.png){kind=icon}
.png){kind=icon}
Return
Random-interval hybrid reliability analysis based on parallel adaptive Kriging and sequential importance sampling
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compstruc.2026.108370.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
At present, the computational efficiency of random-interval hybrid reliability analysis methods remains insufficient. To address this issue, this paper proposes a parallel learning method tailored for small failure probability problems in random-interval hybrid reliability analysis. Specifically, sequential importance sampling, originally developed for random reliability analysis, is extended to the random-interval hybrid reliability analysis framework to generate importance sampling random samples corresponding to the upper and lower bounds of failure probability. These importance sampling samples are then employed to evaluate the bounds of failure probability. Furthermore, a novel parallel learning function that integrates the Kriging believer strategy with K-means clustering is proposed. By efficiently combining sequential importance sampling, the proposed parallel learning function, and Kriging, an efficient method for random-interval hybrid reliability analysis is further proposed. The accuracy and efficiency of the proposed method are validated through five examples, and the results demonstrate that the proposed method not only maintains high accuracy but also significantly outperforms existing methods in computational efficiency, effectively reducing the number of iterations required for convergence.
Keywords:
Random-interval hybrid reliability analysis
Parallel learning function
Sequential importance sampling
Adaptive Kriging
Small failure probability
Journal
C
IF:
4.8
Papers:
216
Citations:
0
Organization
No organization information available