Optimal and near-optimal advection-diffusion finite-difference schemes. V. Error propagation
نویسنده
چکیده
The wave concept of group (or energy) velocity is used to predict how errors propagate in three numerical schemes for the solution of the diffusion equation with flow and decay. A five-point filter is used to isolate the long and the saw-tooth gridscale errors. Contour plots are presented which illustrate the markedly different propagation of the long and saw-tooth errors. The energy of long errors is carried downstream with the flow. An illustrative example is given for which, as predicted, the saw-tooth errors go upstream.
منابع مشابه
Positivity-preserving nonstandard finite difference Schemes for simulation of advection-diffusion reaction equations
Systems in which reaction terms are coupled to diffusion and advection transports arise in a wide range of chemical engineering applications, physics, biology and environmental. In these cases, the components of the unknown can denote concentrations or population sizes which represent quantities and they need to remain positive. Classical finite difference schemes may produce numerical drawback...
متن کاملOptimal and near-optimal advection-diffusion finite-difference schemes 4. Spatial non-uniformity
the non-uniformity can almost be removed with a local change of variables from c(s, t) to b(x, t). So, in numerical computations non-uniformity need not imply any reduction from optimal accuracy. A computational grid that was uniformly spaced with respect to s becomes non-uniform with respect to the intrinsic xcoordinate. Suitable non-uniform grid, optimal and near-optimal compact finitediffere...
متن کاملOptimal and near - optimal advection - diffusion finite - difference schemes . Part 1 : constant coefficient 1 - D
An implicit finite difference scheme is derived which achieves exact results for the area, centroid, variance, skewness and kurtosis (flatness) of solutions to the unforced advection-diffusion equation ∂ t c + λ c + u ∂ x c − κ ∂ 2 x c = 0. Sufficient conditions for computational stability are that the grid spacing ∆x and time step ∆t be small enough that |u| ∆x κ < 12 1/2 , |u| ∆t ∆x < 1 2 1/2...
متن کاملOptimal and near-optimal advection-diffusion finite-difference schemes. Part 2: Unsteadiness and non-uniform grid
∂tc+ λ c+ u ∂xc− κ ∂ xc = q + ∂xf + ∂ xg . At the grid points the extremely high order of approximation for the numerical solutions is such that if loss of accuracy is to be avoided then interpolation must use values extending beyond the local 3 by 2 computational module. Illustrative examples show that reasonable accuracy is possible with extremely long time steps on sparse non-uniform, moving...
متن کاملApproximation of stochastic advection diffusion equations with finite difference scheme
In this paper, a high-order and conditionally stable stochastic difference scheme is proposed for the numerical solution of $rm Ithat{o}$ stochastic advection diffusion equation with one dimensional white noise process. We applied a finite difference approximation of fourth-order for discretizing space spatial derivative of this equation. The main properties of deterministic difference schemes,...
متن کامل