Highly efficient Bayesian updating using metamodels: An adaptive Kriging-based approach
Highly efficient Bayesian updating using metamodels: An adaptive Kriging-based approach
复制标题
使用元模型的高效贝叶斯更新：基于自适应克里金法的方法
DOI：
10.1016/j.strusafe.2019.101915
复制
发表时间：
2020
影响因子：
5.8
通讯作者：
Shafieezadeh, Abdollah
中科院分区：
文献类型：
--
作者：
Wang, Zeyu;Shafieezadeh, Abdollah
贝叶斯更新提供了一个强大的工具，概率校准和不确定性量化的模型，新的观察变得可用。通过引入辅助随机变量，将贝叶斯更新转化为结构可靠性问题，与基于马尔可夫链蒙特卡罗模拟的传统方法相比，贝叶斯结构可靠性更新方法（BUS）具有更高的精度和效率。然而，总线面临着一些限制。转换可靠性问题往往涉及到一个非常罕见的事件，特别是当观察的数量增加。这沿着的事实是，传统的可靠性分析技术是不有效的，往往不能准确地估计概率的罕见事件，必然导致非常大量的评估的似然函数，同时不足的准确性推导出的后验分布。为了克服这些局限性，我们提出了多辅助随机变量简单拒绝抽样（SRS-MARV），其中的极限状态函数在BUS分解成一个系统的可靠性问题与多个极限状态函数。该方法的主要优点是，各分解极限状态函数的接受率得到了显著提高，有利于将自适应Kriging可靠性分析有效地集成到SRS-MARV中。此外，提出了一种新的停止标准，有效的，自适应训练的克里格模型。所提出的方法calledBUAK被证明是高度的计算效率和准确的基础上的全面调查的结果为三个不同的基准问题。与最先进的方法相比，BUAK大大减少了一到三个数量级的计算需求，因此，有利于贝叶斯更新的计算非常密集的模型的应用。
Bayesian updating offers a powerful tool for probabilistic calibration and uncertainty quantification of models as new observations become available. By reformulating Bayesian updating into a structural reliability problem via introducing an auxiliary random variable, the state-of-the-art Bayesian updating with structural reliability method (BUS)has showcased large potential to achieve higher accuracy and efficiency compared with conventional approaches based on Markov Chain Monte Carlo simulations. However,BUSfaces a number of limitations. The transformed reliability problem often involves a very rare event especially when the number of observations increases. This along with the fact that conventional reliability analysis techniques are not efficient, and often not capable of accurately estimating the probability of rare events, unavoidably lead to a very large number of evaluations of the likelihood function and simultaneously insufficient accuracy of the derived posterior distributions. To overcome these limitations, we propose Simple Rejection Sampling with Multiple Auxiliary Random Variables (SRS-MARV), where the limit state function inBUSis decomposed into a system reliability problem with multiple limit state functions. The main advantage of this approach is that the acceptance rate of each decomposed limit state function is significantly improved, which facilitates effective integration of adaptive Kriging-based reliability analysis intoSRS-MARV. Moreover, a new stopping criterion is proposed for efficient, adaptive training of the Kriging model. The proposed method calledBUAKis shown to be highly computationally efficient and accurate based on results of comprehensive investigations for three diverse benchmark problems. Compared to the state-of-the-art methods,BUAKsubstantially reduces the computational demand by one to three orders of magnitude, therefore, facilitating the application of Bayesian updating to computationally very intensive models.
登录
查看更多内容
在不知道似然函数最大值的情况下使用可靠性方法进行贝叶斯推理
影响因子：
2.6
作者：
W. Betz;J. Beck;I. Papaioannou;D. Štraub
通讯作者：
D. Štraub
影响因子：
3.9
作者：
Zeyu Wang;A. Shafieezadeh
通讯作者：
Zeyu Wang;A. Shafieezadeh
影响因子：
7.4
作者：
M. Rahimi;A. Shafieezadeh;Dylan Wood;E. Kubatko;N. Dormady
通讯作者：
M. Rahimi;A. Shafieezadeh;Dylan Wood;E. Kubatko;N. Dormady
DOI：
10.1016/j.ress.2019.106758
发表时间：
2020-04
期刊：
Reliab. Eng. Syst. Saf.
影响因子：
--
作者：
Zeyu Wang;A. Shafieezadeh
通讯作者：
Zeyu Wang;A. Shafieezadeh
DOI：
10.1016/j.cma.2017.02.025
发表时间：
2017-06-01
影响因子：
7.2
作者：
Giovanis, Dimitris G.;Papaioannou, Iason;Papadopoulos, Vissarion
通讯作者：
Papadopoulos, Vissarion