An automatic differentiation techniqueis applied here to anindustrial computational fluid dynamics code with the goal of efficiently evaluating aerodynamic coefficients. Basic concepts about the numerical technique to automatically differentiate a code are first discussed, and then methodologies to apply the capabilities of the Institut National de Recherche en Informatique et en Automatique derivation tool to an industrial finite volume solver are discussed. Afterward, two examples of application to representative two-dimensional and three-dimensional test cases are shown and, for each one, comparisons are carried out between the results gathered using the automatic differentiation and the classical second-order finite difference technique for the evaluation of derivatives. Such comparison highlights the capability of automatic differentiation to produce reliable results even if applied to complex configurations and in very complex numerical structures. Finally, tradeoff considerations are made as a conclusion.
Industrial computational fluid dynamics tools for the evaluation of aerodynamic coefficients / Necci, C; Ceresola, N; Guglieri, Giorgio; Quagliotti, Fulvia. - In: JOURNAL OF AIRCRAFT. - ISSN 0021-8669. - 46:6(2009), pp. 1973-1983. [10.2514/1.42547]
Industrial computational fluid dynamics tools for the evaluation of aerodynamic coefficients
GUGLIERI, GIORGIO;QUAGLIOTTI, Fulvia
2009
Abstract
An automatic differentiation techniqueis applied here to anindustrial computational fluid dynamics code with the goal of efficiently evaluating aerodynamic coefficients. Basic concepts about the numerical technique to automatically differentiate a code are first discussed, and then methodologies to apply the capabilities of the Institut National de Recherche en Informatique et en Automatique derivation tool to an industrial finite volume solver are discussed. Afterward, two examples of application to representative two-dimensional and three-dimensional test cases are shown and, for each one, comparisons are carried out between the results gathered using the automatic differentiation and the classical second-order finite difference technique for the evaluation of derivatives. Such comparison highlights the capability of automatic differentiation to produce reliable results even if applied to complex configurations and in very complex numerical structures. Finally, tradeoff considerations are made as a conclusion.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2370363
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo