Bulletin of the American Physical Society
71st Annual Meeting of the APS Division of Fluid Dynamics
Volume 63, Number 13
Sunday–Tuesday, November 18–20, 2018; Atlanta, Georgia
Session L32: Immersed Boundary Methods |
Hide Abstracts |
Chair: Haoxiang Luo, Vanderbilt University Room: Georgia World Congress Center B404 |
Monday, November 19, 2018 4:05PM - 4:18PM |
L32.00001: A sharp-interface method for computing the shock-induced evaporation rate of droplet Pratik Das, Oishik Sen, Gustaaf B Jacobs, H.S. Udaykumar Shock-induced evaporation of droplets governs the rate of combustion and subsequent deposition of energy in scramjet engines, rocket motors, heterogenous explosions etc. Therefore, it is important to accurately predict the evaporation rate of the droplets during interaction with shock waves. In this work, a sharp-interface method is developed to calculate the evaporation rate of the droplets. The level-set method is used to track the liquid-gas interface. A robust ghost fluid method is developed to couple the flow fields of the liquid and the gaseous phase at the interface. A modified interfacial Riemann problem is solved at every point on the interface to account for the jump in the pressure and the normal velocity fields due to surface tension and evaporation. The tangential components of the velocity fields for the ghost fluid are obtained such that the jump in the stresses at the interface due to the Marangoni effect is respected. The current sharp-interface method is validated against analytical solutions of 1-D Riemann problems and experimental studies of shock interaction with water-column. Subsequently, simulations of shock-droplet interactions are performed to estimate the shock-induced evaporation rate of the droplets. |
Monday, November 19, 2018 4:18PM - 4:31PM |
L32.00002: Large-scale parallel simulation of fluid-structure interaction based on immersed-boundary method Ye Chen, Zheng Li, Haoxiang Luo A three-dimensional parallel simulation tool of the fluid-structure interaction (FSI) using a direct-forcing immersed-boundary method is presented. The computational framework is based on a partitioned approach and can handle a range of biological FSI problems involving large deformations. We have implemented an efficient parallel strategy based on a simple idea of domain decomposition. For the flow, the computational domain is decomposed into subdomains in either 1D, 2D, or 3D, thus allowing the use tens of thousands of processors for parallel computing. The immersed-boundary treatment, including identification of the IBLANK (i.e., determination of the fluid and solid nodes on the grid) and direct forcing, is done within each subdomain and is well scaled like the other parts of the solution process for the numerical PDE. Furthermore, the total memory allocation is nearly constant regardless the number of processors. These features allow simulations using a mesh size of one billion points. We will demonstrate applications in several problems, including glottal airflow/vocal fold, heart valve, and animal flight. |
Monday, November 19, 2018 4:31PM - 4:44PM |
L32.00003: Fluid-structure interaction of engineering geometries using a combined immersed finite element method and finite volume incompressible multiphase solver for high density and high shear flows Nishant Nangia, Amneet Pal Singh Bhalla, Neelesh Ashok Patankar Many industrial fluid flow problems involve the interaction between rigid, heavy objects and one or more fluid phases. In the past few years, the constraint-based immersed boundary method (CIB) has been successfully used to simulate a wide range of fluid-structure interaction (FSI) problems. This method is robust since it can simulate arbitrarily moving bodies on regular Cartesian meshes, making use of adaptive mesh refinement near the fluid-solid interface to adequately resolve boundary layers. In this work, we extend the CIB formulation and method to allow for a finite element based representation of the structure, enabling fully resolved simulation of industrial geometries. The method is coupled to a novel monolithic incompressible multiphase fluid solver. We show preliminary validation cases for both simple and complex geometries. Novel applications of this method include simulation of self-propelled vehicle aerodynamics and wave energy converter devices, in which the density ratio between the fluid and solid regions are more than three orders of magnitude. |
Monday, November 19, 2018 4:44PM - 4:57PM |
L32.00004: Fast level-set-based boundary reconstruction of highly complex body configurations for studying bio-inspired locomotion using a sharp-interface immersed boundary method Xiaolong Deng, Haibo Dong The 3-D finite-difference sharp-interface immersed boundary method (Mittal et al., JCP 2008) has been widely applied to various bio-inspired flow simulations. Immersed boundary reconstruction is the first step before the flow solution. Its efficiency has strong impact on the simulations with moving bodies, in which the reconstruction needs to be done in each time step. In this talk, we present a new fast level-set-based reconstruction method for flow simulation of highly complex bodies. The level set values are calculated in the direct neighbor grids around the solid boundary as a signed distance function, and then propagated to other grids through the entire computational domain in a highly-efficient way. Then the interior/exterior grids are identified by the sign of level set value. After this the flow field can be solved with the sharp-interface immersed boundary method. Algorithm analysis shows the new method has smaller computational complexity than the direct searching method and the ray-tracing method. Applications to swimming dolphin, manta ray and trout with dense meshes show the new method not only significantly saves boundary reconstruction time, but also improves the error-tolerance of reconstruction results. |
