An efficient immersed boundary treatment for complex moving object

Cheng Liu, Changhong Hu

Research output: Contribution to journalArticlepeer-review

79 Citations (Scopus)

Abstract

An efficient immersed boundary treatment for simulation of flexible moving body immersed in fluid is presented. The level set signed distance function is used to indicate body surface. A simple mapping strategy is proposed to avoid costly signed distance re-initialization computations. The strategy is efficient for both rigid and deformable structures, and can be extended to 3-D case easily. With the use of signed distance function, reconstruction of flow variables on a body surface can be easily implemented through linear, bilinear or quadratic interpolation. In order to suppress the pressure oscillations caused by the role conversion of forcing points and fluid points, a modified interpolation scheme is presented by introducing a dynamic weight term to the fluid point. Together with local grid refinement and using larger time step, the pressure oscillations can be effectively diminished by the modified interpolation scheme. The proposed method is validated by 2-D numerical simulations on a uniform flow past a fixed cylinder, in-line and transversely oscillating cylinders, a flapping wing, a 3-D simulation on a uniform flow past a rectangle plane with an initial attack angle and a moving anguilliform swimmer. Present numerical results are compared with other CFD solutions as well as experimental results.

Original languageEnglish
Pages (from-to)654-680
Number of pages27
JournalJournal of Computational Physics
Volume274
DOIs
Publication statusPublished - Oct 1 2014

All Science Journal Classification (ASJC) codes

  • Numerical Analysis
  • Modelling and Simulation
  • Physics and Astronomy (miscellaneous)
  • General Physics and Astronomy
  • Computer Science Applications
  • Computational Mathematics
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'An efficient immersed boundary treatment for complex moving object'. Together they form a unique fingerprint.

Cite this