Transonic drag prediction using an unstructured multigrid solver
Author(s) -
Dimitri J. Mavriplis,
David Levy
Publication year - 2002
Publication title -
nasa sti repository (national aeronautics and space administration)
Language(s) - English
Resource type - Conference proceedings
DOI - 10.2514/6.2002-838
Subject(s) - transonic , unstructured grid , solver , computer science , multigrid method , lift to drag ratio , drag , mach number , computational fluid dynamics , grid , drag coefficient , computational science , algorithm , parallel computing , mathematics , mathematical optimization , aerodynamics , aerospace engineering , geometry , engineering , mathematical analysis , partial differential equation
We summarize the results obtained with the NSU3D unstructured multigrid solver for the First AIAA Drag Prediction Workshop held in Anaheim, California, in June 2001. The test case for the workshop consists of a wing-body configuration at transonic flow conditions. Flow analyses for a complete test matrix of lift coefficient values and Mach numbers at a constant Reynolds number are performed, thus producing a set of drag polars and drag rise curves, which are compared with experimental data. Results were obtained independently by both authors using an identical baseline grid and different refined grids. Most cases were run in parallel on commodity cluster-type machines, whereas the largest cases were run on an SGI Origin machine using 128 processors. We study the accuracy of the subject unstructured grid solver for predicting drag in the transonic cruise regime, to assess the efficiency of the method in terms of convergence, CPU time, and memory, and to determine the effects of grid resolution on this predictive ability and its computational efficiency
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom