Visual comparison of CFD datasets in augmented reality

Kurzfassung

Over the years, several studies have been conducted, that focus on employing immersive environments for data visualization and analysis. Many studies have shown the benefits of 3d visualization in immersive technologies such as AR and VR, over the normal 2d screen. Immersive technologies provide a greater spatial understanding compared to the normal desktop applications, which is effective for many applications such as scientific visualization, which is concerned with the visualization of scientific data, that is spatial in nature. The CFD datasets are an example of scientific data. A team in DLR’s institute of software technology is therefore working on developing applications for visualization of high fidelity volumetric CFD simulation data in immersive augmented reality. This thesis aims to leverage the capabilities of AR to determine the anomalies caused by damage to a turbine blade. For this purpose, a system is being designed that visually compares two CFD datasets before and after physical damage on the blade, in a way that the difference is immediately apparent in AR. First, an automated pipeline encompassing data filtering, color mapping, and rendering in the HoloLens 2 AR headset is developed. Furthermore, a point cloud is generated, that represents the difference between the CFD datasets. The overall AR experience is enhanced by establishing a spatial relationship between a physical turbine blade and the CFD data using the Vuforia engine. This will help to analyze the effect of the airflow on the turbine blade. Finally, an intuitive user interface is designed, to enable interaction and comparison with the datasets.