Skip to main navigation Skip to search Skip to main content

Real-time particle fluid simulation with WCSPH

  • Fengquan Zhang
  • , Xukun Shen
  • , Xiang Long
  • , Bin Zhao
  • , Lei Hu
  • Beihang University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, based on weakly compressible smoothed particles hydrodynamics (WCSPH), we present a method for simulating high quality fluid. We propose a new pressure equation to reduce time overhead, speed up convergence towards small density fluctuations and allow for a larger time step than that of traditional approaches. And based on the new equation, we introduce a new adaptive time-stepping approach that automatically adapts the time step according to each particle current individual time step, thus improving the efficiency of computation and contributing much to a real-time particle fluid simulation. For real-time simulation, we make full use of CPU and GPU in a combined way to build a parallel platform, namely Multi-GPU platform. And optimization of the platform design is made to run a simulation with realistic visual effects in real time.

Original languageEnglish
Title of host publication20th Pacific Conference on Computer Graphics and Applications, PG 2012 - Short Papers
EditorsChris Bregler, Pedro Sander, Michael Wimmer
PublisherIEEE Computer Society
Pages29-34
Number of pages6
ISBN (Electronic)9783905673944
DOIs
StatePublished - 2012
Event20th Pacific Conference on Computer Graphics and Applications, PG 2012 - Hong Kong, Hong Kong SAR
Duration: 12 Sep 201214 Sep 2012

Publication series

NameProceedings - Pacific Conference on Computer Graphics and Applications
Volume2012-September
ISSN (Print)1550-4085

Conference

Conference20th Pacific Conference on Computer Graphics and Applications, PG 2012
Country/TerritoryHong Kong SAR
CityHong Kong
Period12/09/1214/09/12

Fingerprint

Dive into the research topics of 'Real-time particle fluid simulation with WCSPH'. Together they form a unique fingerprint.

Cite this