From Success Stories in Scientific Visualization Wiki
A diesel particulate filter (DPF), which collects soot from automotive exhaust, needs to be cleaned periodically to avoid becoming clogged. The filter regeneration process (soot oxidation) needs to be quick, to be as complete as possible, and produce minimum pollution, resulting in a multi-parameter optimization design problem. Interactive visual analysis of large and complex simulation data allows one to understand complex relations within datasets like the gas flow through a DPF during filter regeneration. The interactive analysis here involved the joint investigation of 10 data dimensions (x, y, z, time, velocity, temperature, carbon monoxide, carbon dioxide, oxygen, and soot). Multiple linked views were used to drive the analysis concurrently in attribute space as well as in the space and time of the simulation. The image shows how the oxidation front in the DPF has been first selected by means of interactive brushing in two views of the attribute space of the simulation (on the left side) and then visualized at two different points in time on the right side (color encodes velocity). With this kind of interactive visual analysis it was possible to show that (due to a particular bend of the exhaustion system before the DPF) not enough oxygen was transported to one side in the back of the DPF and that therefore the oxidation dies off there before burning all the soot, requiring an improved design of the exhaustion system. One main research challenge for this type of engineering problem is the improved integration of data semantics within the process of interactive visual analysis, i.e., which features to show in the context of which others.
H. Doleisch, M. Mayer, M. Gasser, R. Wanker, and H. Hauser,
"Case study: Visual analysis of complex, time-dependent
simulation results of a diesel exhaust system." Proceedings
of the 6th Joint IEEE TCVG —EUROGRAPHICS Symposium
on Visualization (VisSym 2004), May 2004, pp. 91-96.
