aviation, etc., computational flow simulation is increasingly
used as an integral part of the product development cycle --
car engines, vehicle bodies, as well as many other products
are subjected to complex simulation procedures (in early stages
of the development) in order to evaluate combustion processes,
aerodynamic properties, etc., before heavy-weight models are
built at high expenses. Simulations of computational fluid
dynamics (CFD) usually result in large and complex datasets.
Time-dependent flow in a 3D domain with dozens of flow attributes at each and every cell of the computational grid
usually causes result files to be many Megabytes to Gigabytes
large (even Terabytes in some cases). Analysis of these simulated flow datasets poses special challenges. Visualization is one powerful technology to efficiently assess the flow characteristics within such a dataset. Feature-based flow visualization allows to focus on interesting
parts of the flow and thereby offers especially interesting means to access very large datasets. In this talk SimVis technology (as developed at the VRVis Research Center in
Vienna, Austria) will be presented as one interesting approach
to interactive visual flow exploration and analysis. The
combination of views of different kinds allows to access all
the data attributes in an intuitive way. Features are interactively specified by brushing data attributes of special
interest (similar as domain experts describe flow features).
Apart from presenting the general approach, also examples will be shown which demonstrate the use of SimVis technology.
If time permits, an interactive demo will be provided during
the talk (or afterwards).