Bulletin of the American Physical Society
75th Annual Meeting of the Division of Fluid Dynamics
Volume 67, Number 19
Sunday–Tuesday, November 20–22, 2022; Indiana Convention Center, Indianapolis, Indiana.
Session L29: CFD: LBM, SPH, Mesh Free |
Hide Abstracts |
Chair: Wenbin Mao, University of South Florida Room: 237 |
Monday, November 21, 2022 8:00AM - 8:13AM |
L29.00001: Radial-basis-function-based numerical methods for solving compressible flow equations at different Mach numbers Jesus E Arias, Spencer H Bryngelson We present the results of a meshless numerical method, based on radial basis functions (RBFs), for the compressible Navier–Stokes equations at different Mach numbers. The method is based on an RBF– Finite Difference (RBF–FD) discretization of the domain using scattered node sets. Local RBF–FD stencils weights are computed at each node location. The RBF–FD stencil weights encode linear operators like differentiation and interpolation and ultimately comprise global differentiation matrices. We use polyharmonic spline RBFs with appended polynomials for the weight generation process. Boundary conditions are enforced via ghost nodes whose values are computed via the local RBF–FD system at each time step. A hyperviscosity operator stabilizes the resulting system of equations. The governing equations are transformed into a system of ODEs using the global differential operators, boundary conditions, and hyperviscosities. The system is solved via an explicit time integrator. We investigate several benchmark problems to determine the efficacy of this approach for simulating high-Mach flows. |
Monday, November 21, 2022 8:13AM - 8:26AM |
L29.00002: An improved off-lattice algorithm for the open boundary in the lattice Boltzmann method Zongze Li, WENBIN MAO Lattice Boltzmann method (LBM) becomes a popular fluid solver in recent decades due to its natural parallelization and ease of handling complex geometries. Boundary conditions in LBM need special treatments to employ macroscopic values (velocity, pressure, etc.) into discrete distribution functions on these boundary nodes. Many no-slip wall treatments already exist in this field, such as Zou/He, Bouzidi, and Guo algorithms, but the open boundary is still a field needed more study. Although many existing wall treatments can be extended to velocity boundary conditions by adding additional terms, the accuracy might be a concerning shortage, especially on the corner nodes (in 2D) or the edge nodes (in 3D). Hence, in this presentation, we will discuss an improved algorithm based on the Bouzidi algorithm with a focus on the corner/edge node treatments and extend it into the Dirichlet pressure boundary condition. Several numerical tests with different flow types will also be presented to show the better accuracy of our algorithm. |
Monday, November 21, 2022 8:26AM - 8:39AM |
L29.00003: Meshfree methods for fluid flow and applications in the automotive industry Pratik Suchde Meshfree methods have been increasingly studied in the last three decades. The main advantages of meshfree methods are the alleviation of the meshing burden, and the ease of discretizing complex geometries, and the ease of capturing large deformations and displacements. In this talk, I will cover various application of meshfree Lagrangian methods, conducted in collaboration with partners from the automotive industry. |
Monday, November 21, 2022 8:39AM - 8:52AM |
L29.00004: Multi-architecture Adaptive Mesh Refinement Lattice-Boltzmann Method for multi-phase and porous media flows Evangelos Stavropoulos Vasilakis, Pierre Kestener, Alain Cartalade, Alain Genty The work presented herein proposes an implementation of an Adaptive Mesh Refinement (AMR) algorithm forLattice Boltzmann Methods, focusing on code portability and optimisation on different architectures. The Kokkos C++ scientific computing library is used, to exploit various available high performance computing (HPC) resources, CPUs or GPUs, shared or distributed memory systems alike, and stay up-to-date with their technological advancements. |
Monday, November 21, 2022 8:52AM - 9:05AM Not Participating |
L29.00005: Accelerated Transport of Highly Viscous Liquid through Microchannels by Multiphase Flow over Superhydrophobic Surface. Faisal Tushar, Zhen Li Flow resistance of a fluid moving through a cylindrical pipe increases rapidly with decreasing pipe diameter. However, as a channel decreases in size, the surface area-to-volume ratio becomes large, wherein the surface forces dominate the fluid transport. By taking the advantage of interfacial forces in drop-based transport of liquid, we investigate the transport of highly viscous droplets moving through microchannels with superhydrophobic surface using many-body dissipative particle dynamics simulations. Two liquid droplets with similar surface tension but significant differences in viscosity are considered. When the same pressure gradient is applied to drive the two different droplets, a faster motion of the higher viscous droplet than the lower viscous droplet is observed by a comparison of their center of mass velocities. This observation is opposite to traditional continuous fluid flow through microchannels but is consistent with recent experiments on viscosity-enhanced droplet motion. We quantify how viscosity, surface wettability, skin friction, and air-liquid surface tension affect the motion of viscous droplets as well as the internal flow field induced inside the moving droplets to improve our understanding on the mechanism of this anomalous flow phenomenon. |
Monday, November 21, 2022 9:05AM - 9:18AM |
L29.00006: Physics Informed Machine Learning with Smoothed Particle Hydrodynamics: Compressiblity and Shocks Michael Woodward, Yifeng Tian, Criston M Hyett, Chris L Fryer, Daniel Livescu, Mikhail Stepanov, Michael Chertkov Many formulations of Smoothed Particle Hydrodynamics (SPH) have been developed for handling compressible flows, such as those seen in astrophysics and engineering applications. Often, these formulations are first tested and developed on simple 1D and 2D shock problems in order to validate the modeling and numerical implementations. Nevertheless, the parameters of the formulation are usually adjusted in an ad-hoc manner, by trial and error. In this work, we explore the physics informed machine learning methods developed in [Woodward et al, arxiv:2110.13311, 2021], but using a compressible SPH formulation, to learn the SPH parameters to model shock waves. The formulation is verified against analytic solutions such as the 1D Sod and 2D Taylor-Sedov problems. Specifically, combining deep learning, automatic differentiation, and local sensitivity analysis, learn-able and parameterized compressible SPH formulations are constructed and fit to the analytical solutions of 1D and 2D shocks. These parameterizations are used to learn new parameterized smoothing kernels and artificial viscosity terms, with the goal of improving the SPH framework with respect to accuracy and modeling. |
