.png){kind=icon}
.png){kind=icon}
Return
An efficient time-variant reliability analysis method based on sparse grid and improved saddlepoint approximation
.png){kind=icon}
PRE
AI
H
W
S
H
T
H
DOI：10.1016/j.probengmech.2026.103981.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To address the issues of low computational efficiency and high computational cost in time-variant reliability analysis for practical engineering applications, a time-variant reliability analysis method based on sparse grid integration and an improved saddlepoint approximation is proposed. First, the input stochastic process is discretized into a linear combination of multiple random variables using the Expansion Optimal Linear Estimation (EOLE) method. On this basis, the Efficient Global Optimization (EGO) algorithm is employed to transform the system's time-variant reliability problem into a time-invariant problem in which only the extreme value response needs to be considered. Subsequently, sparse grid numerical integration is applied to compute the first four statistical moments of the extreme value distribution of the time-variant performance function. Finally, an improved saddlepoint approximation method is used to estimate the extreme value distribution of the response, from which the time-variant structural reliability is determined. The effectiveness of the proposed method is validated through one numerical example and three practical engineering case studies. The results demonstrate that the proposed method simplifies the entire time-variant reliability analysis process and significantly improves computational efficiency while maintaining high accuracy, by converting the complex time-variant problem into a conventional time-invariant problem.
Journal
IF:
3.5
Papers:
1.7K
Citations:
4.1K
Organization
Cited Papers