Effect of two-phase turbulent modalities and bubble-induced turbulence on polydispersed bubbly flow in continuous casting mold

A crucial challenge for gas-liquid two-phase flow is modeling two-phase turbulent modalities and bubble-induced turbulence (BIT). Although many shear-induced turbulence (SIT) models have been established and widely used in single-phase flow, there is no consensus on extending single-phase flow to gas-liquid flow. This work presents the development of the novel population balance model (PBM) for simulating multiphase flows in continuous casting (CC) mold. The effect of two-phase turbulent modalities, SIT and BIT mechanisms on flow pattern, bubble distribution and bubble diameter were studied and against with the experimental data. The results show that the RNG model shows better agreement than other models for predicting flow patterns and bubble size. Compared with the experimental values, the mean relative error of Sauter mean diameter is 5.74 %. Furthermore, the correlation between three turbulent modalities and the SIT has been revealed. The turbulence properties predicted by dispersed-modality and per-modality are highly consistent, and the mixture-modality extremely overestimates the bubble size. The dispersed-modality is suitable for simulating gas-liquid flow for CC mold. Finally, we reveal the coupling mechanism between BIT and SIT models. The Simonin model is insufficient to describe the BIT effect due to the low momentum exchange. The Sato model only caused slight perturbation for the Standard model and significantly increased the turbulence viscosity predicted by RNG model. The coupling between the RNG and Sato models can achieve a circulation feedback effect.