Experimental study on multi-scale characteristics of cavitating flows with holographic imaging measurement

Cavitating flows are characterized by multi-phase and multi-scale features, with evolutionary processes involving coupled interactions between the convection evolution of macroscale vapor structures and the growth and motion of microbubbles. The quantitative information and intrinsic physical mechanism are poorly understood, due to limitations of traditional methods in quantitatively measuring the three-dimensional distribution of microbubbles within cavity structures. In the present work, an experimental study integrating high-speed imaging of macroscale cavity convection evolution and quantitative digital in-line holography (DIH) measurement of microbubbles is conducted to investigate multiscale characteristics of cavitating flows. Results demonstrate that cavitation morphology progresses through inception, sheet, and cloud stages with decreasing cavitation numbers, accompanied by gradual increases in maximum attached cavity length and significant growth in discrete bubble quantities. Mesoscale bubbles are predominantly distributed at vapor-liquid interfaces of macroscale cavities, surrounding shedding cloud cavities, and within wake regions of turbulent cavitating flows. Meanwhile, the Sauter mean diameter of microbubbles progressively decreases along the streamwise direction. As the cavitation number decreases, within the cavity-shedding region, shed cavities gradually manifest as large scale cavities, the time-averaged number density of discrete microbubbles first increases and then paradoxically decreases. In contrast, within the wake flow region, shed cavities undergo complete fragmentation into discrete bubbles, resulting in a persistent increase in detectable mesoscale discrete bubbles with decreasing cavitation number. Across all cavitation regimes and the holographic measurement zone, the number of discrete bubbles initially increased then decreased with increasing bubble diameter, with spectral peaks in bubble size distribution (BSD) at 30-40 μm. Turbulent flow structures significantly affect bubble dynamic evolution. Consequently, dual power-law scaling governs the microbubble size distribution, relative to the Hinze scale at approximately 55–65 μm. Sub-Hinze-scale bubbles follow a − 4/3 scaling exponent, whereas super-Hinze-scale bubbles obey a − 10/3 scaling law.