This paper presents recent progress on prediction of bubbly flows around ships, including bubble entrainment modeling, bubble transport and numerical issues. The bubbly flow is described by a polydisperse two-fluid model that can predict the bubble entrainment locations and rates, bubble dissolution, breakup and coalescence rates, bubble velocities, turbulence quantities and bubble size distribution. To test the performance of the two phase flow model, several simulations are conducted on canonical bubbly flows with wave breaking. These well experimentally studied flows provide important information for the design of the bubble entrainment model, which is the weakest link in the model chain but crucial for prediction of the bubbly wake. The results are compared with experimental data to study the model’s accuracy and to calibrate the entrainment model constants. Full scale simulations for the flat bottom Kann boat and the Athena R/V are performed to evaluate the model under more complex flows of naval relevance that have considerable data available. It is found that the model calibrated with canonical problems predicts good results for Athena R/V, but the current turbulent entrainment model significantly underestimates the entrainment at the bow of Kann boat due to other entrainment mechanisms involved (entrainment due to impact, droplets, etc.). The breakup model, which currently considers turbulent mechanisms, underestimates the population of small bubbles in the boundary layer where strong shear is present. Finally, a grid study is carried on Athena R/V to test grid convergence. Void fraction and size distribution are compared against available experimental data and discussed in detail. Overall, the simulations show encouraging results considering the complexity of two phase flow involved in ship applications, and the model is proven to be grid independent, a very important property for practical applications.

This content is only available via PDF.
You do not currently have access to this content.