2022/02/16 by Geert Buckinx, Buckinx, Geert
Engineering · #Aerodynamics and Fluid Dynamics Research #FOS: Physical sciences #Fluid Dynamics (physics.flu-dyn) #Fluid Dynamics and Vibration Analysis #Lattice Boltzmann Simulation Studies
paper · pdf · doi:10.48550/arxiv.2202.08245
openalex publication_date 2022/02/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We present a macro-scale description of quasi-periodically developed flow in channels, which relies on double volume-averaging. We show that quasi-developed macro-scale flow is characterized by velocity modes which decay exponentially in the main flow direction. We prove that the closure force can be represented by an exact permeability tensor consisting of two parts. The first part, which is due to the developed macro-scale flow, is uniform everywhere, except in the side-wall region, where it is affected by the macro-scale velocity profile and its slip length. The second part expresses the resistance against the velocity mode, so it decays exponentially as the flow develops. It satisfies a specific closure problem on a transversal row of the array. From these properties, we assess the validity of the classical closure problem for the volume-averaged flow equations. We show that all its underlying assumptions are partly violated by an exponentially vanishing error during flow development. Furthermore, we show that it modifies the eigenvalues, modes, and onset point of quasi-developed flow, when it is applied to reconstruct the macro-scale flow. The former theoretical aspects are illustrated for high-aspect-ratio channels with high-porosity arrays of equidistant in-line square cylinders, by means of direct numerical simulation and explicit filtering of the flow. In particular, we present extensive solutions of the classical closure problem for Reynolds numbers up to 600, porosities between 0.2 and 0.95, and flow directions between 0 and 45 degrees, though the channel height has been kept equal to the cylinder spacing. These closure solutions are compared with the actual closure force in channels with cylinder arrays of a porosity between 0.75 and 0.94, for Reynolds numbers up to 300.