Reliability assessment with correlated variables using support vector machines

This paper presents an approach to estimate probabilities of failure in cases where the random variables are correlated. An explicit limit state function is constructed in the uncorrelated standard normal space using the Nataf transformation and a support vector machine (SVM). The method of explicitly constructing the limit state function is referred to as explicit design space decomposition (EDSD), which also includes an adaptive sampling strategy to build an accurate SVM approximation. Several analytical examples with various distributions and also multiple failure modes are presented.

[1]  Samy Missoum,et al.  Constrained Ecient Global Optimization with Probabilistic Support Vector Machines , 2010 .

[2]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[3]  R. Brereton,et al.  Support vector machines for classification and regression. , 2010, The Analyst.

[4]  Antonio Harrison Sánchez,et al.  Limit state function identification using Support Vector Machines for discontinuous responses and disjoint failure domains , 2008 .

[5]  Qiang Du,et al.  Centroidal Voronoi Tessellations: Applications and Algorithms , 1999, SIAM Rev..

[6]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[7]  Samy Missoum,et al.  A Sampling-Based Approach for Probabilistic Design with Random Fields , 2008 .

[8]  Robert E. Melchers,et al.  Structural Reliability: Analysis and Prediction , 1987 .

[9]  Jorge E. Hurtado,et al.  An examination of methods for approximating implicit limit state functions from the viewpoint of statistical learning theory , 2004 .

[10]  P. E. James T. P. Yao,et al.  Probability, Reliability and Statistical Methods in Engineering Design , 2001 .

[11]  Samy Missoum,et al.  Local update of support vector machine decision boundaries , 2009 .

[12]  A. Basudhar,et al.  Adaptive explicit decision functions for probabilistic design and optimization using support vector machines , 2008 .

[13]  Parviz E. Nikravesh,et al.  Reliability-Based Optimal Design and Tolerancing for Multibody Systems Using Explicit Design Space Decomposition , 2010 .

[14]  A. Kiureghian,et al.  Multivariate distribution models with prescribed marginals and covariances , 1986 .

[15]  M. Rosenblatt.  Remarks on a Multivariate Transformation , 1952 .

[16]  Samy Missoum,et al.  Simulation and probabilistic failure prediction of grafts for aortic aneurysm , 2010 .

[17]  Raphael T. Haftka,et al.  A convex hull approach for the reliability-based design optimization of nonlinear transient dynamic problems , 2007 .

[18]  K. K. Choi,et al.  Reliability-based design optimization of problems with correlated input variables using a Gaussian Copula , 2009 .

[19]  Vicente J. Romero,et al.  Comparison of pure and "Latinized" centroidal Voronoi tessellation against various other statistical sampling methods , 2006, Reliab. Eng. Syst. Saf..

[20]  Samy Missoum,et al.  Reliability Assessment using Probabilistic Support Vector Machines (PSVMs) , 2010 .

[21]  A. Kiureghian,et al.  STRUCTURAL RELIABILITY UNDER INCOMPLETE PROBABILITY INFORMATION , 1986 .

[22]  A. Basudhar,et al.  An improved adaptive sampling scheme for the construction of explicit boundaries , 2010 .