In the realm of energy and aerodynamics research, managing and analyzing vast amounts of data from computational simulations is becoming an increasingly significant challenge. A team of researchers from the German Aerospace Center (DLR), including Janis Geise, Sebastian Spinner, Richard Semaan, and Andre Weiner, has developed an improved algorithm to tackle this issue, aiming to make data analysis more efficient and accessible.
The researchers have enhanced an existing algorithm called Sparse Spatial Sampling (S3), which is designed to reduce the data from time-dependent flow simulations. The improved S3 algorithm works by creating a time-invariant octree grid based on a user-defined metric. This grid allows for efficient down-sampling of the data while preserving as much of the metric as possible. The goal is to make post-processing of the data more efficient and enable memory-intensive tasks, such as computing the modal decomposition of flow snapshots.
The enhanced S3 algorithm was tested on three different flow simulation scenarios: the flow past a tandem configuration of airfoils in the transonic regime, the incompressible turbulent flow past a circular cylinder, and the flow around an aircraft half-model at high Reynolds and Mach numbers. In all test cases, the algorithm significantly reduced the number of mesh cells by 35% to 98% while accurately preserving the dominant flow dynamics. This reduction in data size enables post-processing of computational fluid dynamics (CFD) data on a local workstation rather than requiring high-performance computing (HPC) resources for many cases.
The practical applications of this research for the energy sector are notable. For instance, in wind energy, CFD simulations are used to optimize the design of wind turbines and their blades. The improved S3 algorithm could make these simulations more efficient and accessible, potentially speeding up the development of more efficient wind turbines. Similarly, in the aerospace industry, the algorithm could aid in the design and optimization of aircraft, leading to more fuel-efficient and environmentally friendly planes. The research was published in the Journal of Computational Physics, a peer-reviewed scientific journal that focuses on the development and application of computational methods to problems in physics and engineering.
This article is based on research available at arXiv.