Shock Optimization for Airfoil Design Problems
We begin by recalling the classical approach to solve shape design problems by gradient methods. Then we proceed to explain how automatic differentiation can simplify the analysis and illustrate this approach to a shape design problem where the shock is required to be at a certain place. The object of this chapter is to show that automatic differentiation seems to work even though the gradients and the calculus of variation have to be extended by Distribution theory.
KeywordsMach Number Dirac Masse Automatic Differentiation NACA Airfoil Transonic Speed
Unable to display preview. Download preview PDF.
- 1.A. Griewank: Evaluating Derivatives, Principles and Techniques of Algorithmic Differentiation. Vol 19, Frontiers in Applied Mathematics. SIAM, Philadelphia. (2000).Google Scholar
- 2.B. Mohammadi and O. Pironneau: Applied Numerical Optimal Shape Design. Oxford U. Press. (2001).Google Scholar
- 3.A. Caughey and M. Hafez (ed.): Pros and Cons of Airfoil Optimization, Frontier of Computational Fluid Dynamics, World Scientific, Singapore. (1998).Google Scholar
- 4.C. Bardos and O. Pironneau : Derivatives and Control in the Presence of Shocks. Computational Fluid Dynamics Journal, vol 12, no.1 (April 2003).Google Scholar