Properties of median-dual regions on triangulations in $\mathbb{R}^{4}$ with extensions to higher dimensions
David M. Williams, Hiroaki Nishikawa
TL;DR
The paper addresses efficiency in 4D space-time CFD by enabling node-centered edge-based schemes that rely on median-dual regions but avoid explicit dual-cell construction. It defines median-dual regions in $\mathbb{R}^d$ and derives a dimension-general hypervolume formula and a 4D directed-hyperarea formula, both expressed entirely in terms of simplex geometry. The hypervolume identity holds for all $d$, while the directed-hyperarea formula is proven for $d=4$, with detailed implementation and boundary considerations. The results enable on-the-fly residual calculations using precomputed edge-directed-hyperarea data, facilitating robust, conservation-compatible schemes on general triangulations and guiding future extensions to higher dimensions.
Abstract
Many time-dependent problems in the field of computational fluid dynamics can be solved in a four-dimensional space-time setting. However, such problems are computationally expensive to solve using modern high-order numerical methods. In order to address this issue, efficient, node-centered edge-based schemes are currently being developed. In these schemes, a median-dual tessellation of the space-time domain is constructed based on an initial triangulation. Unfortunately, it is not straightforward to construct median-dual regions or deduce their properties on triangulations for $d \geq 4$. In this work, we provide the first rigorous definition of median-dual regions on triangulations in any number of dimensions. In addition, we present the first methods for calculating the geometric properties of these dual regions. We introduce a new method for computing the hypervolume of a median-dual region in $\mathbb{R}^d$. Furthermore, we provide a new approach for computing the directed-hyperarea vectors for facets of a median-dual region in $\mathbb{R}^{4}$. These geometric properties are key for facilitating the construction of node-centered edge-based schemes in higher dimensions.
