, 进而影响模拟的长期稳定性与物理可信度. 为此, 本文提出了一种高阶的、保结构的不可压缩理想 MHD混合有限元离散格式. 该方法引入涡量演化方程, 构建了速度-涡量-磁场演化形式的控制方程; 在时间离散上采用隐式中点法, 并构造时间交错网格, 同时对耦合项进行了适当的显-隐处理, 实现了整个系统的完全解耦, 避免了由非线性项导致的迭代求解过程. 在三维收敛性测试中, 该方法展现出最优阶收敛阶; 在经典二维Orszag-Tang涡旋问题中, 速度场的无散性可精确控制在机器精度(仅含舍入误差), 验证了该格式在保持物理结构和捕捉复杂演化过程中的稳定性与准确性.

     

    Abstract: Numerical simulations of the incompressible ideal magnetohydrodynamics (MHD) equations involve the coupling of multiple physical fields, including the fluid velocity, magnetic field, and electric field. Due to the strong nonlinearity and multi-physics interactions inherent in the system, designing robust and accurate numerical methods is a significant challenge. Traditional numerical approaches often struggle to preserve physical conservation properties, such as mass conservation and charge conservation at the discrete level, which can compromise the long-term stability and physical fidelity of the simulations. In this work, we propose a high-order, strictly mass-conserving mixed finite element discretization for the incompressible ideal MHD equations. Our method introduces a velocity-vorticity-magnetic formulation by incorporating the evolution equation for vorticity. For temporal discretization, we adopt the implicit midpoint method and construct a temporally staggered grid. Furthermore, we apply an explicit-implicit splitting strategy to the coupling terms, which allows the entire system to be fully decoupled and avoid the need for iterative solvers typically required by nonlinear systems. The proposed scheme demonstrates optimal spatial convergence rates in three-dimensional convergence test. In the classical two-dimensional Orszag-Tang vortex simulation, the divergence-free condition of the velocity field is maintained at machine precision, with only rounding errors present. These results confirm the stability, accuracy, and structure-preserving capabilities of the proposed method in capturing complex MHD phenomena.

     

/

返回文章
返回