.png){kind=icon}
.png){kind=icon}
Return
A dimensional reduction integration based stochastic collocation algorithm for estimating local reliability sensitivity
.png){kind=icon}
PRE
AI
DOI：10.1016/j.istruc.2025.110655.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Local reliability sensitivity (LRS) serves to quantify the influence exerted by the distribution parameter of random input vector on the failure probability, and it can provide directions of searching the optimal designable distribution parameter in solving structural reliability-based design optimization model. However, estimating LRS still poses a challenge in efficiency for implicit performance functions. To address this challenge, this paper derives a dimensional reduction integration (DRI) format for LRS, and leverages the continuity of the integrand in the derived DRI format to design a stochastic collocation algorithm for efficiently estimating LRS estimation. Firstly, the proposed algorithm improves the behavior of integrand involved in integration of estimating LRS by DRI. Secondly, the proposed algorithm employs an information sharing strategy to enable a single set of stochastic collocation nodes to estimate LRS for all distribution parameters of all inputs, and this strategy eliminates the computational dependency on the dimensions of inputs and their distribution parameters, significantly improving efficiency. The pronounced efficiency advantage of the proposed method is sufficiently validated by means of numerical examples.
Keywords:
Structural reliability
Local reliability sensitivity
Dimensional reduction integration
Stochastic collocation algorithm
Journal
IF:
4.3
Papers:
1.2W
Citations:
2.7W