Novel method for measuring interface behavior and flow parameters in bubble-particle detachment using PIV-LIF and machine learning image segmentation

Turbulence is a critical factor inducing bubble-particle detachment, and investigating this mechanism is crucial for improving the flotation recovery of coarse particles. However, current research is constrained by the lack of synchronous measurement methods, causing most studies to analyze interface behavior and flow parameters separately, thereby limiting comprehensive insights into detachment processes. This study presents a novel method combining particle image velocimetry (PIV) and laser-induced fluorescence (LIF) with machine learningbased image segmentation to investigate detachment mechanisms in shear flow fields. The results demonstrate that this method enables real-time measurement of bubble-particle detachment while simultaneously capturing interface behavior and flow parameters. Furthermore, the study elucidates the shear detachment mechanism: lateral vortices dominate bubble deflection, whereas forward shear flow plays a crucial role in bubble detachment. This work provides a new method for bubble-particle detachment research and advances the understanding of turbulence-induced detachment mechanisms.