Loading [MathJax]/extensions/tex2jax.js

Error-informed parallel adaptive Kriging method for time-dependent reliability analysis

Research output: Contribution to journalArticleResearchpeer review

Authors

Research Organisations

External Research Organisations

  • Central South University of Forestry & Technology
  • TU Dortmund University
  • University of Liverpool
  • Tongji University
  • Changsha University of Science and Technology

Details

Original languageEnglish
Article number111194
Number of pages12
JournalReliability Engineering and System Safety
Volume262
Early online date9 May 2025
Publication statusE-pub ahead of print - 9 May 2025

Abstract

Active learning single-loop Kriging methods have gained significant attention for time-dependent reliability analysis. However, it still remains a challenge to estimate the time-dependent failure probability efficiently and accurately in practical engineering problems. This study proposes a new method, called ‘Error-informed Parallel Adaptive Kriging’ (EPAK) for efficient time-dependent reliability analysis. First, a sequential variance-amplified importance sampling technique is developed to estimate the time-dependent failure probability based on the trained global response Kriging model of the true performance function. Then, the maximum relative error of the time-dependent failure probability is derived to facilitate the construction of stopping criterion. Finally, a parallel sampling strategy is proposed through combining the relative error contribution and an influence function, which enables parallel computing and reduces the unnecessary limit state function evaluations caused by excessive clustering. The superior performance of the proposed method is validated through several examples. Numerical results demonstrate that the method can accurately estimate the time-dependent failure probability with higher efficiency than several compared methods.

Keywords

    Active learning, Estimation error, Importance sampling, Kriging model, Parallel computing, Time-dependent reliability analysis

ASJC Scopus subject areas

Cite this

Error-informed parallel adaptive Kriging method for time-dependent reliability analysis. / Hu, Zhuo; Dang, Chao; Wang, Da et al.
In: Reliability Engineering and System Safety, Vol. 262, 111194, 10.2025.

Research output: Contribution to journalArticleResearchpeer review

Hu Z, Dang C, Wang D, Beer M, Wang L. Error-informed parallel adaptive Kriging method for time-dependent reliability analysis. Reliability Engineering and System Safety. 2025 Oct;262:111194. Epub 2025 May 9. doi: 10.1016/j.ress.2025.111194
Download
@article{5b2f842e5652445fa2b2f6b42936fec8,
title = "Error-informed parallel adaptive Kriging method for time-dependent reliability analysis",
abstract = "Active learning single-loop Kriging methods have gained significant attention for time-dependent reliability analysis. However, it still remains a challenge to estimate the time-dependent failure probability efficiently and accurately in practical engineering problems. This study proposes a new method, called {\textquoteleft}Error-informed Parallel Adaptive Kriging{\textquoteright} (EPAK) for efficient time-dependent reliability analysis. First, a sequential variance-amplified importance sampling technique is developed to estimate the time-dependent failure probability based on the trained global response Kriging model of the true performance function. Then, the maximum relative error of the time-dependent failure probability is derived to facilitate the construction of stopping criterion. Finally, a parallel sampling strategy is proposed through combining the relative error contribution and an influence function, which enables parallel computing and reduces the unnecessary limit state function evaluations caused by excessive clustering. The superior performance of the proposed method is validated through several examples. Numerical results demonstrate that the method can accurately estimate the time-dependent failure probability with higher efficiency than several compared methods.",
keywords = "Active learning, Estimation error, Importance sampling, Kriging model, Parallel computing, Time-dependent reliability analysis",
author = "Zhuo Hu and Chao Dang and Da Wang and Michael Beer and Lei Wang",
note = "Publisher Copyright: {\textcopyright} 2025 Elsevier Ltd",
year = "2025",
month = may,
day = "9",
doi = "10.1016/j.ress.2025.111194",
language = "English",
volume = "262",
journal = "Reliability Engineering and System Safety",
issn = "0951-8320",
publisher = "Elsevier Ltd.",

}

Download

TY - JOUR

T1 - Error-informed parallel adaptive Kriging method for time-dependent reliability analysis

AU - Hu, Zhuo

AU - Dang, Chao

AU - Wang, Da

AU - Beer, Michael

AU - Wang, Lei

N1 - Publisher Copyright: © 2025 Elsevier Ltd

PY - 2025/5/9

Y1 - 2025/5/9

N2 - Active learning single-loop Kriging methods have gained significant attention for time-dependent reliability analysis. However, it still remains a challenge to estimate the time-dependent failure probability efficiently and accurately in practical engineering problems. This study proposes a new method, called ‘Error-informed Parallel Adaptive Kriging’ (EPAK) for efficient time-dependent reliability analysis. First, a sequential variance-amplified importance sampling technique is developed to estimate the time-dependent failure probability based on the trained global response Kriging model of the true performance function. Then, the maximum relative error of the time-dependent failure probability is derived to facilitate the construction of stopping criterion. Finally, a parallel sampling strategy is proposed through combining the relative error contribution and an influence function, which enables parallel computing and reduces the unnecessary limit state function evaluations caused by excessive clustering. The superior performance of the proposed method is validated through several examples. Numerical results demonstrate that the method can accurately estimate the time-dependent failure probability with higher efficiency than several compared methods.

AB - Active learning single-loop Kriging methods have gained significant attention for time-dependent reliability analysis. However, it still remains a challenge to estimate the time-dependent failure probability efficiently and accurately in practical engineering problems. This study proposes a new method, called ‘Error-informed Parallel Adaptive Kriging’ (EPAK) for efficient time-dependent reliability analysis. First, a sequential variance-amplified importance sampling technique is developed to estimate the time-dependent failure probability based on the trained global response Kriging model of the true performance function. Then, the maximum relative error of the time-dependent failure probability is derived to facilitate the construction of stopping criterion. Finally, a parallel sampling strategy is proposed through combining the relative error contribution and an influence function, which enables parallel computing and reduces the unnecessary limit state function evaluations caused by excessive clustering. The superior performance of the proposed method is validated through several examples. Numerical results demonstrate that the method can accurately estimate the time-dependent failure probability with higher efficiency than several compared methods.

KW - Active learning

KW - Estimation error

KW - Importance sampling

KW - Kriging model

KW - Parallel computing

KW - Time-dependent reliability analysis

UR - http://www.scopus.com/inward/record.url?scp=105004798807&partnerID=8YFLogxK

U2 - 10.1016/j.ress.2025.111194

DO - 10.1016/j.ress.2025.111194

M3 - Article

AN - SCOPUS:105004798807

VL - 262

JO - Reliability Engineering and System Safety

JF - Reliability Engineering and System Safety

SN - 0951-8320

M1 - 111194

ER -

By the same author(s)