www.Autodiff.org - Publication: Delayed Propagation of Derivatives in a Two-dimensional Aircraft Design Optimization Problem

Delayed Propagation of Derivatives in a Two-dimensional Aircraft Design Optimization Problem

- Part of a collection -

Area
Computational Fluid Dynamics

Author(s)
H. Martin Bücker , A. Rasch , E. Slusanschi , Christian H. Bischof

Published in
Proceedings of the 17th Annual International Symposium on High Performance Computing Systems and Applications and OSCAR Symposium, Sherbrooke, Québec, Canada, May 11--14

Editor(s)
D. Sénéchal

Year
2003

Publisher
NRC Research Press

Abstract
In computational fluid dynamics (CFD), the use of high-performance computers is often indispensable, simply to manage the high request for computation time and storage. We consider a particular two-dimensional aircraft design problem and report on the use of automatic differentiation (ad) to obtain accurate derivatives rather than approximations based on numerical differentiation. More precisely, we apply the ad system ADIFOR to the large-scale CFD solver TFS and quantify the computational savings of a technique starting the derivative computations in a delayed manner over a black box ad approach.

AD Tools
ADIFOR

AD Theory and Techniques
Iteration

Related Applications
- Differentiation of the TFS Package

BibTeX
@INPROCEEDINGS{
         Bucker2003DPo,
       TITLE = "Delayed Propagation of Derivatives in a Two-dimensional Aircraft Design Optimization
         Problem",
       BOOKTITLE = "Proceedings of the 17th Annual International Symposium on High Performance
         Computing Systems and Applications and OSCAR Symposium, Sherbrooke, Qu\'{e}bec, Canada,
         May 11--14",
       editor = "D. S\'{e}n\'{e}chal",
       PUBLISHER = "NRC Research Press",
       pages = "123--126",
       address = "Ottawa",
       abstract = "In computational fluid dynamics (CFD), the use of high-performance computers is
         often indispensable, simply to manage the high request for computation time and storage. We consider
         a particular two-dimensional aircraft design problem and report on the use of automatic
         differentiation~(AD) to obtain accurate derivatives rather than approximations based on numerical
         differentiation. More precisely, we apply the AD system ADIFOR to the large-scale CFD solver TFS and
         quantify the computational savings of a technique starting the derivative computations in a delayed
         manner over a black box AD approach.",
       ad_area = "Computational Fluid Dynamics",
       ad_tools = "ADIFOR",
       ad_theotech = "Iteration",
       YEAR = "2003",
       author = "H. Martin B{\"u}cker and A. Rasch and E. Slusanschi and Christian
         H.~Bischof"
}

back

autodiff.org
Username:
Password: