Junior Research Group "Advanced modelling of multiphase flows"
The Junior Research Group is part of the project "Crossing borders and scales - an interdisciplinary approach (CROSSING)", which is funded by the Helmholtz Association of German Research Centres in the frame of the program „Helmholtz European Partnering“. The funding is granted for 3 years (2019 - 2021) and might be extended for 2 years.
Helmholtz-Zentrum Dresden - Rossendorf (HZDR): Computational Fluid Dynamics
The aim is to establish a long-term strategic partnership between HZDR and JSI in Slovenia and to conduct excellent research in the field of numerical simulations of multiphase flows.
The focus of the work is on numerical simulation of condensation-induced water hammers, which cause serious damage in industrial facilities. Despite the enormous progress in modelling of multiphase flows, condensation induced water hammers are a complex problem with many challenges. However, JSI and HZDR have a long expertise in this field and in the frame of the Junior Research Group the knowledge will be brought together.
The work is embedded into the OpenFOAM activities at HZDR.
Qualification of OpenFOAM for turbulent interfaces in two-fluid model
Similarly to modelling of interfacial flows, turbulent eddies are separated into large and small, sub-filter scales. In LES, large scale structures are resolved in space and time, while the small turbulent structures need modelling. In the framework of multiphase flows in a two-fluid model, additionally to the "classical" turbulent stress, a number of sub-filter scale contributions is present, which describe the dynamics and interaction of interfaces and turbulence. For these sub-filter contributions, different models and their combinations are assessed in a-posteriori investigations for application to technically relevant simulations.
Qualification of OpenFOAM for stratified flows with heat and mass transfer
One of our goals is to advance the capabilities of current two-fluid (Euler-Euler) based modelling tools towards simulation of industrially relevant turbulent two-phase flows. Present work is focused on the development, implementation and validation of improved two-phase heat and mass transfer models for stratified flows. Advancement includes improvements concerning the treatment of interaction between gas-liquid interfaces and turbulence.
Develop and apply machine learning methods for population balance modelling
The inhomogeneous class method developed at HZDR and contributed to the OpenFOAM Foundation release allows for modelling of the size distribution of bubbles, droplets or fractal agglomerates. However, solving the population balance equation requires almost 90% of a time step in a numerical simulation. The implementation of a class method on graphical processor units (Cuda) and the application of suitable machine learning methods can lead to a significant reduction in computing time and due to that allow for more challenging and larger multiphase flow simulations.
Dr. Fabian Schlegel, Computational Fluid Dynamics Department, Helmholtz-Zentrum Dresden - Rossendorf
Dr. Matej Tekavčič, Postdoc, Reactor Engineering Division, Jožef Stefan Institute, Slovenia
Richard Meller, PhD, Computational Fluid Dynamics Department, Helmholtz-Zentrum Dresden - Rossendorf
Gašper Petelin, PhD, Computer Systems Department, Jožef Stefan Institute, Slovenia
- Liao, Y.; Upadhyay, K.; Schlegel, F.
Eulerian-Eulerian two-fluid model for laminar bubbly pipe flows: validation of the baseline model
Computers & Fluids 202(2020), 104496 (10.1016/j.compfluid.2020.104496)