Application of the Lattice-Boltzmann Method for Particle-laden Flows: Point-particles and Fully Resolved Particles

摘要

The Lattice-Boltzmann-Method (LBM) is a powerful and robust approach for calculating fluid flows over or through complex geometries. This method was further developed for allowing the calculation of several problems relevant to dispersed particle-laden flows. For that purpose two approaches have been developed. The first approach concerns the coupling of the LBM with a classical Lagrangian procedure where the particles are considered as point-masses and hence the particles and the flow around them are numerically not resolved. As an example of use, the flow through a single pore representing a single element of a filter medium was considered and the deposition of nano-scale particles was simulated. The temporal evolution of the deposit structures is visualised and both the filtration efficiency and the pressure drop are simulated and compared with measurements. In the second developed LBM-approach, the particles are fully resolved by the numerical grid whereby the flow around particles is also captured and it is possible to effectively calculate forces on complex particles from the bounce-back boundary condition. As a case study the flow around spherical agglomerates consisting of poly-sized spherical primary particles with sintering contact is examined. Using local grid refinement and curved wall boundary condition, accurate simulations of the drag coefficient of such complex particles were performed. Especially the effect of porosity on the drag was analysed. Moreover, the flow about very porous fractal flocks, generated by a random process, was simulated for different flock size and fractal dimension. The drag coefficients resulting from LBM simulations were compared to theoretical results for Stokes flow. Finally, scenarios with moving particles were considered. First, the sedimentation of a single particle towards a plane wall was simulated and compared with measurements for validation. Secondly, the temporal sedimentation of a cluster of 13 particles was studied. Here, the primary particles were allowed to stick together and form agglomerates. This research will be the basis for further analysing agglomerate formation in laminar and turbulent flows.