Computing volume fractions and signed distances from triangulated surfaces immersed in unstructured meshes

Abstract

Available algorithms for the initialization of volume fractions typically utilize exact functions to model fluid interfaces, or they rely on computationally costly intersections between volume meshes. Here, a new algorithm is proposed that computes signed distances and volume fractions on unstructured meshes from arbitrarily shaped surfaces, e.g. originating from experimental data. The proposed algorithm calculates signed distances geometrically near the fluid interface, approximated as a triangle surface mesh, and propagates the inside/outside information by an approximate solution of a Laplace equation. Volume fractions are computed based on signed distances, using either geometrical intersections between cells of the unstructured mesh and a sub-set of the surface mesh that represents the interface, or using a polynomial approximation and adaptive mesh refinement. Although primarily developed for multiphase flow simulations, the proposed algorithm can potentially be used for other problems that require a phase-indicator: inside/outside information with respect to an arbitrarily shaped surface on arbitrarily unstructured meshes.