Applied-Information Technology with an Approximation Method for Black Model

Abstract:

The complex black-box models, for example, the simulation model, finite element analysis model have been used in lots applications such as parameter test, sensitivity analysis, and semi physical simulation. Most of these models have a feature of multiple inputs multiple outputs (MIMO). When dealing with MIMO, the traditional methods of one objective per response surface not only cause a large number of memory consumption, but also reduce the efficiency of response surface building and valuation. In this paper, a new global approximation method based on the response surface group is proposed. Multi-objective can be approximated only once by response surface group, which greatly simplifies the construction of response surface. Through standard test, the effectiveness and advanced of proposed method is verified.

• Periodical: Applied Mechanics and Materials (Volume 685)
• Pages: 507-511
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.685.507
• Online since: October 2014
• Authors: Xiao Liang Yin, Yi Zhong Wu*, Shu Ting Wang
• Keywords: Black-Box Function, Multiple Inputs, Multiple Outputs, Response Surface Set

References

[1] Mathworks, Look-Up Table, Simulink Reference.

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

DOI: 10.1115/1.2429697

[3] Wei X, Wu Y Z, Chen L P: A new sequential optimal sampling method for radial basis functions. Applied Mathematics and Computation, 2012, 218(19): 9635-9646.

DOI: 10.1016/j.amc.2012.02.067

[4] Jones D R: Direct global optimization algorithm. Encyclopedia of optimization. Springer US, 2001: 431-440.