A staggered grid finite difference method based on global optimization

Staggered grid finite difference techniques have been widely used in numerical simulation of seismic wave field， but the interpolation might enlarge the simulation errors when the elastic moduli change dramatically. Rotated staggered grid can overcome this drawback thus be more applicable for modeling the anisotropic media. However， for the grid cells with the same size， rotated staggered grid needs larger step than conventional staggered grid does， consequently， the gradient operator and divergence operator generate larger errors which are more likely to cause spatial numerical dispersion. Aiming at the above problems， a combination of rotated staggered grid and compact finite difference method is proposed， simultaneously， global optimization is preformed based on simulated annealing algorithm to suppress numerical dispersion and to broaden the range of wavenumber. Numerical simulation results show that this method can effectively suppress the numerical dispersion and exhibits high simulation accuracy at the same time.