Improvements in speed for explicit, transient compressible flow solvers

Abstract

Several explicit high‐resolution schemes for transient compressible flows with moving shocks are combined in such a way so as to achieve the highest possible speed without compromising accuracy. The main algorithmic changes considered comprise the following: replacing limiting and approximate Riemann solvers by simpler schemes during the initial stages of Runge–Kutta solvers, and only using limiting and approximate Riemann solvers for the last stage; automatically switching to simpler schemes for smooth flow regions; automatic deactivation of quiescent regions; and unstructured grids with cartesian cores or embedded cartesian grids. The results from several examples demonstrate that speedup factors of 1:4 are attainable without compromising the accuracy of the traditional FCT schemes. Copyright © 2007 John Wiley & Sons, Ltd.

Detailed Simulation of Viral Propagation In The Built Environment

“…-Use of unstructured grids (automatic grid generation and mesh refinement); -Finite element discretization of space; -Separate flow modules for compressible and incompressible flows; -Edge-based data structures for speed; -Optimal data structures for different architectures; -Bottom-up coding from the subroutine level to assure an open-ended, expandable architecture. The code has had a long history of relevant applications involving compressible flow simulations in the areas of transonic flow [81,82,83,84,63,85], store separation [7,10,12,14,15], blast-structure interaction [6,8,11,13,16], [66,72,94,105,97], incompressible flows [91,93,65,68,5,76], free-surface hydrodynamics [62,69,70], dispersion [22,23,67,24], patient-based haemodynamics [26,63,27,1,73] and aeroacoustics [57]. The code has been ported to vector [64], shared memory [61,96], distributed memory…”

Spatial accuracy and performance of a mixed‐order, explicit multi‐stage method for unsteady flows

“…Ideally, we would like to limit the use of higher-order flux calculations to once per time step while simultaneously using the largest time step allowed by stability constraints, which corresponds to a Courant number of 1.0. This type of method was briefly discussed in [13], albeit within the context of a number of speed enhancement techniques and the maximum number of stages was also limited to three.…”