CFD Simulation of Particle-Laden Flow in a 3D Differentially Heated Cavity Using Coarse Large Eddy Simulation

Area of Science:

• Fluid Dynamics
• Aerosol Science
• Computational Engineering

Background:

• Particulate flow in enclosed spaces is crucial for many engineering applications.
• Accurate modeling of particle removal is essential for optimizing these processes.
• Previous studies often require high-resolution meshes, increasing computational expense.

Purpose of the Study:

• To investigate the prediction of particle removal in a thermally driven 3D cavity using Coarse Large Eddy Simulation (CLES).
• To assess the impact of subgrid-scale (SGS) motions on particulate flow predictions with coarse mesh resolution.
• To evaluate the feasibility of modeling complex particulate flows with reduced computational cost.

Main Methods:

• Utilized Coarse Large Eddy Simulation (CLES) for a turbulent Rayleigh number of 10^9.
• Employed an Euler/Lagrange framework to track SiO2 aerosol particles (1.4–14 µm).
• Validated carrier fluid flow against well-resolved LES and experimental data.

Main Results:

• CLES showed good agreement for first-moment Eulerian statistics, with underprediction in higher moments due to coarse resolution.
• Particles with low inertia (<3.5 µm) were more influenced by SGS effects.
• Larger particles (3.5–14 µm) showed less grid dependency.
• Lagrangian statistics for particle depletion agreed well with reference data.

Conclusions:

• CLES is a feasible approach for modeling particulate flow in buoyancy-driven cavities with coarse meshes.
• Reduced mesh resolution significantly impacts the prediction of smaller, less inertial particles.
• This method offers a computationally efficient alternative for simulating complex particulate flows.