A quantitative evaluation of the prediction performance of a one-dimensional multifluid population balance model in continuous and semi-batch bubble columns

Abstract

The objective of this study was to evaluate the efficacy of a one-dimensional multifluid population balance model (1D-MPB) in predicting the axial variation of the bubble size distribution (BSD), gas phase velocity, and gas volume fraction in a bubble column. The bubble column was operated in either a semi-batch mode, without liquid feed, or a continuous mode, with co-current liquid and gas feed. The model’s predictions were compared with experimental data obtained through minimal invasive fiber optic needle probes. The experiments were carried out with different spargers and varying gas and liquid fluxes. The model parameters were previously determined in a different study. The findings indicate that the model accurately predicts the BSD, Sauter mean diameter and local gas volume fractions, particularly at medium gas fluxes in semi-batch mode. The application of a modified sparger model to predict inlet BSDs demonstrated potential, although limitations were observed at higher fluxes. In continuous co-current mode, the model (parameterized with data from semi-batch experiments), exhibited robust prediction performance without the need for recalibration, indicating the soundness of the underlying physical principles and the value of the method for scale-up. The mean absolute percentage error for the Sauter mean diameter and for the local gas volume fraction was about 10 %, depending on the mode of operation, type of sparger, fluid inlet flux and source of the inlet BSD (experimental or from the sparger model), and in many cases significantly less. Future work will focus on refining the turbulent energy dissipation model to enhance the model’s accuracy and applicability for industrial bubble column design and optimization.