Large-scale Simulations in Complex Flows

NAISS 2024/1-13


NAISS Large Compute

Principal Investigator:

Outi Tammisola


Kungliga Tekniska högskolan

Start Date:


End Date:


Primary Classification:

20306: Fluid Mechanics and Acoustics

Secondary Classification:

10105: Computational Mathematics



We summarize here the projects by the research groups working on complex and multiphase flows at FLOW, Department of Engineering Mechanics (a previous VR Excellence Centre) at KTH Engineering Mechanics. In this proposal we list five successful research areas, each with many PhD students and postdocs whose projects rely on HPC resources. Many researchers participate in and contribute to more than one area. We actively promote collaboration within our large user group to facilitate HPC support, sharing of simulation methods and codes, and user experience. 1. Complex fluids and multiphase flows: Non-Newtonian fluids, multiphase flows and flows with a microstructure (particles, fibers, bubbles, droplets in laminar and turbulent regimes). 2. Flow at interfaces: Flows over porous, elastic, poro-elastic and micro-structured interfaces. 3. Bio-physical flows: Flows in the cardiovascular and respiratory systems, and in artificial devices (blood pumps, oxygenators), and cells in microfluidics. 4. Flows with phase change: Flows with phase change, absorption/desorption, buoyancy-induced convection to study bubble/droplet breakup during evaporation and boiling process, and melting of glaciers. 5. Unsteady flows for clean vehicles: Simulations of compressible flows, with heat- and mass-transfer associated with energy conversion and propulsion systems, to reduce pollutants and noise and improve efficiency. In the associated activity report, we show that resources previously allocated could be used efficiently and led to a significant number of relevant scientific publications. The consortium has been awarded a significant number of prestigious grants. Profs.Tammisola, Bagheri and Prahl Wittberg have all been awarded ongoing ERC grants, Bagheri chosen the SSF Future research leader and Wallenberg Academy Fellow in 2017. We are a large part of the environment for multiscale modelling, INTERFACE, funded by VR (24 millions sek). We are leading and participating in several large EU and national projects, i.e. AdThERM (), EU Doctoral Networks YIELDGAP (12 PhDs), SCALE (15 PhDs) and . Projects AdTherm funded by Vinnova (2 PhD students at KTH) and recently MedTechLab granted center project and XX are about to start. In addition, a new assistant professor fluid mechanics has been recruited and will join this allocation in Fall 2024. During the last years we used the allocated time, sometimes largely exceeding it (e.g. 131 % of the allocated resources on Dardel1 and 137 % of the allocated resources on Tetralith). Given this and the developments mentioned above, we apply for a significant amount of time on most systems, in particular on Dardel (5 · 106 core hours) and Tetralith (2 · 106 core hours). In agreement with the recent effort in machine learning computations on GPU-accelerated architectures, we also ask 1500 GPU-h/month on Alvis. The different projects listed below demand large-scale resources. For instance, we are now performing the first ever resolved DNS simulation of three-dimensional elasto-inertial turbulence in a duct and turbulent flow over liquid-infused surfaces.