A supervised neural network for drag prediction of arbitrary 2D shapes in laminar flows at low Reynolds number

被引:38
|
作者
Viquerat, Jonathan [1 ]
Hachem, Elie [1 ]
机构
[1] PSL Res Univ, MINES Paristech, CEMEF, Paris, France
关键词
Machine learning; Neural networks; Convolutional networks; Computational fluid dynamics; Immersed method;
D O I
10.1016/j.compfluid.2020.104645
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Despite the significant breakthrough of neural networks in the last few years, their spreading in the field of computational fluid dynamics is very recent, and many applications remain to explore. In this paper, we explore the drag prediction capabilities of convolutional neural networks for laminar, low-Reynolds number flows past arbitrary 2D shapes. A set of random shapes exhibiting a rich variety of geometrical features is built using Bezier curves. The efficient labelling of the shapes is provided using an immersed method to solve a unified Eulerian formulation of the Navier-Stokes equation. The network is then trained and optimized on the obtained dataset, and its predictive efficiency assessed on several real-life shapes, including NACA airfoils. (C) 2020 Elsevier Ltd. All rights reserved.
引用
收藏
页数:11
相关论文
共 36 条
  • [1] A hybrid asymptotic-numerical method for calculating drag coefficients in 2-D low Reynolds number flows
    Sarah Hormozi
    Michael J. Ward
    Journal of Engineering Mathematics, 2017, 102 : 3 - 33
  • [2] A hybrid asymptotic-numerical method for calculating drag coefficients in 2-D low Reynolds number flows
    Hormozi, Sarah
    Ward, Michael J.
    JOURNAL OF ENGINEERING MATHEMATICS, 2017, 102 (01) : 3 - 33
  • [3] An Assessment of the Laminar Kinetic Energy Concept for the Prediction of High-Lift, Low-Reynolds Number Cascade Flows
    Pacciani, R.
    Marconcini, M.
    Arnone, A.
    Bertini, F.
    9TH EUROPEAN CONFERENCE ON TURBOMACHINERY: FLUID DYNAMICS AND THERMODYNAMICS, VOLS I AND II, 2011, : 297 - 307
  • [4] An assessment of the laminar kinetic energy concept for the prediction of high-lift, low-Reynolds number cascade flows
    Pacciani, R.
    Marconcini, M.
    Arnone, A.
    Bertini, F.
    PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART A-JOURNAL OF POWER AND ENERGY, 2011, 225 (A7) : 995 - 1003
  • [5] Modeling 2D unsteady flows at moderate Reynolds numbers using a 3D convolutional neural network and a mixture of experts
    Zigon, Bob
    Zhu, Luoding
    COMPUTER PHYSICS COMMUNICATIONS, 2025, 310
  • [6] A multipole direct and indirect BEM for 2D cavity flow at low Reynolds number
    Gomez, JE
    Power, H
    ENGINEERING ANALYSIS WITH BOUNDARY ELEMENTS, 1997, 19 (01) : 17 - 31
  • [7] Automatic landmarking of 2D medical shapes using the growing neural gas network
    Angelopoulou, A
    Psarrou, A
    Rodríguez, JG
    Revett, K
    COMPUTER VISION FOR BIOMEDICAL IMAGE APPLICATIONS, PROCEEDINGS, 2005, 3765 : 210 - 219
  • [8] A STUDY ON CHAOTIC MIXING IN 2D CAVITY FLOWS (EFFECTS OF REYNOLDS-NUMBER AND AMPLITUDE OF LID VELOCITY)
    TAKASAKI, S
    OGAWARA, K
    IIDA, S
    JSME INTERNATIONAL JOURNAL SERIES B-FLUIDS AND THERMAL ENGINEERING, 1994, 37 (02) : 237 - 241
  • [9] Convolutional Neural Network Model Based on 2D Fingerprint for Bioactivity Prediction
    Hentabli, Hamza
    Bengherbia, Billel
    Saeed, Faisal
    Salim, Naomie
    Nafea, Ibtehal
    Toubal, Abdelmoughni
    Nasser, Maged
    INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES, 2022, 23 (21)
  • [10] High and Low Reynolds Number Two-phase Flows over a Backward-facing Step by 2D and 3D Large Eddy Simulation
    Yu, K. F.
    Lee, Eric W. M.
    Yuen, Jason K. K.
    INTERNATIONAL JOURNAL OF NONLINEAR SCIENCES AND NUMERICAL SIMULATION, 2009, 10 (09) : 1135 - 1157