Development of an interface resolving numerical method for a large number of Kolmogorov scale sized bubbles interacting with homogeneous isotropic turbulence

• Applicant: Professor Dr.-Ing. Wolfgang Schröder
• Subject Area: Fluid Mechanics
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 536555656

The interaction of a large number of Kolmogorov scale sized deformable bubbles in turbulent flow plays an important role in numerous technical applications in chemical or process engineering. For bubble laden flows, it can be stated that depending on the ratio of bubble size and the minimum characteristic length scale of the turbulent carrier fluid, i.e., the Kolmogorov scale, different numerical methods exist. If the bubble size is large compared to the turbulent length scale, interface resolving methods are used. On the other hand, reduced-order models are state of the art if a large number of small bubbles is considered. In this case, Lagrangian models which are based on simplifying assumptions are used. To determine the limits of those models being caused by simplifying the impact of surface tension and bubble shape variation on the turbulent structures in the carrier fluid, scale resolving gas-liquid interface computations are necessary, which require a highly accurate and efficient numerical method. Therefore, new direct interface resolving numerical methods (Direct Bubble-Fluid Simulation (DBFS)) and parallelization strategies are developed that allow the analysis of the interaction of thousands Kolmogorov scale sized bubbles with turbulent flow. A large number of bubbles has to be considered due to the necessary statistical quantification of the simulation results. Two Lattice Boltzmann solvers are coupled to model the liquid and the gas phase. The interface is captured using a level-set approach, where the geometrical information is used to impose the coupling condition between both flow solvers, including the effect of surface tension. To keep the required number of grid cells in an acceptable range and to use the available high performance computing hardware as efficiently as possible, adaptive mesh refinement and dynamic load balancing techniques are exploited. After validating this novel approach by benchmark cases from the literature, the canonical case of isotropic turbulence is considered to gain fundamental understanding of the multiphase interaction of small deformable bubbles and to determine the impact of surface tension and bubble deformation on the near-field turbulence of the bubbles. The data serve as a ground truth for the evaluation of reduced-order models. The detailed comparison with standard Lagrangian model solutions will show the limits of the low-fidelity approaches and clarify the physical mechanisms that determine these limits.

• DFG Programme: Research Grants
• Co-Investigator: Dr.-Ing. Matthias Meinke