Publikationsrepositorium - Helmholtz-Zentrum Dresden-Rossendorf

The need to optimise processes, e.g. in terms of energy consumption and product properties, is leading to an increasing demand from industry to be able to reliably predict multiphase flows or even to simulate them as a digital twin in parallel with industrial processes. Because of the large range of scales to be considered, the Euler-Euler approach is often the only feasible framework for medium and large industrial scales. Improving closure models and bringing together different modelling approaches for different scales of phase interfaces are important tasks to achieve a better reliability. For the latter, a flexible hybrid framework for OpenFOAM was developed at the HZDR, which is presented in the talk. Improving closure models requires a better understanding of local phenomena that influence the interactions between phases. Enhanced possibilities for measurements (e.g. with high-speed cameras and ultra-fast X-ray tomography), growing feasibility of DNS for multiphase flows and the use of artificial intelligence open up new possibilities here. Some recent developments and requirements for future activities are discussed.