H.M. Bücker, B. Lang, A. Rasch, and C.H. Bischof (Germany)
Automatic Differentiation, Forward Mode, SensitivityAnalysis, Adifor
Derivatives are not only useful in sensitivity analysis but play a vital role in various areas of computational sci ence including parameter optimization of computer mod els. Among the various methods for obtaining derivatives, automatic differentiation (AD) combines freedom from ap proximation errors, high performance, and the ability to handle arbitrarily complex codes arising from large-scale scientific investigations. In this note, we show how AD technology can aid in the sensitivity analysis of a com puter model of an aircraft. To this end, the software tool ADIFOR, implementing the AD technology for functions written in Fortran 77, was applied to a large-scale compu tational fluid dynamics solver, TFS, developed at Aerody namisches Institut, Aachen University of Technology. This solver comprises approximately 24,000 lines of Fortran 77 and is able to resolve steady and unsteady laminar and tur bulent flows in two and three dimensions.
Important Links:
Go Back
