Interactive Particle Tracing for Visualizing Large, Time-Varying Flow Fields

Topics: Data compression, Time, Particle system Pages: 22 (7820 words) Published: January 6, 2013
Delft University Of Technology, 2009

Technical Report VIS2009-01

Interactive Particle Tracing for Visualizing Large, Time-Varying Flow Fields Dylan Dussel, Eric J. Griffith, Michal Koutek and Frits H. Post Data Visualization Group, Delft University of Technology, Netherlands Technical Report VIS 2009-01

Abstract Particle tracing is a classical method of flow field visualization. For interactive exploration, particles must be advected and displayed in real-time. Graphics Processor Unit (GPU) based techniques can advect hundreds of thousands or millions of particles in real-time. We have investigated such GPU-based techniques for interactive exploration of large, time-varying flow fields. Our approach can be roughly divided into three categories: data preprocessing, visualization and interaction. The preprocessing involves data compression, region of interest computation and preparation of multi-resolution data. For flow visualization, we use use the GPU for both data decompression and particle advection. More than 1,000,000 particles can be visualized at interactive frame rates and data rates. We support the standard particle visualization techniques of pathlines, streamlines and streaklines. We also represent particles as flow-oriented ellipsoids, which can additionally be moved over their traversed pathlines to explore their behavior in time. Dynamic features in the data are explored by interactively seeding and tracking particles through time in both a standard display screen and a stereoscopic virtual environment. Further, we have validated our particle system by comparing its particle trajectories with those generated by a Large-eddy Simulation. Categories and Subject Descriptors (according to ACM CCS): I.3.8 [Computer Graphics]: ApplicationsFlow Visualization

1. Introduction Computational Fluid Dynamics (CFD) techniques such as Large-Eddy Simulation (LES) or Direct Numerical Simulation (DNS) can produce very large, time-varying, multi-field data sets. Exploration and analysis of these data sets is difficult due to their size, complexity and time-varying nature. Various techniques have been developed over the years to explore different aspects of the data. One popular technique for studying the fluid flow characteristics, particle tracing, has recently been extended to handle time-varying flow data interactively with the help of the Graphics Processing Unit (GPU) ( [SBK06], [SBK07], and [BSK∗ 07] , Figure 1). One of the main challenges to extending particle tracing to time-varying data has been the sheer size of the data. Interactive exploration requires loading data time steps at interactive rates in addition to interactive frame rates for the particle tracing algorithm. To ensure interactive frame rates, Technical Report VIS2009-01

particle tracing algorithms are executed on the GPU. In order to maintain an interactive data rate, new data must be continuously transferred from a source, such as disk or a network computer, to the CPU and ultimately to the GPU. Both of these transfers represent potential bottlenecks, and, to overcome them, data reduction and filtering techniques must be applied. We present a system that supports interactive exploration of large, time-varying flow fields. With the help of the GPU, we are able to advect more than 1,000,000 particles at interactive frame rates and data rates, see Figure 1. The system is targeted at our primary source of data: LES data on a staggered, Cartesian grid with relatively compact features. The system includes both a preprocessing phase and an interactive particle tracing engine. The preprocessing step downsamples and quantizes the flow fields for each time step, and it identifies, extracts, and quantizes full resolution regions-

2

D. Dussel, E. J. Griffith, M. Koutek & F. H. Post / Interactive Particle Tracing for Time-Varying Flow Fields

describe our GPU-based particle system in Section 5. This includes the GPU-based data decompression and particle advection...

References: [AL77] A RAKAWA A., L AMB V.: Computational design of the basic dynamical processes of the UCLA general circulation model. In General circulation models of the atmosphere. Academic Press, Inc., New York, 1977, pp. 173–265. [BSK∗ 07] B ÜRGER K., S CHNEIDER J., KONDRATIEVA P., K RÜGER J., W ESTERMANN R.: Interactive visual exploration of unsteady 3D-flows. In Eurographics/IEEE VGTC Symposium on Visualization (2007), pp. 251–258. [GKP07] G RIFFITH E. J., KOUTEK M., P OST F. H.: Fast normal vector compression with bounded error. In Proc. Geometry Processing. (2007), pp. 263–272. [GPK∗ 05] G RIFFITH E. J., P OST F. H., KOUTEK M., H EUS T., J ONKER H. J. J.: Feature tracking in VR for cumulus cloud life-cycle studies. In Virtual Environments 2005 (2005), Kjems E., Blach R., (Eds.), pp. 121–128. [KKKW05] K RÜGER J., K IPFER P., KONDRATIEVA P., W ESTERMANN R.: A particle system for interactive visualization of 3D flows. IEEE Transactions on Visualization and Computer Graphics 11, 6 (November 2005), 744–756. [KKW05]
MANN
[Lan94] L ANE D. A.: UFAT: A particle tracer for timedependent flow fields. In Proc. IEEE Visualization (1994), pp. 257–264. [MCG94] M AX N., C RAWFIS R., G RANT C.: Visualizing 3D velocity fields near contour surfaces. In Proc. IEEE Visualization (1994), pp. 248–255. [OLG∗ 05] OWENS J. D., L UEBKE D., G OVINDARAJU N., H ARRIS M., K RÜGER J., L EFOHN A. E., P URCELL T. J.: A survey of general-purpose computation on graphics hardware. In Eurographics 2005, State of the Art Reports (2005), pp. 21–51. [PVH∗ 03] P OST F. H., V ROLIJK B., H AUSER H., L ARAMEE R. S., D OLEISCH H.: The state of the art in flow visualization: Feature extraction and tracking. Computer Graphics Forum 22, 4 (2003), 775–792. [SBK06] S CHIRSKI M., B ISCHOF C., K UHLEN T.: Exploring flow fields with GPU-based stream tracers in virtual environments. In Eurographics 2006 (2006), pp. 115–118. [SBK07] S CHIRSKI M., B ISCHOF C., K UHLEN T.: Interactive exploration of large data in hybrid visualization environments. In Virtual Environments 2007 (2007), pp. 69– 76. [vW93] VAN W IJK J. J.: Flow visualization with surface particles. IEEE Computer Graphics and Applications 13, 4 (1993), 18–24. [ZSH96] Z ÖCKLER M., S TALLING D., H EGE H.-C.: Interactive visualization of 3D-vector fields using illuminated streamlines. In Proc. IEEE Visualization (1996), pp. 107–113.
KONDRATIEVA P., K RÜGER J., W ESTER R.: The application of GPU particle tracing to diffusion tensor field visualization. In Proc. IEEE Visualization (2005), pp. 73–78.
[KSW04] K IPFER P., S EGAL M., W ESTERMANN R.: Uberflow: A GPU-based particle engine. In Proc. ACM SIGGRAPH/EUROGRAPHICS conference on Graphics hardware (2004), pp. 115–122.
Technical Report VIS2009-01
Continue Reading

Please join StudyMode to read the full document

You May Also Find These Documents Helpful

  • Flowchart and Theoretical Flow Time Essay
  • Fluid Flow in a Duct of Varying Cross-Section Report Essay
  • field study Essay
  • Essay on Open flow
  • On The Field Essay
  • Hard Times Essay
  • Essay about no time
  • time Essay

Become a StudyMode Member

Sign Up - It's Free