Abstract
This paper introduces a novel approach for reliability assessment with dependent variables. In this work, the boundary of the failure domain, for a computational problem with expensive function evaluations, is approximated using a Support Vector Machine and an adaptive sampling scheme. The approximation is sequentially refined using a new adaptive sampling scheme referred to as generalized "max-min". This technique efficiently targets high probability density regions of the random space. This is achieved by modifying an adaptive sampling scheme originally tailored for deterministic spaces (Explicit Space Design Decomposition). In particular, the approach can handle any joint probability density function, even if the variables are dependent. In the latter case, the joint distribution might be obtained from copula. In addition, uncertainty on the probability of failure estimate are estimated using bootstrapping. A bootstrapped coefficient of variation of the probability of failure is used as an estimate of the true error to determine convergence. The proposed method is then applied to analytical examples and a beam bending reliability assessment using copulas.
| Original language | English (US) |
|---|---|
| Title of host publication | 34th Computers and Information in Engineering Conference |
| Publisher | American Society of Mechanical Engineers (ASME) |
| Volume | 1A |
| ISBN (Print) | 9780791846285 |
| DOIs | |
| State | Published - 2014 |
| Event | ASME 2014 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE 2014 - Buffalo, United States Duration: Aug 17 2014 → Aug 20 2014 |
Other
| Other | ASME 2014 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE 2014 |
|---|---|
| Country | United States |
| City | Buffalo |
| Period | 8/17/14 → 8/20/14 |
Fingerprint
ASJC Scopus subject areas
- Mechanical Engineering
- Computer Graphics and Computer-Aided Design
- Computer Science Applications
- Modeling and Simulation
Cite this
A generalized "max-min" sample for reliability assessment with dependent variables. / Lacaze, Sylvain; Missoum, Samy.
34th Computers and Information in Engineering Conference. Vol. 1A American Society of Mechanical Engineers (ASME), 2014.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - A generalized "max-min" sample for reliability assessment with dependent variables
AU - Lacaze, Sylvain
AU - Missoum, Samy
PY - 2014
Y1 - 2014
N2 - This paper introduces a novel approach for reliability assessment with dependent variables. In this work, the boundary of the failure domain, for a computational problem with expensive function evaluations, is approximated using a Support Vector Machine and an adaptive sampling scheme. The approximation is sequentially refined using a new adaptive sampling scheme referred to as generalized "max-min". This technique efficiently targets high probability density regions of the random space. This is achieved by modifying an adaptive sampling scheme originally tailored for deterministic spaces (Explicit Space Design Decomposition). In particular, the approach can handle any joint probability density function, even if the variables are dependent. In the latter case, the joint distribution might be obtained from copula. In addition, uncertainty on the probability of failure estimate are estimated using bootstrapping. A bootstrapped coefficient of variation of the probability of failure is used as an estimate of the true error to determine convergence. The proposed method is then applied to analytical examples and a beam bending reliability assessment using copulas.
AB - This paper introduces a novel approach for reliability assessment with dependent variables. In this work, the boundary of the failure domain, for a computational problem with expensive function evaluations, is approximated using a Support Vector Machine and an adaptive sampling scheme. The approximation is sequentially refined using a new adaptive sampling scheme referred to as generalized "max-min". This technique efficiently targets high probability density regions of the random space. This is achieved by modifying an adaptive sampling scheme originally tailored for deterministic spaces (Explicit Space Design Decomposition). In particular, the approach can handle any joint probability density function, even if the variables are dependent. In the latter case, the joint distribution might be obtained from copula. In addition, uncertainty on the probability of failure estimate are estimated using bootstrapping. A bootstrapped coefficient of variation of the probability of failure is used as an estimate of the true error to determine convergence. The proposed method is then applied to analytical examples and a beam bending reliability assessment using copulas.
UR - http://www.scopus.com/inward/record.url?scp=84961323878&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84961323878&partnerID=8YFLogxK
U2 - 10.1115/DETC201434051
DO - 10.1115/DETC201434051
M3 - Conference contribution
AN - SCOPUS:84961323878
SN - 9780791846285
VL - 1A
BT - 34th Computers and Information in Engineering Conference
PB - American Society of Mechanical Engineers (ASME)
ER -