Directional filter combined with active learning for rare failure events

Research output: Contribution to journalArticleResearchpeer review

Authors

  • Jingwen Song
  • Yifan Cui
  • Pengfei Wei
  • Mohsen Rashki
  • Weihong Zhang
  • Michael Beer

Research Organisations

External Research Organisations

  • Northwestern Polytechnical University
  • University of Sistan and Baluchistan
  • University of Liverpool
  • Tongji University
View graph of relations

Details

Original languageEnglish
Article number117105
Number of pages19
JournalComputer Methods in Applied Mechanics and Engineering
Volume428
Early online date7 Jun 2024
Publication statusE-pub ahead of print - 7 Jun 2024

Abstract

Estimating the small failure probability is a crucial task in structural engineering, and directional sampling has long been recognized as one of the most promising stochastic simulation method for problems with multiple disconnected failure domains. However, when applied to problems with expensive-to-evaluate and highly nonlinear limit state functions, it is still less satisfactory in terms of numerical accuracy and efficiency. To fill this gap, this paper develops a directional filter equipped active learning (DirFAL) algorithm. It integrates three key components: (1) an active learning directional filter scheme, which plays as a role for adaptively prioritizing the directional samples that dominantly contribute to the failure probability; (2) a tailored closed-form expression of acquisition function, which is newly defined and incorporated into DirFAL to effectively select new training data for enriching a Gaussian process regression surrogate model; (3) an active learning stratification strategy for tackling the performance degradation issue when applied to problems with relatively high dimension and extremely small failure probability. The performance of the proposed methods is ultimately demonstrated through two numerical examples and two engineering problems, and results show that they are of superiority over other parallel methods in terms of numerical efficiency given required accuracy.

Keywords

    Active learning, Directional filter, Gaussian process regression, Most probable point, Rare failure event

ASJC Scopus subject areas

Cite this

Directional filter combined with active learning for rare failure events. / Song, Jingwen; Cui, Yifan; Wei, Pengfei et al.
In: Computer Methods in Applied Mechanics and Engineering, Vol. 428, 117105, 01.08.2024.

Research output: Contribution to journalArticleResearchpeer review

Song J, Cui Y, Wei P, Rashki M, Zhang W, Beer M. Directional filter combined with active learning for rare failure events. Computer Methods in Applied Mechanics and Engineering. 2024 Aug 1;428:117105. Epub 2024 Jun 7. doi: 10.1016/j.cma.2024.117105
Download
@article{0d313caed6f1434f89872274f2f5bd52,
title = "Directional filter combined with active learning for rare failure events",
abstract = "Estimating the small failure probability is a crucial task in structural engineering, and directional sampling has long been recognized as one of the most promising stochastic simulation method for problems with multiple disconnected failure domains. However, when applied to problems with expensive-to-evaluate and highly nonlinear limit state functions, it is still less satisfactory in terms of numerical accuracy and efficiency. To fill this gap, this paper develops a directional filter equipped active learning (DirFAL) algorithm. It integrates three key components: (1) an active learning directional filter scheme, which plays as a role for adaptively prioritizing the directional samples that dominantly contribute to the failure probability; (2) a tailored closed-form expression of acquisition function, which is newly defined and incorporated into DirFAL to effectively select new training data for enriching a Gaussian process regression surrogate model; (3) an active learning stratification strategy for tackling the performance degradation issue when applied to problems with relatively high dimension and extremely small failure probability. The performance of the proposed methods is ultimately demonstrated through two numerical examples and two engineering problems, and results show that they are of superiority over other parallel methods in terms of numerical efficiency given required accuracy.",
keywords = "Active learning, Directional filter, Gaussian process regression, Most probable point, Rare failure event",
author = "Jingwen Song and Yifan Cui and Pengfei Wei and Mohsen Rashki and Weihong Zhang and Michael Beer",
note = "Publisher Copyright: {\textcopyright} 2024 Elsevier B.V.",
year = "2024",
month = jun,
day = "7",
doi = "10.1016/j.cma.2024.117105",
language = "English",
volume = "428",
journal = "Computer Methods in Applied Mechanics and Engineering",
issn = "0045-7825",
publisher = "Elsevier",

}

Download

TY - JOUR

T1 - Directional filter combined with active learning for rare failure events

AU - Song, Jingwen

AU - Cui, Yifan

AU - Wei, Pengfei

AU - Rashki, Mohsen

AU - Zhang, Weihong

AU - Beer, Michael

N1 - Publisher Copyright: © 2024 Elsevier B.V.

PY - 2024/6/7

Y1 - 2024/6/7

N2 - Estimating the small failure probability is a crucial task in structural engineering, and directional sampling has long been recognized as one of the most promising stochastic simulation method for problems with multiple disconnected failure domains. However, when applied to problems with expensive-to-evaluate and highly nonlinear limit state functions, it is still less satisfactory in terms of numerical accuracy and efficiency. To fill this gap, this paper develops a directional filter equipped active learning (DirFAL) algorithm. It integrates three key components: (1) an active learning directional filter scheme, which plays as a role for adaptively prioritizing the directional samples that dominantly contribute to the failure probability; (2) a tailored closed-form expression of acquisition function, which is newly defined and incorporated into DirFAL to effectively select new training data for enriching a Gaussian process regression surrogate model; (3) an active learning stratification strategy for tackling the performance degradation issue when applied to problems with relatively high dimension and extremely small failure probability. The performance of the proposed methods is ultimately demonstrated through two numerical examples and two engineering problems, and results show that they are of superiority over other parallel methods in terms of numerical efficiency given required accuracy.

AB - Estimating the small failure probability is a crucial task in structural engineering, and directional sampling has long been recognized as one of the most promising stochastic simulation method for problems with multiple disconnected failure domains. However, when applied to problems with expensive-to-evaluate and highly nonlinear limit state functions, it is still less satisfactory in terms of numerical accuracy and efficiency. To fill this gap, this paper develops a directional filter equipped active learning (DirFAL) algorithm. It integrates three key components: (1) an active learning directional filter scheme, which plays as a role for adaptively prioritizing the directional samples that dominantly contribute to the failure probability; (2) a tailored closed-form expression of acquisition function, which is newly defined and incorporated into DirFAL to effectively select new training data for enriching a Gaussian process regression surrogate model; (3) an active learning stratification strategy for tackling the performance degradation issue when applied to problems with relatively high dimension and extremely small failure probability. The performance of the proposed methods is ultimately demonstrated through two numerical examples and two engineering problems, and results show that they are of superiority over other parallel methods in terms of numerical efficiency given required accuracy.

KW - Active learning

KW - Directional filter

KW - Gaussian process regression

KW - Most probable point

KW - Rare failure event

UR - http://www.scopus.com/inward/record.url?scp=85195212268&partnerID=8YFLogxK

U2 - 10.1016/j.cma.2024.117105

DO - 10.1016/j.cma.2024.117105

M3 - Article

AN - SCOPUS:85195212268

VL - 428

JO - Computer Methods in Applied Mechanics and Engineering

JF - Computer Methods in Applied Mechanics and Engineering

SN - 0045-7825

M1 - 117105

ER -

By the same author(s)