GPU accelerated study of a dual-frequency driven single bubble in a 6-dimensional parameter space: The active...

Area of Science:

• Acoustics
• Fluid Dynamics
• Computational Physics

Background:

• Active cavitation is crucial in various applications, requiring precise control over bubble dynamics.
• Understanding the active cavitation threshold is key to optimizing processes involving bubble oscillations.

Purpose of the Study:

• To numerically investigate the active cavitation threshold of single spherical gas bubbles driven by dual frequencies.
• To determine the optimal frequency combinations and their dependence on bubble size and expansion levels.

Main Methods:

• Utilized the Keller-Miksis equation, a second-order ordinary differential equation, to model bubble dynamics.
• Employed a high-performance computing approach using C++ and CUDA C to explore a vast 6-dimensional parameter space (approx. 2 billion combinations).

Main Results:

• For a relative expansion of 2, optimal frequency pairs varied with bubble size: low single frequencies for small bubbles (<3μm), a mix of low and resonance frequencies for medium bubbles (3-6μm), and resonance frequencies for large bubbles (>6μm).
• For a higher relative expansion of 3, single low-frequency driving (20kHz) consistently dominated the active cavitation threshold, indicating the prevalence of the giant response.
• Phase shift between dual frequencies had no impact on the cavitation threshold.

Conclusions:

• Bubble size significantly influences the optimal dual-frequency driving strategy for achieving active cavitation.
• The giant response becomes increasingly dominant at higher expansion levels, favoring single low-frequency excitation.
• Computational fluid dynamics and GPU acceleration are essential for exploring complex parameter spaces in bubble dynamics research.