Reliability Analysis Method Based on Support Vector Machines Classification and Adaptive Sampling Strategy

Abstract:

For probabilistic design problems with implicit limit state functions encountered in practical application, it is difficult to perform reliability analysis due to the expensive computational cost. In this paper, a new reliability analysis method which applies support vector machine classification(SVM-C) and adaptive sampling strategy is proposed to improve the efficiency. The SVM-C constructs a model defining the boundary of failure regions which classifies samples as safe or failed using SVM-C, then this model is used to replace the true limit state function,thus reducing the computational cost. The adaptive sampling strategy is applied to select samples along the constraint boundaries. It can also improves the efficiency of the proposed method. In the end, a probability analysis example is presented to prove the feasible and efficient of the proposed method.

• Periodical: Advanced Materials Research (Volume 544)
• Pages: 212-217
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.544.212
• Online since: June 2012
• Authors: Hong Yan Hao, Hao Bo Qiu, Zhen Zhong Chen, Hua Di Xiong
• Keywords: Adaptive Sampling Strategy, Reliability Analysis, Support Vector Machine Classification

References

[1] Wong FS. Slope reliability and response surface method. J Geotech Eng1985;111(112):32–53.

[2] Faravelli L. Response surface approach for reliability analysis.J Eng Mech1989;115(12):2763– 81.

[3] Youn, BD ,Choi, KK .A new response surface methodology for reliability-based design optimization.Computers & Structures.Volume 82, Issues 2–3, January 2004, Pages 241–256

DOI: 10.1016/j.compstruc.2003.09.002

[4] Xiao-hui Tan , Wei-hua Bi, Xiao-liang Hou, Wei Wang.Reliability analysis using radial basis function networks and support vector machines.Comput Geotech, 38 (2) (2011), p.178–186

DOI: 10.1016/j.compgeo.2010.11.002

[5] Haldar A, Mahadevan S. Probability, reliability, and statistical methods in engineering design. New York: Wiley and Sons; 2000.

[6] Youn BD, Choi KK, Du L. Adaptive probability analysis using an enhanced hybrid mean value method. Structural and Multidisciplinary Optimization 2005;29(2):134–48.

DOI: 10.1007/s00158-004-0452-6

[7] Li HS, Lv ZZ, Yue ZF. Support vector machine for structural reliability analysis.Appl Math Mech 2006;27(10):1135–43.

[8] S. Missoum, P. Ramu, R.T. Haftka, A convex hull approach for the reliability-based design of nonlinear transient dynamic problems, Comput. Meth. Appl.Mech. Engrg. 196 (2007) 2895–2906.

DOI: 10.1016/j.cma.2006.12.008

[9] A. Basudhar, S. Missoum, A.H. Sanchez, Limit state function identification using support vector machines for discontinuous responses and disjoint failure domains, Prob. Engrg. Mech. 23 (1) (2008) 1–11.

DOI: 10.1016/j.probengmech.2007.08.004

[10] A. Basudhar, S. Missoum, Adaptive explicit decision functions for probabilistic design and optimization using support vector machines, Comput. Struct. 86 (19–20) (2008) 1904–1917.

DOI: 10.1016/j.compstruc.2008.02.008

[11] Greg Schohn, David Cohn. Less is More: Active learning with support vector machines. In: Proceedings of the 17th international conference on machine learning. San Francisco, CA, USA: Morgan Kaufmann Publishers Inc.; 2000. p.839–46.

[12] Shawe-Taylor J, Cristianini N. Kernel methods for pattern analysis. Cambridge University Press; 2004.

[13] Tou JT, Gonzalez RC. Pattern recognition principles. Addison-Wesley; 1974.

[14] C.-W. Hsu, C.-C. Chang, C.-J. Lin. A practical guide to support vector classification.

[15] C.-C. Chang and C.-J. Lin. LIBSVM : a library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1--27:27, 2011.

DOI: 10.1145/1961189.1961199

[16] Gunn SR. Support vector machines for classification and regression. Technical Report ISIS-1-98, Department of Electronics and Computer Science, University of Southampton; 1998.

[17] Butler AN. Optimal and orthogonal latin hypercube designs for computer experiments. Biometrika 2001;88(3):847–57.

DOI: 10.1093/biomet/88.3.847

[18] Romero VJ,Burkardt JV,Gunzburger MD,Peterson JS. Comparison of pure and Latinized centroidal Voronoi tesselation against various other statistical sampling method.J Reliab Eng Safety 2006;91:1266-80.

DOI: 10.1016/j.ress.2005.11.023