A Novel RBF Tuning Strategy for Global Surrogate Modeling

Article Preview

Abstract:

In this paper, a research is performed on the tuning strategy for the shape parameter of the RBF to improve accuracy. The shape parameter is an influential factor for the accuracy but the configuration is commonly based on experience. An algorithm is designed to optimize the shape parameter. Three high-order mathematical benchmarks are used to compare the common RBF and the optimized RBF. This paper shows that the optimized RBF is more accurate for various test functions.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

881-885

Citation:

Online since:

October 2014

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Bi Z, Liu Y, Jiang W, et al. A brief review of dual-frequency capacitively coupled discharges[J]. Current Applied Physics, 2011, 11(5): S2-S8.

DOI: 10.1016/j.cap.2011.07.002

Google Scholar

[2] Sacks J, Welch WJ, Mitchell TJ, Wynn HP (1989) Design and Analysis of Computer Experiments. 409-423. doi: 10. 1214/ss/1177012413.

Google Scholar

[3] Wang GG, Shan S (2007) Review of metamodeling techniques in support of engineering design optimization. Journal of Mechanical Design 129 (4): 370-380.

DOI: 10.1115/1.2429697

Google Scholar

[4] Viana FAC, Simpson TW, Balabanov V, Toropov V (2014) Metamodeling in Multidisciplinary Design Optimization: How Far Have We Really Come? AIAA Journal 52 (4): 670-690. doi: 10. 2514/1. J052375.

DOI: 10.2514/1.j052375

Google Scholar

[5] Fang H, Horstemeyer M F. Global response approximation with radial basis functions[J]. Engineering Optimization, 2006, 38(04): 407-424.

DOI: 10.1080/03052150500422294

Google Scholar