×

Designing an aerofoil with a Fowler flap using artificial neural networks. (English) Zbl 1484.76063

Summary: The paper considers the problem of designing an aerofoil with a Fowler flap. The proposed approach is based on the use of artificial neural networks for rapid evaluation of aerodynamic characteristics. The linear method of principal component analysis (PCA) is used to reduce the dimensionality of design parameter space and to generate “random” airfiols. The simulated annealing method is used to find the optimal shape of the airfoil and flap.

MSC:

76N25 Flow control and optimization for compressible fluids and gas dynamics
76H05 Transonic flows
76M99 Basic methods in fluid mechanics
68T05 Learning and adaptive systems in artificial intelligence

Software:

XFOIL
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Application of Artificial Neural Networks in Problem of Applied Aerodynamics, Collection of Articles, Ed. by Yu. N. Sviridenko, Vol. 2678 of TsAGI Works (TsAGI, Moscow, 2008) [in Russian].
[2] Kahaner, D.; Mouler, K.; Nash, S., Numerical Methods and Software (1988), Hoboken: Prentice Hall, Hoboken
[3] S. Watanabe, ‘‘Karhunen-Loéve expansion and factor analysis theoretical remarks and applications,’’ in Proceedings of the 4th Prague Conference on Information Theory (1965), pp. 635-660.
[4] A. V. Bernstein, E. V. Burnaev, E. A. Dorofeev, Yu. N. Sviridenko, and S. S. Chernova, ‘‘Cascade dimensionality reduction procedures,’’ in Proceedings of the 11th National Conference on Artificial Intelligence (2008), Vol. 1, pp. 241-250.
[5] M. Drela, XFOIL 6.9 (MIT Aero and Astro, Harold Youngren, Aerocraft, Inc).
[6] Wolkov, A. V.; Lyapunov, S. V., Numerical prediction of transonic viscous separated flow past an airfoil, Theor. Comput. Fluid Dyn., 6, 1 (1994) · Zbl 0792.76058 · doi:10.1007/BF00417926
[7] S. V. Lyapunov and A. V. Wolkov, ‘‘Application of viscous-inviscid interaction methods for a separated flow calculation about airfoils and high-lift systems,’’ ICAS Proc. ICAS-96-1.10.2 (1996).
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.