Near-wall resolution and wall modeling for lattice Boltzmann simulations of mechanically agitated vessels. (31st December 2022)

• Record Type: Journal Article
• Authors: Flamm, Matthew H.; Kalal, Zbynek
• Is Part Of: Chemical engineering science. Volume 264(2022)
• Publication Date: 2022-12-31
• Subjects: Mixing -- Computational fluid dynamics -- Lattice Boltzmann -- Wall modeling -- Large Eddy Simulation; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2022.118172
• Languages: English
• ISSNs: 0009-2509

Abstract

Highlights: Under-resolving the near-wall region leads to incorrect flow prediction in stirred tanks. Using a wall model improves prediction at coarse resolutions. Better wall models are still needed to remove bias at fine resolutions. Abstract: We demonstrate the importance of near-wall treatment in lattice Boltzmann models (LBM) for mechanically agitated vessels at high Reynolds numbers typical of industrial mixing. LBM simulations were performed with and without wall models at multiple resolutions. Results were compared to experimental data in the impeller stream and the near-wall region. Neglecting wall models when under-resolving the wall regions, which is the typical practice for LBM simulations of stirred vessels, led to incorrect flow prediction. Either extreme spatial resolution in the near-wall region or the use of wall models improved predictions when comparing to other software methodologies (Finite Volume Method) and experimental data. However, the use of wall stress-based models in this work led to a small bias that persists even at higher resolutions whether using finite volume or lattice Boltzmann methods. More accurate wall models should be investigated for more accurate predictions in turbulent mechanically agitated vessels.