Automated estimation of air entrainment in plunging jet flows below onset velocity

Air entrainment due to plunging jet flows has significant applications across various fields, including environmental and industrial processes. In such flows, the onset velocity is the critical jet velocity required to sustain continuous air entrainment. However, even below the onset velocity, the initial jet impact and subsequent bubble re-entrainment can result in considerable air entrainment for a finite duration. This study investigates air entrainment by plunging jets operating below the onset velocity. Experiments were carried out, and images were captured using a high-speed camera with backlighting at different jet velocities up to the onset velocity. One of the primary challenges in analyzing air entrainment by plunging jets is quantifying entrained air, which is difficult due to the random nature of bubble dynamics. Traditional methods do not provide an accurate bubble distribution. This challenge was addressed using a U-Net-based deep learning approach for bubble segmentation. Overlapping bubbles in the segmented images posed a challenge for automatic identification. An algorithm was developed to separate overlapping bubbles and extract accurate bubble distributions. The analysis revealed that bubbles generated by the initial jet impact tend to persist longer due to their entry into recirculation zones and subsequent re-entrainment. Bubble size decreases over time due to breakage during re-entrainment. It was observed that the time required for the gas holdup to reach zero increases with jet velocity. This work presents fundamental observations of air entrainment below the onset velocity.