A comparative study of Lattice-Boltzmann methods for supersonic flow simulations
D. Krasauskas (Student TU Delft)
D. Casalino (TU Delft - Aerospace Engineering)
S. Hulshoff (TU Delft - Aerospace Engineering)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This paper presents a comparative benchmark study of different lattice Boltzmann schemes for simulating ideal-gas supersonic flows. In particular, two approaches are investigated: a D3Q33 scheme that solves the momentum and total energy equations using two distinct particle distribution functions, and a D2Q49 scheme based on an entropy Lattice Boltzmann formulation, also employing two populations. These methods are implemented in two different computational frameworks, namely a CPU-based industrial solver and a GPU-based research solver. Their numerical properties are first analyzed in terms of dissipation and dispersion errors through the propagation of a Gaussian plane wave. Subsequently, the classical two-dimensional Schardin’s shock tube problem is used to evaluate both accuracy and computational performance. In addition to validation and algorithmic comparison, the study introduces an original GPU-oriented optimization strategy, which is essential for enabling future three-dimensional extensions of the entropic LBM scheme and represents a key contribution of this work.