Monday, November 21, 2022 9:18AM - 9:31AM Author not Attending |
L29.00007: A High-Order Meshless Method for Solving Multiphysics Problems in Complex Domains Surya P Vanka, Shantanu Shahane We present a high-order meshless method based on scattered data interpolation for solving Multiphysics problems in complex domains. A complex domain is first represented by scattered points generated by any method, including as vertices of a finite element grid. The flow variables are located at the scattered points and interpolated over a local cloud of points using a radial basis function (RBF). We consider Polyharmonic splines (PHS) which do not need a shape parameter. We further append a polynomial of desired degree to achieve high order discretization accuracy. To solve the Navier-Stokes equations, the partial-derivatives are evaluated by differentiating the radial basis functions and the polynomials. We have developed an explicit fractional step method and a semi-implicit method to integrate the Navier-Stokes and energy equations in time with high spatial and second-order temporal accuracies. The explicit fractional step method is limited by the combined convective and diffusive Courant numbers while the semi-implicit method can use time steps 10-20 times the explicit time step. A three-dimensional extension for periodic problems has been developed with Fourier expansions in the periodic direction. We have demonstrated the method has high spatial discretization accuracy. |
Monday, November 21, 2022 9:31AM - 9:44AM |
L29.00008: Study of Taylor Couette flow between an elliptical enclosure and an inner rotating cylinder Akash Unnikrishnan, Shantanu Shahane, Surya P Vanka, Vinod Narayanan Taylor-Couette flows have been extensively studied in a gap between concentric cylinders. However, fewer studies exist of Taylor cells in the gap between a non-circular enclosure and an inner rotating circular cylinder. In this study, a meshless method is used to study such Taylor cells in the gap formed by an elliptical enclosure and a cylinder. We have considered ellipses of aspect ratios two and three. The meshless method uses Polyharmonic spline radial basis functions (PHS-RBF) with appended polynomials for interpolating scattered data. In the periodic streamwise direction, Fourier expansions are used. The discrete equations are obtained by differentiation of radial basis functions and collocation of the partial-differential equations at the scattered points. A second-order accurate fractional step algorithm solves the discrete equations in time. The first onset of Taylor cells and their characteristics at increasing Reynolds numbers are studied for two aspect ratio of outer enclosure. The Taylor Couette cells are observed to stretch and compress to fit the varying gap between the outer enclosure and the inner cylinder. The torque on the inner cylinder and the velocity distributions are presented. These flow patterns can have important effects on mixing and particle dynamics. |
Monday, November 21, 2022 9:44AM - 9:57AM |
L29.00009: Study of film dynamics using lattice Boltzmann method Garima Singh, Naveen Tiwari In this work the dynamics of a spreading film is numerically modelled using the phase-field lattice Boltzmann approach. The fluid-fluid interface is mesoscopic in nature, making the lattice Boltzmann method (LBM) a useful technique for modelling two-phase systems. A geometry-based wetting boundary condition has been used to model the three-phase contact point. The two-dimensional interfacial pattern obtained using LBM matches with the profile obtained using the analytical model derived within the lubrication limit for a thin film spreading on an inclined plane. We present a detailed analysis of the effect of aspect ratio (film-thickness versus capillary length) and viscosity ratio (bottom to top fluid) on the spreading dynamics. The study reveals that the dimensionless height of the ridge's tip approaches unity rapidly as we approach the thick films away from the validity of the lubrication approximation. The ridge completely vanishes at a critical aspect ratio and is shown to be independent of the advancing contact angle. A protruding structure at the contact point has been observed for aspect ratios beyond the critical value and is also found sensitive to contact angle. Furthermore, the viscous effects of surrounding fluid have been investigated on the interfacial pattern of spreading film. Increasing the viscous effects of surrounding fluid results in an enhanced ridge height and, as a result, makes the film more prone to instability. |
Monday, November 21, 2022 9:57AM - 10:10AM |
L29.00010: Lattice Boltzmann Simulations of Thermocapillary Convection in Self-rewetting Fluid Layers and Bubble Dynamics William T Schupbach, Bashir Elbousefi, Kannan Premnath Self-rewetting fluids (e.g., aqueous solutions of long-chain alcohols, certain liquid binary alloys, and nematic liquid crystals) exhibit anomalous nonlinear dependence of surface tension on temperature with a positive gradient. They involve significantly altered interfacial dynamics compared to ordinary fluids, which have recently been exploited in various applications. We will perform a computational study of thermocapillary convection in self-rewetting fluids in various configurations. To simulate such flows, we develop and utilize a robust central moment lattice Boltzmann (LB) method, which involve evolving three distribution functions; one to compute the fluid motion with attendant Marangoni stresses due to surface tension gradient, another one for interface tracking represented by the conservative Allen-Cahn equation, and finally, the third one to compute energy transport. First, thermocapillary convection in superimposed self-rewetting fluid layers in a microchannel driven by a periodic heating will be simulated and compared to newly developed analytical solution under creeping flow regime. Then, rising bubbles in self-rewetting fluids will be studied. Finally, the effect of anomalous surface tension behavior in mass transfer due to phase change will be presented. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700