Conservative, density-based smoothed particle hydrodynamics with improved partition of the unity and better estimation of gradients
Domingo Garc\'ia-Senz, Rub\'en M. Cabez\'on, Jos\'e A. Escart\'in

TL;DR
This paper introduces a new volume element approach in smoothed particle hydrodynamics that improves gradient estimation accuracy and partition of unity, enhancing the method's robustness and conservation properties in complex flow simulations.
Contribution
It proposes a novel volume element variant that improves partition of unity and gradient accuracy, compatible with Lagrangian SPH, and introduces a flexible scheme for handling sharp density contrasts.
Findings
Improved partition of unity enhances gradient estimation accuracy.
The new scheme performs well in standard tests with contact discontinuities.
Conservation of energy, momentum, and entropy remains robust with the new approach.
Abstract
The correct evaluation of gradients is at the cornerstone of the smoothed particle hydrodynamics (SPH) technique. Using an integral approach to estimate gradients has proven to enhance accuracy substantially. Such approach retains the Lagrangian structure of SPH equations and is fully conservative. But, in practice, it is difficult to make the Lagrangian formulation totally consistent to an exact partition of the unity. In this paper we study, among other things, the connection between the choice of the volume elements (VEs), which enters in the SPH summations, and the accuracy in the gradient estimation within the integral approach scheme (ISPH). A new variant of VEs are proposed which improve the partition of the unity and is fully compatible with the Lagrangian formulation of SPH, including the grad-h corrections. Using analytic considerations, simple static toy models in 1D, and a…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
