Strategy Combining Regionally Adapted Processes (SCRAP)

Abstract

A local Navier-Stokes (NS) algorithm to simulate a physically and geometrically complex flow is presented. The domain is subdivided into blocks which facilitate the generation of global mesh and splitting of schemes. Basic types are inviscid, viscous and interaction blocks. In the interaction block a local procedure is used to decide whether to use NS or Euler equations. Each block also employs subdomain adaptation to capture small scale features. The method concentrates resources by modeling complex physics in only important regions. Physics adaptation treats those cells for which local stress is more than 1% of the maximum value as viscous cells. Subdomain adaptation is carried out by monitoring the first differences of selected variables. The cells for which these differences are large are locally divided. The SCRAP technique is demonstrated via a two dimensional model problem. It significantly reduces computational requirements and is comparatively more accurate than conventional schemes. Up to two orders of magnitude savings in computer time have been achieved, compared to a globally fine solution of full NS equations. The agreement with previous full NS computations over fine grids is excellent. Future work will extend the approach to three dimensional configurations. Phase II will simulate flows for complex configurations at large attack angles.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 30, 1991
Accession Number
ADA232158

Entities

People

  • M. M. Pervaiz

Tags

Communities of Interest

  • Energy and Power Technologies
  • Weapons Technologies

DTIC Thesaurus Topics

  • Boundary Layer
  • Central Processing Units
  • Computational Fluid Dynamics
  • Computational Science
  • Equations
  • Euler Equations
  • Fluid Dynamics
  • Fluid Flow
  • Geometry
  • Heat Flux
  • Mach Number
  • Reynolds Number
  • Specific Heat
  • Supersonic Flow
  • Three Dimensional
  • Two Dimensional
  • Viscous Flow

Fields of Study

  • Physics

Readers

  • Computational Fluid Dynamics (CFD)
  • Computational Modeling and Simulation