REVIEW 1 cited by
Method of Lines Transpose: A Fast Implicit Wave Propagator
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
abstract
As a follow up to \cite{Causley2013}, we provide a detailed description of the numerical implementation of an O(N), A-stable, second order accurate solution of the wave equation, constructed from semi-discrete boundary value problems. We improve on the previous algorithm by replacing the Lax-type correction used in \cite{Causley2013}, which was necessary for convergence when $\Delta t < \Delta x/c$, with a more accurate spatial quadrature, which we prove is convergent. We also demonstrate that the resulting solver remains fast even in the case of unstructured meshes, can incorporate domain decomposition, and allows for the implementation of Dirichlet, Neumann, periodic and outflow boundary conditions. Building upon results for the 1d formulation, we utilize alternate direction implicit (ADI) splitting to achieve a fast O(N) solver in higher spatial dimensions. Our solver is built upon line objects and, combined with the flexibility of the integral solver, allows us to solve problems on arbitrary spatial domains, by embedding the boundary in a regular Cartesian mesh. Our solver is designed to couple with particle codes, where scale separation is an issue. We therefore demonstrate the ability of our solver to take time steps well beyond that of the Courant-Friedrichs-Lewy (CFL) stability limit of explicit codes.
Forward citations
Cited by 1 Pith paper
-
A Kernel Based High Order "Explicit" Unconditionally Stable Constrained Transport Method for Ideal Magnetohydrodynamics
A kernel-based solver for the magnetic vector potential gives a high-order, diffusion-free constrained transport update for ideal MHD in 2D and 3D.
Discussion (0). Continue with ORCID to comment.