HPC Cluster
High-performance computer for wind energy research
Computing power for the research of the future
Numerical simulations, along with free-field measurements and laboratory experiments, form an important pillar in current research questions to achieve a comprehensive understanding of turbulent wind conditions and their interaction with wind energy systems. The high-performance computers at the Carl von Ossietzky University of Oldenburg will profitably combine different modeling and simulation methods in order to be able to depict the physical properties of wind flow even more precisely – both with regard to individual wind turbines and entire wind farms.
Research focus
Mainframe until 2023
For the investigation of wind flows at the rotor blade, individual turbines and also entire wind farms, ForWind – Center for Wind Energy Research has a powerful high-performance computing (HPC) cluster. This HPC cluster, named “Eddy”, can compute simulations from the field of numerical flow simulation with 5,856 computational cores and a computing power of up to 170 TFlop/s. Additional computing power is also available on the HPC cluster “CARL” at Carl von Ossietzky University, which has an additional 7,592 computing cores or up to 230 TFlop/s. For the numerical calculations, software frameworks such as PALM for large-eddy simulation or OpenFOAM are used to calculate problems in energy meteorology, rotor blade optimization, wind turbine simulation or wind farm simulation.
Mainframes from 2023
With STORM & MOUSE, the previous HPC cluster has been replaced by a more powerful system for computational fluid dynamics (CFD). Funding was provided by two projects acquired by ForWind Oldenburg with the participation of the Scientific Computing Department of the Carl von Ossietzky University of Oldenburg. The latest generation HPC cluster, which has been in use at the Carl von Ossietzky University of Oldenburg since summer 2023, enables researchers to jointly examine physical phenomena across several orders of magnitude and time scales and to use machine learning methods in the process. The new, more powerful CFD cluster replaces the HPC cluster EDDY currently in use and will be jointly used by ForWind and Fraunhofer IWES.
