Bulletin of the American Physical Society
73rd Annual Meeting of the APS Division of Fluid Dynamics
Volume 65, Number 13
Sunday–Tuesday, November 22–24, 2020; Virtual, CT (Chicago time)
Session R09: Computational Fluid Dynamics: SPH and Mesh Free Methods (5:00pm - 5:45pm CST)Interactive On Demand
|
Hide Abstracts |
|
R09.00001: SPH simulations of helicopter ditching on calm water and in waves. Guillaume Oger, Alban Vergnaud, Benjamin Bouscasse, Séverin Halbout During helicopter ditching events, an Emergency Floatation System (EFS) is deployed prior to the impact so that to improve the occupants' chances of survival by keeping the helicopter afloat for a sufficiently long duration. However, predicting the helicopter behavior (equipped with its EFS) during the impact is not straightforward, especially in presence of waves. Thanks to the European H2020 project SARAH, experiments were performed in the wave tank of Ecole Centrale Nantes in partnership with Airbus Helicopters, providing reference results for comparison with CFD solvers. During this experimental campaign, a large set of helicopter impact cases with and without waves were tested. The Smoothed Particle Hydrodynamics (SPH) method appeared as a good candidate for such numerical simulations, due to its meshless and Lagrangian features together with its ability to deal with complex geometries in interaction with strong free surface deformations. In the present study, simulations are performed for various impact configurations and different wave conditions. The numerical solutions are systematically compared with the experimental results, especially regarding the trajectories, impact forces and local pressures. [Preview Abstract] |
|
R09.00002: Graph Neural Network for Lagrangian Fluid Simulation Zijie Li, Amir Barati Farimani Despite the advances in computing power, computing high-quality fluid simulation is still computationally expensive. Meanwhile, data-driven model serves as an attractive alternative. In this work, we use graph to describe fluid field under Lagrangian system and build a neural network model upon graph representation, where physical quantities are encoded as node and edge features. Instead of directly predicting the acceleration or position correction given current state, we decompose the simulation scheme into separate parts-advection, collision and pressure projection. With different sub-networks each responsible for a specific reasoning task, the learned model is able to give reasonable prediction and remain stable in long-term simulation. The network is build upon simple graph aggregation and standard multi-layer perceptron, but we show that it can accurately learn and simulate the underlying complex fluid dynamics based on observations. Our tests demonstrates that, first, it can remain stable and be extrapolated to situation with different geometries and conditions. Second, the simplicity of the model enables its fast inference during simulation. Third, the learned model is able to maintain low velocity divergence and generate reasonable pressure distribution. [Preview Abstract] |
|
R09.00003: Turbulence Modeling in Smoothed Particle Hydrodynamics Francesco Ricci, Renato Vacondio, Angelantonio Tafuni The continuous growth of computational power in Computational Fluid Dynamics (CFD) has made it possible to study flow at high Reynolds numbers, which demand reliable models for the simulation of turbulence. For Eulerian methods, this has led to a shifting from Reynolds Averaged Navier-Stokes (RANS) models to Large Eddy Simulation (LES), especially for industrial applications. Among Lagrangian approaches is Smoothed Particle Hydrodynamics (SPH), a meshless method often used to study free-surface flow in several fluids engineering problems. In the present work, the standard SPH method is compared with an Eulerian SPH model, the latter being a modification of the standard SPH approach in which the position of the particles is kept fixed and additional convective terms are added to the governing equations. This has allowed the identification of the main source of error for the standard SPH approach when simulating isotropic turbulence decay, i.e. the discretization error due to the irregular distribution of SPH particles. Such error leads to an inaccurate description of the decay of the kinetic energy and the turbulent structures in the flow. Different strategies to address these issues are then proposed and described. [Preview Abstract] |
|
R09.00004: DualSPHysics: from fluid dynamics to multiphysics problems Angelantonio Tafuni, Jose Dominguez, Georgios Fourtakas, Corrado Altomare, Ricardo Canelas, Orlando García-Feal, Iván Martínez-Estévez, Athanasios Mokos, Renato Vacondio, Alejandro Crespo, Benedict Rogers, Peter Stansby, Moncho Gómez-Gesteira DualSPHysics is a smoothed particle hydrodynamics (SPH) Navier-Stokes solver initially developed for coastal engineering problems. Since its first release in 2011, the code has undergone continuous improvements in performance thanks to the use of latest hardware technologies, but also thanks to the coupling with wave propagating models such as SWASH and OceanWave3D. Numerous functionalities have been included over the last few years to simulate fluid driven objects. The use of the discrete element method (DEM) has enabled simulating the interactions among different bodies (e.g., sliding rocks), providing a unique tool to analyze debris flows. In addition, the recent coupling with other solvers like Project Chrono and MoorDyn has been a milestone in the development of the solver. DualSPHysics includes a multi-phase solver for simulations with gas-liquid and a combination of Newtonian and non-Newtonian models, further expanding the capabilities and range of applications. These advancements and functionalities make DualSPHysics a state-of-the-art meshless solver with emphasis on free-surface flows modeling. [Preview Abstract] |
|
R09.00005: Coupling the Finite Volume Particle Method with the Finite Element Method for fluid-structure interaction for large deformations Maryrose McLoone, Nathan Quinlan The finite volume particle method (FVPM) is a meshless CFD method. FVPM is advantageous for fluid-structure interaction (FSI) as walls are defined as exact geometry, therefore fictitious particles are not required for wall modelling, saving on computational cost. To date, FVPM has not been coupled with an external solid solver for two-way FSI. FVPM is here coupled with a mesh-based finite element solid mechanics solver, FEBio (Maas et al., 2012), for 2D FSI. This allows for the modelling of highly deformable solids. The FSI method is applied to the elastic dambreak case of Antoci et al. (2007). The method is validated by comparing the gate displacement against the experiment. The FSI-computed gate displacement agrees well with the experimental data. Antoci, C, Gallati, M, {\&} Sibilla, S, (2007). Numerical simulation of fluid--structure interaction by SPH. Computers {\&} Structures, 85(11-14), 879-890. Maas, S. A, Ellis, B. J, Ateshian, G. A, {\&} Weiss, J. A, (2012). FEBio: finite elements for biomechanics. Journal of biomechanical engineering, 134(1). [Preview Abstract] |
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