Combining Sparse and Dense Databases to Form a Robust Aerodynamic Model for a Long-Range High-Speed Projectile
Abstract
Several aerodynamic databases are combined to form an aerodynamic model for a high-speed long-range missile. A large test matrix of flight conditions was formed and exercised on a semi-empirical prediction code and an inviscid computation fluid dynamics (CFD) solver, providing full-density databases of low fidelity. Higher-fidelity methods, specifically Navier-Stokes CFD and wind tunnel tests, were applied to a small subset of the test matrix resulting in sparse but accurate predictions of the aerodynamic forces and moments. In this work we demonstrate a novel approach to combining the databases such that features of the low-fidelity predictions are preserved as they are tuned to intersect the sparse data provided by the high-fidelity sources. This is done by modeling the dense predictions with a set of basis functions, then tuning the basis functions to intersect the points from the high-fidelity sources. Harmonic basis functions are used to reflect the cyclical symmetry of the projectile used in this study. Examples show that this method outperforms established kriging methods.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2021
- Accession Number
- AD1142707
Entities
People
- Bradley T. Burchett
- Joseph D. Vasile
Organizations
- United States Army