Lattice Boltzmann Method for 3-D Flows with Curved Boundary

Abstract

In this work, we investigate two issues that are important to computational efficiency and reliability in fluid dynamics applications of the lattice Boltzmann equation (LBE): (1) Computational stability and accuracy of different lattice Boltzmann models and (2) the treatment of the boundary conditions on curved solid boundaries and their 3-D implementations. Three a thermal 3-D LBE models (D3Q15, D3Q27) are studied and compared in terms of efficiency, accuracy, and robustness. The boundary treatment recently developed by Filippova and Hanel and Mei et al. in 2-D is extended to and implemented for 3-D. The convergence, stability, and computational efficiency of the 3-D LBE models with the boundary treatment for curved boundaries were tested in simulations of four 3-D flows: (1) Fully developed flows in a square duct, (2) flow in a 3-D lid-driven cavity, (3) fully developed flows in a circular pipe, and (4) a uniform flow over a sphere. We found that while the fifteen-velocity 3-D (D3Q15) model is more prone to numerical instability and the D3Q27 is more computationally intensive, the D3Q19 model provides a balance between computational reliability and efficiency. Through numerical simulations, we demonstrated that the boundary treatment for 3-D arbitrary curved geometry has second-order accuracy and possesses satisfactory stability characteristics.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2002
Accession Number
ADA404858

Entities

People

  • Dazhi Yu
  • Renwei Mei
  • Wei Shyy

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Boltzmann Equation
  • Boundaries
  • Computational Fluid Dynamics
  • Distribution Functions
  • Engineering
  • Equations
  • Fluid Dynamics
  • Geometry
  • Instability
  • Mechanics
  • Navier Stokes Equations
  • Reliability
  • Reynolds Number
  • Simulations
  • Three Dimensional
  • Two Dimensional

Readers

  • Artificial Intelligence
  • Computational Fluid Dynamics (CFD)
  • Fluid Mechanics and Fluid Dynamics.