1 Copyright © 2012 by ASME
Proceedings of the ASME 2012 International Design Engineering Technical Conferences &
Computers and Information in Engineering Conference
IDETC/CIE 2012
August 12-15, 2012, Chicago, Illinois, USA
DETC2012-70715
Sampling-based RBDO using Probabilistic Sensitivity
Analysis and Virtual Support Vector Machine
Hyeongjin Song, K.K. Choi*
Department of Mechanical & Industrial Engineering
College of Engineering
The University of Iowa, Iowa City, IA 52242, USA
hyesong@engineering.uiowa.edu, kkchoi@engineering.uiowa.edu
*Corresponding author
Ikjin Lee
Department of Mechanical
Engineering
The University of Connecticut,
Storrs, CT 06269, USA
ilee@engr.uconn.edu
Liang Zhao
Schlumberger
1310 Rankin Rd, Houston, TX
77073, USA
lzhao01@slb.com
David Lamb
US Army RDECOM/TARDEC
Warren, MI 48397-5000, USA
david.lamb@us.army.mil
ABSTRACT
In this paper, a sampling-based RBDO method using a
classification method is presented. The probabilistic
sensitivity analysis is used to compute sensitivities of
probabilistic constraints with respect to random variables.
Since the probabilistic sensitivity analysis requires only the
limit state function, and not the response surface or sensitivity
of the response, an efficient classification method can be used
for a sampling-based RBDO. The proposed virtual support
vector machine (VSVM), which is a classification method, is a
support vector machine (SVM) with virtual samples. By
introducing virtual samples, VSVM overcomes the deficiency
in existing SVM that uses only classification information as
their input. In this paper, the universal Kriging method is used
to obtain locations of virtual samples to improve the accuracy
of the limit state function for highly nonlinear problems. A
sequential sampling strategy effectively inserts new samples
near the limit state function. In sampling-based RBDO,
Monte Carlo simulation (MCS) is used for the reliability
analysis and probabilistic sensitivity analysis. Since SVM is
an explicit classification method, unlike implicit methods,
computational cost for evaluating a large number of MCS
samples can be significantly reduced. Several efficiency
strategies, such as the hyper-spherical local window for
generation of the limit state function and the
Transformations/Gibbs sampling method to generate uniform
samples in the hyper-sphere, are also applied. Examples show
that the proposed sampling-based RBDO using VSVM yields
better efficiency in terms of the number of required samples
and the computational cost for evaluating MCS samples while
maintaining accuracy similar to that of sampling-based RBDO
using the implicit dynamic Kriging (D-Kriging) method.
KEYWORDS
Surrogate Model, Sampling-based RBDO, Probabilistic
Sensitivity Analysis, Support Vector Machine (SVM),
Sequential Sampling, Virtual Samples, Virtual Support Vector
Machine (VSVM), Hyper-spherical Local Window
1. Introduction
The reliability-based design optimization (RBDO) methods
that use sensitivity have been widely applied to various
engineering applications. Using sensitivities, the most probable
point (MPP)-based reliability analysis can be carried out for
approximating the reliability of the system [1-6]. However, the
sensitivity often is not available or is difficult to obtain in
complex multi-physics or multidisciplinary simulation-based
engineering design applications. In this case, the surrogate-
based method can provide approximations of otherwise
expensive computer simulations for design optimization [7]. In
RBDO, once accurate surrogate models are constructed, Monte
Carlo simulation (MCS) [8-11] can be directly applied to the
surrogate model for the reliability analysis and probabilistic