CSAR Seminar
SPEAKER: Manuel N. D. Barcelos Jr.,
University of Colorado at Boulder
TITLE:
Aeroelastic Optimization Methodology for Viscous and Turbulent Flows
DATE: Tuesday, April 24, 2007
TIME: 10:00 A.M.
PLACE: 2240 DCL
1304 W. Springfield Ave., Urbana, IL
ABSTRACT
In recent years, the development of faster computers and parallel
processing allowed the application of high-fidelity analysis methods to
the aeroelastic design of aircraft. However, these methods are
restricted to the final design verification, mainly due to the
computational cost involved in iterative design processes. Therefore,
this work is concerned with the creation of a robust and efficient
aeroelastic optimization methodology for inviscid, viscous and
turbulent flows by using high-fidelity analysis and sensitivity
analysis techniques. Most of the research in aeroelastic optimization,
for practical reasons, treats the aeroelastic system as a quasi-static
inviscid problem. In this work, as a first step toward the creation of
a more complete aeroelastic optimization methodology for realistic
problems, an analytical sensitivity computation technique was developed
and tested for quasi-static aeroelastic viscous and turbulent flow
configurations. Viscous and turbulent effects are included by using an
averaged discretization of the Navier-Stokes equations, coupled with an
eddy viscosity turbulence model. For quasi-static aeroelastic problems,
the traditional staggered solution strategy has unsatisfactory
performance when applied to cases where there is a strong
fluid-structure coupling. Consequently, this work also proposes a
solution methodology for aeroelastic and sensitivity analyses of
quasi-static problems based on the fixed point of an iterative
nonlinear block Gauss-Seidel scheme. The methodology can also be
interpreted as the solution of the Schur complement of the linearized
systems of equations from aeroelastic and sensitivity analyses. The
methodologies developed in this work are tested and validated by using
realistic aeroelastic systems.