Application of Gaussian moment closure to microscale flows with moving embedded boundaries

Research output: Contribution to journalArticlepeer-review

Abstract

The application of the Gaussian moment closure to continuum and microscale flows with embedded, and possibly moving, boundariesis considered. The Gaussian moment closure is briefly reviewed, asis an extension that allows for the treatment of flow of diatomic gases. Aparallel upwind, finite volume scheme with adaptive mesh refinement using a Roe-type numerical flux function is described for solving the hyperbolic system of partial differential equations arising from this closure on multiblock meshes with embedded and possibly moving boundaries. The purely hyperbolic nature of moment equations makes them particularly insensitive to discretizations involving grids with irregularities. Typical of adaptive mesh-refinement, embedded-boundary, and Cartesian cut-cell treatments, mesh irregularities are difficultto deal with when second derivatives are requiredby the physical model. Suchisthe case for the Navier-Stokes equations. Numerical solutions to mathematical descriptions involving second derivatives show significantly degraded solution quality as compared to solutions of first-order quasi-linear moment equations. Solidwall boundary conditions are implemented via a Knudsen-layer approximation. Comparisons are made between numerical solutions of the Gaussian modelon both body-fitted meshes and meshes with embedded boundaries, aswell as to experimental and approximate analytic results for a variety of flow problems. The benefits and potential of the proposed approach for unsteady microscale flow applications having complex geometries are clearly demonstrated.

Original languageEnglish (US)
Pages (from-to)1839-1857
Number of pages19
JournalAIAA journal
Volume52
Issue number9
DOIs
StatePublished - Sep 2014
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering

Fingerprint

Dive into the research topics of 'Application of Gaussian moment closure to microscale flows with moving embedded boundaries'. Together they form a unique fingerprint.

Cite this