Predictive model for bubble size distribution in a vertical pipe

Abstract

The development of the MUSIG (MUltiple SIze Group) method for modeling two-fluid flows has enabled us to take into account bubble break-up and coalescence, providing a better approximation of bubble interactions and turbulence in a flow. Nonetheless, the development of this model comes with long computation times, making it complicated for industrial use. In this article, we propose a new predictive approach, which leverages the MUSIG population balance but at a much lower computational cost, through a 1D MUSIG model, to calculate bubble size distributions in pipe flow. Our approach is validated by comparing with both the full iMUSIG simulations from Liao et al. (2015) and experimental measurements. We show that it is possible to deduce a relationship between Sauter diameter and flow parameters that enhances the Euler–Euler model’s industrial applicability by accounting for bubble interactions without fully resorting to expensive multi-size-group simulations.