A new matrix dissipation model for central scheme

Xinrong Su, Satoru Yamamoto

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)


In this paper, a simple and efficient improvement to the famous Swanson-Turkel matrix dissipation model for the central scheme is proposed. In the new matrix dissipation model, the accuracy is improved by eliminating the second-difference dissipation added to the characteristic fields representing the vorticity waves. This strategy is proposed based on analyzing the flow-physics about shock-vortex interaction using the Rankine-Hugoniot jump condition. In this paper, the behavior of central scheme for rotational flow is also theoretically and numerically analyzed. Results show a newfound problem of the original scalar and matrix dissipation models, in which for rotational flow excessive second-difference dissipation is added due to the pressure-based shock sensor. With current new matrix dissipation model improved accuracy is obtained at minimal cost overhead, especially, in the highly vortical region where the second-difference dissipation is reduced. At the same time, it preserves the excellent shock capturing capability and convergence speed of original method. Numerical properties of this new matrix dissipation model are validated with a series of numerical experiments and results comparison with original model verifies improved performance of current method.

Original languageEnglish
Pages (from-to)494-513
Number of pages20
JournalInternational Journal for Numerical Methods in Fluids
Issue number7
Publication statusPublished - 2014 Mar 10


  • Artificial dissipation
  • Central scheme
  • Matrix dissipation
  • Rankine-Hugoniot jump condition
  • Turbulent flow
  • Vortical flow


Dive into the research topics of 'A new matrix dissipation model for central scheme'. Together they form a unique fingerprint.

Cite this