An efficient and accurate penalty-projection eddy viscosity algorithm for stochastic magnetohydrodynamic flow problems (Q6608073)

This paper proposes a stabilized penalty projection algorithm to numerically approach the solution of the stochastic magnetohydrodynamic (MHD) flow problems in convection dominated regimes. The method mainly combines a stochastic collocation method (SCM) and a stabilized penalty projection finite element method (SPP-FEM). The system is decoupled in four subproblems to solve at each time step. The authors manage that the system matrix remains the same at each time step, which provides a significant computational time reduction. The main mathematical result is that the proposed method is unconditionally stable with respect to the time step size. Moreover, the authors prove that the proposed numerical scheme converges to a non projection based MHD scheme for large stabilization parameters. Numerical simulations successfully illustrate the effectiveness of the algorithm and the theoretical results.