Fast micromagnetic simulations using an analytic mathematical model

In this paper an analytic mathematical model is presented for fast micromagnetic simulations. In dynamic micromagnetic simulations the Landau–Lifshitz–Gilbert (LLG) equation is solved for the observation of the reversal magnetisation mechanisms. In stiff micromagnetic simulations the large system of ordinary differential equations has to be solved with an appropriate method, such as the Backward Differentiation Formulas (BDF) method, which leads to the solution of a large linear system. The latter is solved efficiently employing matrix-free techniques, such as Krylov methods with preconditioning. Within the Krylov methods framework a product of a matrix times a vector is involved which is usually approximated with directional differences. This paper provides an analytic mathematical model to calculate efficiently this product, leading to more accurate calculations and consequently faster micromagnetic simulations due to better convergence properties.