论文信息 - Application of Neural Networks to Wind Tunnel Data Response Surface Methods

Application of Neural Networks to Wind Tunnel Data Response Surface Methods

The integration of nonlinear neural network methods with conventional linear regression techniques is demonstrated for representative wind tunnel force balance data modeling. This work was motivated by a desire to formulate precision intervals for response surfaces produced by neural networks. Applications are demonstrated for representative wind tunnel data acquired at NASA Langley Research Center and the Arnold Engineering Development Center in Tullahoma, TN.

Ching F. Lo | Richard DeLoach | Jun L. Zhao

[1] John Moody,et al. Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[2] R. DeLoach. Improved Quality in Aerospace Testing Through the Modern Design of Experiments , 2000 .

[3] Shang-Liang Chen,et al. Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[4] R Fisher,et al. Design of Experiments , 1936 .

[5] Elizabeth A. Peck,et al. Introduction to Linear Regression Analysis , 2001 .

[6] Yoh-Han Pao,et al. Adaptive pattern recognition and neural networks , 1989 .

[7] Egon Stanewsky,et al. Reynold Number Scaling at Transonic Speeds , 1998 .

[8] Richard DeLoach,et al. Applications of modern experiment design to wind tunnel testing at NASA Langley Research Center , 1998 .