Lattice boltzmann method on composite grids

Area of Science:

• Computational Fluid Dynamics
• Numerical Methods
• Fluid Mechanics

Background:

• Simulating incompressible fluid flows requires efficient numerical methods.
• Local grid refinement is crucial for capturing complex flow features accurately.
• Existing methods may face challenges in balancing accuracy and computational cost.

Purpose of the Study:

• To propose a composite block-structured lattice Boltzmann method (LBM) for 2D incompressible fluid flow simulations.
• To enable efficient local grid refinement within the LBM framework.
• To reduce computational cost while maintaining high accuracy.

Main Methods:

• Developed a composite grid structure with a coarse base grid and one or more fine grids.
• Implemented a two-stage simulation process: initial coarse grid run followed by simultaneous multi-grid advancement.
• Utilized base grid solutions to provide boundary conditions for fine grids at interfaces.

Main Results:

• The method successfully simulated the lid-driven cavity flow.
• Results showed excellent agreement with benchmark data and finite-volume Reynolds-averaged Navier–Stokes (RANS) model.
• Demonstrated significant savings in CPU time compared to traditional methods.

Conclusions:

• The composite block-structured LBM is an effective method for accurate 2D incompressible flow simulation.
• The approach offers substantial computational efficiency through local grid refinement.
• This method provides a viable alternative for complex fluid dynamics problems.