Monday, November 19, 2018 4:57PM - 5:10PM |
L32.00005: Flapping dynamics of an inverted flag with different shapes in a uniform flow Jeong Woo Park, Jaeha Ryu, Hyung Jin Sung The flapping motion of an inverted flag with different shapes in a uniform flow was simulated using an immersed boundary method. The shapes of the flag were distinguished by the shape ratio (S=W_{T }/ W_{L}) defined as the ratio of the trailing edge width (W_{T}) to the leading edge width (W_{L}). To investigate the effects of the shape ratio on the dynamics of the inverted flag, the peak-to-peak amplitude (A/L) and the Strouhal number (St) were analyzed as a function of the bending rigidity (0.1 ≤ γ ≤ 0.3) and the shape ratio (0.5 ≤ S ≤ 2). The vortical structures behind the inverted flag were visualized by the Q-criterion to examine the vortex dynamics. The hydrodynamic forces exerted on the flag were explored to reveal the correlation between the kinematics of the inverted flag and the vortex formation during a flapping period. Finally, we estimated the strain energy (E_{s}) stored on the inverted flag and the conversion ratio (R) of the flow kinetic energy to the strain energy. |
Monday, November 19, 2018 5:10PM - 5:23PM |
L32.00006: A consistent and conservative immersed boundary method for moving boundary problems Jun-Hua Pan, Nian-Mei Zhang, Ming-Jiu Ni A consistent and conservative immersed boundary method has been developed to accurately and efficiently solve moving boundary problems. Based on a least square interpolation reconstruction method, a unified form of boundary condition is described to handle a general boundary condition. In order to achieve the consistency with a desired wall boundary condition, a so-called approximation-correction procedure is done. Besides, a consistent and conservative scheme is implemented to satisfy mass conservation laws on cells around the immersed surface. Then, a conservative interpolation is reconstructed for velocity in moving boundary problems. At last, the applied numerical method is validated by stationary and moving boundary cases and an excellent agreement is obtained with good accuracy, efficiency and conservation. Especially, the consistent and conservative immersed boundary method can obtain almost the same accurate results as those from the cut cell technique (J.H. Seo, R. Mittal, J. Comp. Phys. 230 (2011) 7347-7363.) for a moving boundary problem by reducing the spurious pressure. |
Monday, November 19, 2018 5:23PM - 5:36PM |
L32.00007: Fluid Structure Interaction of Linear Viscoelastic Splitter Plate at Low Reynolds Number Rahul Mishra, Salil S. Kulkarni, Rajneesh Bhardwaj The flow-induced deformation of a viscoelastic thin plate, attached to the rear of a circular cylinder subjected to laminar flow is numerically investigated. An in-house fluid-structure interaction code couples the sharp-interface immersed boundary method for fluid dynamics with a finite-element method for structural dynamics. Standard linear solid (SLS) model is used to represent viscoelasticity of plate which is governed by two parameters. First is the ratio of non equilibrium to equilibrium Young modulus (R) and second is the ratio of structural viscosity to non equilibrium Young modulus (ζ). The present study is to examine the dynamic response of the viscoelastic plate by varying the parameter R and ζ for Reynolds number, Re =100. The tip displacement amplitude (A_{Y_tip}) and time to achieve dynamic steady state has been investigated. Tip displacement amplitude is non monotonic function of ζ. When forcing frequency is lesser (greater) than natural frequency of the plate, A_{Y_tip} decreases (increases asymptotically) . The theoretical analysis of simple spring, mass, dashpot model of SLS is done to understand effect of sinusoidal force on the plate dynamics. Numerically computed A_{Y_tip} is observed to be similar trend as the analytical result. |
Monday, November 19, 2018 5:36PM - 5:49PM |
L32.00008: Evaluation of temperature wall functions for large-eddy simulation over curvilinear geometries Randall J McDermott, Marcos Vanella Recently, a novel transport scheme combining a cut-cell finite volume method for scalars and a direct-forcing immersed boundary method for momentum has been implemented in the open-source, low-Mach LES solver called the Fire Dynamics Simulator (FDS). This talk discusses the implementation and testing of heat transfer wall functions over curvilinear and non-grid-aligned geometries. Nusselt number correlations are compared for forced and natural convection. |
