(Georgia Institute of Technology, 1998-10-01)
Scott, Jeffrey
This paper presents a method of transforming aerodynamic datasets generated in
Aerodynamic Preliminary Analysis System (APAS) into parametric equations which may
subsequently be used in a multidisciplinary design optimization (MDO) environment for
analyzing aerospace vehicles.
APAS is an analysis code which allows the user to create a simple geometric
model of a vehicle and then calculate the aerodynamic force coefficients of lift, drag, and
pitching moment over a wide range of flight conditions. As such, APAS is a very useful
tool for conceptual vehicle designs since it allows the force coefficients for a given
design to be calculated relatively quickly and easily.
However, APAS suffers from an outdated user interface and, because it is tedious
to generate a new dataset during each design iteration, it is quite difficult to integrate into
an MDO framework. Hence the desire for a method of transforming the APAS output
into a more usable form.
The approach taken and described in this paper involves the use of regression
analysis techniques to accomplish the data transformation with three goals in mind. The
first goal was to develop a parametric model for calculating the aerodynamic coefficients
for a single unique geometry. The second goal was to extend this model to capture the
effects of changes in vehicle geometry. The third goal was to write a Fortran program
that would be capable of automatically carry out the regression analysis on a given APAS
data set and produce the desired parametric equations. This paper presents the results and
gives the model developed for analyzing a sample vehicle with a fixed geometry as well
as the results of a sample vehicle with a variable geometry. The Fortran computer code is
also given.