Search

An immersed boundary-discrete unified gas kinetic scheme (IB-DUGKS) is developed in this paper for interface-resolved simulation of particle suspension flows. Compared with the conventional IB methods which consider the interface as a generator of external force, the no-slip boundary condition in the present scheme is implemented directly by correcting the distribution functions near the interface. Therefore, it makes good use of the intrinsic nature for the DUGKS as a kinetic method, and removes simultaneously the necessity to construct the models to evaluate and adsorb the external force in DUGKS. Furthermore, the present IB-DUGKS promotes the accuracy of the IB methods from first- to second-order, which is verified reasonably in the cylindrical Couette flow. After that, several well-established particle-fluid flows are simulated, including the motion of a neutrally buoyancy particle moving in the Poiseuille and Couette flows, and sedimentation of a particle in an enclosure. In all test cases, the results are in good agreement with the data available in the literature. The robust of the present IB-DUGKS is also demonstrated in the simulations. [ABSTRACT FROM AUTHOR]

The lattice Boltzmann equation (LBE) is an efficient kinetic method for particulate flows. Two key issues should be addressed in the implementation of LBE for such systems, i.e., how to treat the curved surface of a solid particle on a uniform Cartesian grid, and how to initialize the state of a fresh node coming from the moving particle. These two key issues are usually considered separately in previous studies. In this work, we propose an efficient unified iterative scheme (UIS) to treat both the issues simultaneously. On one hand, the present method provides a consistent treatment for both the boundary nodes and fresh nodes, on the other hand, to enforce the no-slip boundary condition and decrease the inconsistency between the constructed distribution functions and those evolutionary ones, an enforced iteration (EI) is employed. To describe the inconsistency quantitatively, the inconsistency degree is defined. Simulations of several typical problems are conducted, and the numerical accuracy, computational efficiency and ability to treat moving boundaries are validated. Compared with the combination method, the inconsistency degree around the moving body and spurious force fluctuation are suppressed significantly due to the improved consistency. [ABSTRACT FROM AUTHOR]

This paper studies the robustness of the Smoothed Profile Method (SPM) in fluid-solid interaction problems where inertia and added mass effects are significant. The performance of SPM is compared with that of a Sharp Interface Method (SIM), which is constructed with second-order discretization schemes. The applicability of SPM is extended to the moderate Reynolds numbers regime where the boundary layer and wake dynamics are likely to be influenced by the thickness of the solid-fluid interface. It is shown that SPM is able to correctly predict the drag and surface vorticity for R e < O ( 10 4 ) . In comparison with SIM, SPM is more robust on coarse grids for R e < O ( 10 4 ) ; however, SPM needs fine grids to produce accurate results when R e ∼ O ( 10 4 ) . SPM and SIM are also applied to fluid structure interaction (FSI) problems using an implicit fluid-solid strong coupling framework. This allows computations of FSI problems with moderate inertia and low particle to fluid density ratios (high added mass effects). The effect of the smoothed interface on accuracy and robustness is assessed for such FSI problems. Results indicate that SPM is more robust than SIM when the density of the particle is comparable to the fluid, i.e. for high added mass effects. [ABSTRACT FROM AUTHOR]

Abstract: Fluid particulate flows are common phenomena in nature and industry. Modeling of such flows at micro and macro levels as well establishing relationships between these approaches are needed to understand properties of the particulate matter. We propose a computational technique based on the direct numerical simulation of the particulate flows. The numerical method is based on the distributed Lagrange multiplier technique following the ideas of Glowinski et al. and Patankar . Each particle is explicitly resolved on an Eulerian grid as a separate domain, using solid volume fractions. The fluid equations are solved through the entire computational domain, however, Lagrange multiplier constrains are applied inside the particle domain such that the fluid within any volume associated with a solid particle moves as an incompressible rigid body. Mutual forces for the fluid-particle interactions are internal to the system. Particles interact with the fluid via fluid dynamic equations, resulting in implicit fluid-rigid body coupling relations that produce realistic fluid flow around the particles (i.e., no-slip boundary conditions). The particle-particle interactions are implemented using explicit force-displacement interactions for frictional inelastic particles similar to the DEM method of Cundall et al. with some modifications using a volume of an overlapping region as an input to the contact forces. The method is flexible enough to handle arbitrary particle shapes and size distributions. A parallel implementation of the method is based on the SAMRAI (Structured Adaptive Mesh Refinement Application Infrastructure) library, which allows handling of large amounts of rigid particles and enables local grid refinement. Accuracy and convergence of the presented method has been tested against known solutions for a falling particle as well as by examining fluid flows through stationary particle beds (periodic and cubic packing). To evaluate code performance and validate particle contact physics algorithm, we performed simulations of a representative experiment conducted at the U.C. Berkeley Thermal Hydraulic Lab for pebble flow through a narrow opening. [Copyright &y& Elsevier]

We present a distributed memory many-GPU implementation of the Physalis method for resolving spherical particles in disperse multiphase flow simulations. The current work extends a previous single-GPU computational procedure by implementing a distributed memory Poisson solver and distributed finite-size particle methods using MPI. We document the changes required to move to a distributed memory model for both the fluid and solid phases. We benchmark the code with up to one million resolved particles in a domain size of 192 0 3 on 216 GPUs at the Maryland Advanced Research Computing Center and present strong and weak scaling results. Finally, by taking advantage of the realization that the solution procedure for the pressure Poisson equation can be implemented using a symmetric matrix, we are able to replace the biconjugate gradient stabilized algorithm used in previous work with the conjugate gradient algorithm. • We present a many-GPU implementation of the Physalis method for resolved particle-laden flows. • The implementation uses MPI to distribute fluid and particle memory among GPUs. • We benchmark the strong and weak scaling performance of the code with up to one million resolved particles on 216 GPUs at the Maryland Advanced Research Computing Center. [ABSTRACT FROM AUTHOR]