A VIRTUAL FINITE VOLUME METHOD FOR COMPUTATIONAL FLUID DYNAMICS

被引:0
|
作者
Runchal, Akshai [1 ]
Rajagopal, P. S. [2 ]
Rao, Madhukar M. [2 ]
Addaguduri, Chakrapani [2 ]
机构
[1] Analyt & Computat Res Inc, 1931 Stradella Rd, Los Angeles, CA 90077 USA
[2] ACRi Infotech Pvt Ltd, Dharmshala, India
来源
COMPUTATIONAL THERMAL SCIENCES | 2024年 / 16卷 / 06期
关键词
computational fluid dynamics; machine learning; Voronoi mesh; virtual finite volume; point cloud; polyhedral;
D O I
10.1615/ComputThermalScien.2024053399
中图分类号
O414.1 [热力学];
学科分类号
摘要
The virtual finite volume method (VFVM), based on Voronoi diagrams, is developed, and applied to two benchmark simulations of laminar and turbulent flows. The results of the VFVM are found to be closest to the benchmark for turbulent flow compared to all-quad and triangular meshes. Additionally, the VFVM produced the smallest RMS error for laminar flow on the coarsest mesh in comparison. The resources required to generate a high-quality mesh for computational fluid dynamics (CFD) computations may well exceed those required for CFD computations. Further, mesh-generation is a labor-intensive process that is difficult to automate. Meshless computing is therefore an active area of research. The article describes a finite volume method (FVM) that is mesh-less as far as the user is concerned. The mesh is generated from a cloud of distributed points automatically without any intervention by the CFD practitioner. We call this the virtual finite volume method. A key feature of the VFVM is that the mesh is generated from a point cloud that is converted into Voronoi diagrams. In the context of the FVM, Voronoi polygons (2D) or polyhedra (3D) have the important property that the line connecting two adjacent nodes is perpendicular to the common intervening face and is bisected by it. This orthogonality allows for very efficient computation of diffusion and convection terms and minimizes the numerical diffusion. The ANSWER (R) CFD solver has been extended to generate these Voronoi mesh systems automatically and allow for adaptive gridding to yield a high accuracy solution with optimal computational resources.
引用
收藏
页码:45 / 58
页数:14
相关论文
共 50 条
  • [1] Quantum Approach to Accelerate Finite Volume Method on Steady Computational Fluid Dynamics Problems
    Zhao-Yun Chen
    Cheng Xue
    Si-Ming Chen
    Bing-Han Lu
    Yu-Chun Wu
    Ju-Chun Ding
    Sheng-Hong Huang
    Guo-Ping Guo
    Quantum Information Processing, 21
  • [2] Quantum Approach to Accelerate Finite Volume Method on Steady Computational Fluid Dynamics Problems
    Chen, Zhao-Yun
    Xue, Cheng
    Chen, Si-Ming
    Lu, Bing-Han
    Wu, Yu-Chun
    Ding, Ju-Chun
    Huang, Sheng-Hong
    Guo, Guo-Ping
    QUANTUM INFORMATION PROCESSING, 2022, 21 (04)
  • [3] On combining finite element methods and finite volume methods in computational fluid dynamics
    Pascal, F
    PROGRESS IN ANALYSIS, VOLS I AND II, 2003, : 1205 - 1213
  • [4] Computational fluid dynamics simulation for the prediction of the venturi scrubber performance using finite volume method
    Attaullah
    Niazi, Muhammad Bilal Khan
    Ahsan, Muhammad
    Ali, Majid
    INTERNATIONAL JOURNAL OF COMPUTING SCIENCE AND MATHEMATICS, 2020, 11 (04) : 338 - 346
  • [5] Limitations of traditional finite volume discretizations for unsteady computational fluid dynamics
    Manson, JR
    Pender, G
    Wallis, SG
    AIAA JOURNAL, 1996, 34 (05) : 1074 - 1076
  • [6] A unified-grid finite volume formulation for computational fluid dynamics
    Chakravarthy, S
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS, 1999, 31 (01) : 309 - 323
  • [7] A Massively Parallel Hybrid Finite Volume/Finite Element Scheme for Computational Fluid Dynamics
    Rio-Martin, Laura
    Busto, Saray
    Dumbser, Michael
    MATHEMATICS, 2021, 9 (18)
  • [8] Computational fluid dynamics (CFD) modeling of heat transfer in a polymeric membrane using finite volume method
    Muhammad Ahsan
    Arshad Hussain
    Journal of Thermal Science, 2016, 25 : 564 - 570
  • [9] Experimental methods in chemical engineering: Computational fluid dynamics/finite volume method-CFD/FVM
    Van Hoecke, Laurens
    Boeye, Dieter
    Gonzalez-Quiroga, Arturo
    Patience, Gregory S.
    Perreault, Patrice
    CANADIAN JOURNAL OF CHEMICAL ENGINEERING, 2023, 101 (02): : 545 - 561
  • [10] Computational Fluid Dynamics(CFD) Modeling of Heat Transfer in a Polymeric Membrane using Finite Volume Method
    Muhammad Ahsan
    Arshad Hussain
    JournalofThermalScience, 2016, 25 (06) : 564 - 570