Predicting bubble size and bubble rate data in water and in froth flotation-like slurry from computational fluid dynamics (CFD) by applying deep neural networks (DNN)

Accurate characterization of two phase bubbly flows is crucial in many industrial processes such as fluidized reactors, ore froth flotation, etc. The bubble size determines the rate at which components present in the gas phase are transferred to the surroundings and vice versa while bubble rate defines the appropriate bubbly flow regime occurring in the heterogeneous system. This research work employs deep neural networks (DNNs) to predict bubble size and bubble rate using data obtained from validated computational fluid dynamics (CFD) computations. Pure water and slurry (in conditions similar to those employed in mineral froth flotation) case studies are evaluated. It is found that the DNN can predict the CFD results accurately when using four hidden layers, describing discontinuities in the bubbly flow regime. The relative errors computed between the CFD data and the prediction obtained by the DNN is as low as 8.8% and 1.8% for bubble size and bubble rate, respectively. These results confirm that the DNN can be applied to sophisticated fluid dynamics systems and allow developing better control process strategies since once the DNN is trained critical variables can be computed very efficiently. The slurry case study, although restricted to the application of mineral froth flotation, can also be generalized to other industrial operations keeping the exact same procedure.