Monday, November 19, 2018 5:49PM - 6:02PM |
L32.00009: An Immersed Boundary Projection Method for Dynamics of Rigid Bodies Interacting with Fluid Flows Hsieh-Chen Tsai The fast immersed boundary projection method using a nullspace approach and multi-domain far-field boundary conditions developed by Colonius and Taira (Comput. Methods Appl. Engrg., 2008) is further extended to simulate dynamics of rigid bodies interacting with fluid flows. Dynamics of rigid bodies is characterized by the velocity of the center of mass and the angular velocity about the center of mass, which are governed by the translational and rotational equations of motion involving boundary forces. By introducing the summation and distribution operators, equations of motion are coupled with the incompressible Navier-Stokes equations through the no-slip constraint and boundary forces. Boundary forces are regarded as Lagrange multipliers that enable the no-slip constraint to be implicitly determined to arbitrary precision without associated time-step restrictions. Through projections, the current method removes not only slip component of the velocity field but also components of rigid-body dynamics that do not satisfy equations of motion. Results from simulating an impulsively-rotated circular shell in a quiescent fluid are in good agreement with numerical solutions of the analytical model. |
Monday, November 19, 2018 6:02PM - 6:15PM |
L32.00010: Moving geometry in compressible flows using Cartesian grid and high order Discontinuous Galerkin Method Neda Ebrahimi Pour, Sabine Roller In this contribution we present how the interactions between moving geometry and compressible flows can be realized using high order method on Cartesian grids. To maintain the accuracy of the boundary conditions on geometry equivalent to the scheme order, the geometry has to be represented accordingly. Therefore we consider an immersed boundary method, representing the geometry as a porous material with artificial porosity and permeability coefficients as functions in space and time, known as Brinkman penalization[1]. The surface is obtained by projecting the surface function to a representing polynomial on the computational grid, resulting in the same accuracy as the underlying scheme. Our simulations are carried out using our in-house CFD Framework called APES[2]. REFERENCE [1]Q. Liu, O.V. Vasilyev, A Brinkman method for compressible flows in complex geometries, In Journal of Computational Physics 227, Elsevier Inc. 2007 [2]S. Roller, J. Bernsdorf, H. Klimach, M. Hasert, D. Harlacher, M. Cakircali, S. Zimny, K. Masilamani, L. Didinger, and J. Zudrop. An adaptable simulation framework based on a linearized octree. In M. Resch, X. Wang, W. Bez, E. Focht, H. Kobayashi, and S. Roller, editors, High Performance Computing on Vector Systems 2011, Springer Berlin Heidelberg, 2012 |
Monday, November 19, 2018 6:15PM - 6:28PM |
L32.00011: A Method to Derive Fluid-Structure Interaction Reduced-Order Models for Biomedical Applications Suyue Han, Yahya Modarres-Sadeghi We discuss a method to derive Fluid-Structure Interaction (FSI) Reduced-Order Models (ROMs) for biomedical applications. We use Proper Orthogonal Decomposition (POD) method, which is widely used in deriving ROMs based on results from Computational Fluid Dynamic (CFD), combined with the Immersed Boundary (IB) method. The IB method enhances the POD method to handle large-amplitude displacements without changing the mesh connectivity, thus consuming much less time compared with the topology-changing CFD methods. Another major advantage of the IB method is that it could directly be combined with the Snapshot-POD method to create FSI ROMs. We first conduct training CFD simulations with pre-defined structural motion using the IB method, and then use the snapshot POD method to generate POD modes of the flow field. Then we couple these POD modes with structural modes to create FSI ROMs. |
Monday, November 19, 2018 6:28PM - 6:41PM |
L32.00012: A Parallel dynamic overset grid framework for CFD applications Mohammadali Hedayat, Iman Borazjani The overset grid technique enables the flow solvers to handle complex geometries that can’t be represented well using a single grid, or unsteady moving grid simulations without dealing with remeshing. A new scalable parallel grid assembly and interpolation framework is developed and integrated with a sharp interface curvilinear immersed boundary (CURVIB) flow solver to handle multiple overlapping domains and structures in the fluid domain. In order to achieve a good parallel scalability several steps are implemented in our framework: 1) Oriented bounding boxes (OBB) are constructed instead of axis-aligned bounding boxes; 2) Gradient search for identifying the donor; 3) Removing the I/O file for communication with flow solver; 4) Efficient vectorized implementation for velocity interpolation; and 5) Using general non-inertial frame of reference to prevent the recomputation of curvilinear grid metrics. This framework is validated against several test cases to insure the accuracy of the solvers. Our results show a good scalability and accuracy for this new framework. |
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