Kinetic-Based Multiphase Flow Simulation.

Area of Science:

• Computational fluid dynamics
• Computer graphics
• Multiphase flow simulation

Background:

• Existing fluid solvers excel at individual effects (bubbling, wetting) but lack generality.
• Computational fluid dynamics offers general models but struggles with high density ratios and Reynolds numbers for realistic simulations.
• A gap exists in methods capable of simulating complex multiphase phenomena concurrently for applications like film production.

Purpose of the Study:

• To develop a general multiphase flow solver capable of simulating diverse behaviors simultaneously.
• To improve upon existing kinetic-based approaches for enhanced stability and accuracy.
• To address limitations in simulating high density ratios and Reynolds numbers required for realistic visual effects.

Main Methods:

• Proposed a kinetic model coupling Navier-Stokes equations with a conservative phase-field equation.
• Implemented numerical improvements over existing kinetic-based fluid solvers.
• Developed an embarrassingly parallel and conservative algorithm.

Main Results:

• The new solver demonstrates superior stability and generality for simulating multiphase flows.
• Successfully captures complex behaviors like bubbling, glugging, wetting, and splashing concurrently.
• Achieved accurate simulations comparable to real footage and previous methods, even under challenging real-life conditions.

Conclusions:

• The proposed kinetic model provides a robust and general solution for multiphase flow simulation.
• The numerical improvements enhance stability and efficiency, making it suitable for demanding applications.
• This method advances the state-of-the-art in simulating realistic fluid dynamics for computer graphics and scientific research.