Numerical Drag and Lift Prediction Framework for Superellipsoidal Particles in Multiphase Flows

The numerical treatment of industrial and environmental problems, involving multiphase flows with particles, has gained significant interest of researchers over the recent years. For large-scale problems, involving an increased number of particles, authors mostly rely on the Lagrangian particle tracking approach, where particle-fluid interaction is generally unresolved and has to be modelled. Significant research efforts have already been made in developing various models to predict particle-fluid interac- tion, where applications involving complex particle shapes are especially intriguing. In the majority of encountered problems, particle dynamics is primarily governed by drag forces exerted on the particle by the carrier fluid. Following from that it is unsurprising that precise particle trajectories can only be established from accurate particle drag prediction model. In this context, we present a steady-state particle-resolved numerical model, based on OpenFOAM, for numerical drag prediction of superel- lipsoidal particles in Stokesian flow regime. The idea behind particle-resolved model is to benefit from multi parameter drag prediction, which considers not only the flow regime and particle size but also detailed geometric features (expressed by four independent parameters) and particle orientation. The proposed numerical model will also benefit from a parametric geometry formulation, which will allow to evaluate the drag force for the entire range of particle shapes, offered by the superellipsoidal param- etrization. For a vast amount of non-spherical particles, this significantly improves the accuracy of the predicted drag force in comparison to traditional drag models, which do not account for the entire range of influencing factors. The numerical model is further supported by an automated parametric mesh generation algorithm, which makes it possible to autonomously address the full range of particle orientations in parallel. The parametric algorithm also enables the specification of various flow regimes, which are captured in the analysis. Thus, with a single set of input parameters, one can quickly obtain the drag function for given particle shape, with respect to the entire range of orientations and flow regimes. The authors believe that the proposed solution will significantly reduce the effort to obtain an accurate drag model for a vast amount of non-spherical particle shapes.