On pressure boundary conditions for the incompressible Navier‐Stokes equations

Abstract

SUMMARYThe pressure is a somewhat mysterious quantity in incompressible flows. It is not a thermodynamic variable as there is no 'equation of state' for an incompressible fluid. It is in one sense a mathematical artefact-a Lagrange multiplier that constrains the velocity field to remain divergence-free; i.e., incompressible-yet its gradient is a relevant physical quantity: a force per unit volume. It propagates at infinite speed in order to keep the flow always and everywhere incompressible; i.e., it is always in equilibrium with a time-varying divergencefree velocity field. It is also often difficult and/or expensive to compute. While the pressure is perfectly welldefined (at least up to an arbitrary additive constant) by the governing equations describing the conservation of mass and momentum, it is (ironically) less so when more directly expressed in terms of a Poisson equation that is both derivable from the original conservation equations and used (or misused) to replace the mass conservation equation. This is because in this latter form it is also necessary to address directly the subject of pressure boundary conditions, whose proper specification is crucial (in many ways) and forms the basis of this work. Herein we show that the same principles of mass and momentum conservation, combined with a continuity argument, lead to the correct boundary conditions for the pressure Poisson equation: viz., a Neumann condition that is derived simply by applying the normal component of the momentum equation at the boundary. It usually follows, but is not so crucial, that the tangential momentum equation is also satisfied at the boundary.

Left Ventricular Fluid Mechanics: The Long Way from Theoretical Models to Clinical Applications

“…(3), although care must be taken in imposing appropriate boundary conditions. 13,25,37 Because Eq. (4) is second order and the average pressure gradient (tri-linear terms) follows from the boundary conditions only.…”

A cell-centered adaptive projection method for the incompressible Navier–Stokes equations in three dimensions

“…Domain boundary conditions are the standard projection boundary conditions ( ∂φ ∂n = 0 at solid wall boundaries) [14]. Then the face-centered velocity field is corrected:…”

Authors' Reply

“…This criticism is irrelevant to [1]. Furthermore, the claims in [2] were made on a rather informal basis in which case, just like a conjecture, one can later publish results that contradict a conjecture without having to be accused of being wrong.…”