Practical Large-Eddy simulation for complex turbulent flowfield with adaptive cartesian mesh and data compression technique

Ryotaro Sakai, Shigeru Obayashi, Yuichi Matsuo, Kazuhiro Nakahashi

Research output: Contribution to conferencePaperpeer-review

1 Citation (Scopus)

Abstract

A practical turbulent flow simulation environment is established by incorporating a turbulent wall model and a data compression method as post-processing in a flow solver. The solver is based on the block-structured Cartesian mesh method named Building-Cube Method which achieves robust and automated mesh generation and efficient numerical simulation. The turbulent wall model which is based on immersed boundary method and the logarithmic law improves solution reliabilities on Cartesian mesh, without losing its computational efficiency. The data compression method consists of four techniques relevant to image encoding, and it efficiently compresses a flow simulation data set by making use of the mesh structure of the Building-Cube Method. The validation demonstrates that the wall model properly yields a turbulent boundary layer in terms of mean velocity and Reynolds stress, even in relatively coarse mesh. It is also demonstrated that in the simulation of turbulent flow past vortex generators the data compression method reduces the data size to nearly 10% of the original data, while its computational load is less than 1% of the main numerical simulation.

Original languageEnglish
Publication statusPublished - 2013
Event21st AIAA Computational Fluid Dynamics Conference - San Diego, CA, United States
Duration: 2013 Jun 242013 Jun 27

Other

Other21st AIAA Computational Fluid Dynamics Conference
CountryUnited States
CitySan Diego, CA
Period13/6/2413/6/27

ASJC Scopus subject areas

  • Fluid Flow and Transfer Processes
  • Energy Engineering and Power Technology
  • Aerospace Engineering
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Practical Large-Eddy simulation for complex turbulent flowfield with adaptive cartesian mesh and data compression technique'. Together they form a unique fingerprint.

Cite this