Optimization of particle–bubble collision dynamics in turbulence via clustering algorithms and microscale vortex enrichment analysis

Abstract

The interaction dynamics between particles and bubbles in turbulent flow fields are crucial for optimizing multiphase flow systems. In this work, direct numerical simulation is combined with advanced K-means++ clustering algorithms to quantify the spatial distribution and enrichment effects of particle–bubble clusters under different turbulence conditions. The results indicate that the Stokes number increases with particle and bubble size, demonstrating stronger inertial effects, but decreases with higher turbulence intensity. Radial relative velocity and collision frequency also exhibit a positive correlation with size and turbulence intensity. Clustering analysis reveals that larger particles and bubbles form more pronounced clusters, particularly in high turbulence conditions, leading to higher local densities and interaction frequencies. Overlap ratios suggest increased interactions with growing size and turbulence intensity. These findings highlight the importance of optimizing particle and bubble sizes to match specific turbulence conditions, enhancing interaction dynamics in multiphase flow systems. This research provides valuable insights for improving various industrial processes involving particle–bubble interactions.