Pressure-velocity algorithms

Note
Under construction - please check again later

The background to the SIMPLE, PISO and PIMPLE pressure-velocity algorithms can be demonstrated using the incompressible, inviscid flow equations, comprising the momentum equation:

\[ \ddt{\u} + \div (\u \otimes \u) = - \grad p, \]

and continuity equation

\[ \div \u = 0. \]

Disctretising the momentum equation leads to a set of algebraic equations of the form:

\[ M[\u] = - \grad p \]

where the matrix \(M[\u]\) comprises the diagonal and off-diagonal contributions using the decomposition:

\[ M[\u] = A\u - \vec{H} \]

The discretised momentum equation therefore becomes:

\[ A\u - \vec{H} = - \grad p \]

which on re-arranging leads to the velocity correction equation:

\[ \u = \frac{\vec{H}}{A} - \frac{1}{A} \grad p. \]

The volumetric flux corrector equation is then derived by interpolating \( \u \) to the faces and dotting the result with the face area vectors, \( \vec{S}_f \):

\[ \phi = \u_f \dprod \vec{S}_f = \left( \frac{\vec{H}}{A} \right)_f \dprod \vec{S}_f - \left( \frac{1}{A} \right)_f \vec{S}_f \dprod \snGrad p \]

Discretistation of the continuity equation yields the constraint:

\[ \div \phi = 0. \]

Substituting the flux equation leads to the pressure equation:

\[ \div \left[ \left( \frac{1}{A} \right)_f \grad p \right] = \div \left( \frac{\vec{H}}{A} \right)_f. \]


Would you like to suggest an improvement to this page? Create an issue

Copyright © 2018-2019 OpenCFD Ltd.

Licensed under the Creative Commons License BY-NC-ND Creative Commons License