--------------------------------------------------------------------
I-AIM Seminar on Computation and Modelling 
3:10 PM Friday December 3 in SS 5017A (100 St. George Street)

AERODYNAMIC OPTIMIZATION: A NEWTON-KRYLOV APPROACH

Professor David Zingg (UofT Institute for Aerospace Studies) 

Numerical aerodynamic optimization is the process of computing an 
aerodynamic shape that minimizes or maximizes a specified performance 
objective subject to imposed constraints. Over the past few years, 
great progress has been made in this area, primarily as a result of the 
use of adjoint methods. This presentation describes a Newton-Krylov 
approach to aerodynamic optimization consisting of a Newton-Krylov 
solver for the compressible Navier-Stokes equations, a Krylov solver 
for the discrete adjoint equation solved in order to compute the 
gradient, and a quasi-Newton gradient-based optimizer. Several examples 
are given to illustrate the issues involved, including single- and 
multi-element airfoil optimization at subsonic and transonic speeds, 
optimization at multiple operating points, and a Pareto front for a 
multi-objective problem. The multi-objective results are compared with 
those computed using a genetic algorithm. We conclude with some recent 
results and a discussion of future research directions. The 
presentation is aimed at a broad audience, and no specific knowledge of 
aerodynamics is assumed.

Refreshments to follow in the adjacent Math Lounge.

---------------------------------------------------------------------