An efficient physics-based preconditioner for the fully implicit solution of small-scale thermally driven atmospheric flows

摘要: In atmospheric flow situations typical of a small-scale thermal, separation time scales exists between the fast sound wave scale and advective scale. Atmospheric models have been designed to take advantage this disparity with numerical approaches such as semi-implicit or split-explicit approach being used efficiently step over waves. Some these are first order in time. To improve accuracy methods, fully implicit nonlinearly consistent (INC) solver has developed for Navier-Stokes equation set. our INC method, set is solved by use Jacobian-free Newton-Krylov (JFNK) method. An efficient preconditioner which uses method solve governing equations. Being that was attack fastest waves system not other features advection turbulent diffusion, preconditioning technique labeled physics-based preconditioner. A variety linear solvers including SSOR, Krylov methods and/or multigrid approximately invert pressure matrix algorithm. suite simulations will be conducted utilizing different simple problem bouyant rise warm bubble. The also document ability achieve second accuracy.