Investigation of bubble transport in a turbulent recurrent flow

The high number of degrees of freedom present in turbulent flows implies extremely expensive calculations making detailed, long-term and large-scale studies unfeasible. Therefore, the challenge of maintaining a reasonable accuracy and reducing the computational cost still exists. “Recurrence CFD” (rCFD) is a data-assisted approach to time-extrapolate the dynamics of recurrent systems. In other words, based on the temporal life-cycle characteristic of dominant flow structures obtained from a short-time detailed simulation, we study the long-time evolution of a system. Along with this time-extrapolation, we could study long-term transport processes according to the underlying flow field by solving only the passive transport equation, which renders rCFD an excellent and reliable method to study large-scale systems.

Here, we evaluate the method’s capability for a submerged double-jet coupled with argon bubble injection as an example of weakly coupled, Lagrangian transport. We study and develop rCFD to reduce the database size in turbulent flows. Our rCFD results are comparable with the full computational fluid dynamics–discrete element method (CFD-DEM) simulations by speed-up factors of more than 35.