Sequential Bayesian optimal experimental design for structural reliability analysis

DOI10.1007/S11222-021-10000-2zbMATH Open1475.62013arXiv2007.00402OpenAlexW3038622104MaRDI QIDQ2058731FDOQ2058731

Christian Agrell, Kristina Rognlien Dahl

Publication date: 9 December 2021

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: Structural reliability analysis is concerned with estimation of the probability of a critical event taking place, described by $P(g(extbfX)leq0)$ for some $n$-dimensional random variable $extbfX$ and some real-valued function $g$. In many applications the function $g$ is practically unknown, as function evaluation involves time consuming numerical simulation or some other form of experiment that is expensive to perform. The problem we address in this paper is how to optimally design experiments, in a Bayesian decision theoretic fashion, when the goal is to estimate the probability $P(g(extbfX)leq0)$ using a minimal amount of resources. As opposed to existing methods that have been proposed for this purpose, we consider a general structural reliability model given in hierarchical form. We therefore introduce a general formulation of the experimental design problem, where we distinguish between the uncertainty related to the random variable $extbfX$ and any additional epistemic uncertainty that we want to reduce through experimentation. The effectiveness of a design strategy is evaluated through a measure of residual uncertainty, and efficient approximation of this quantity is crucial if we want to apply algorithms that search for an optimal strategy. The method we propose is based on importance sampling combined with the unscented transform for epistemic uncertainty propagation. We implement this for the myopic (one-step look ahead) alternative, and demonstrate the effectiveness through a series of numerical experiments.

Full work available at URL: https://arxiv.org/abs/2007.00402

Recommendations

• Efficient reliability analysis based on adaptive sequential sampling design and cross-validation
• scientific article
• Sequential Bayesian experimental design for estimation of extreme-event probability in stochastic input-to-response systems
• Sequential design of computer experiments for the estimation of a probability of failure
• A combined radial basis function and adaptive sequential sampling method for structural reliability analysis

zbMATH Keywords

optimal experimental designstructural reliabilityprobability of failureepistemic and aleatory uncertaintyunscented transform

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Reliability and life testing (62N05) Optimal statistical designs (62K05) Sequential statistical design (62L05)

Cites Work

• Bayesian calibration of computer models. (With discussion)
• Gaussian processes for machine learning.
• Sequential design of computer experiments for the estimation of a probability of failure
• A supermartingale approach to Gaussian process based sequential design of experiments
• Bayes Linear Statistics
• Optimal control of systems with noisy memory and BSDEs with Malliavin derivatives
• A Systematization of the Unscented Kalman Filter Theory
• A robust iterative algorithm for structural reliability analysis
• Coupling Computer Models through Linking Their Statistical Emulators
• Bayes linear analysis of risks in sequential optimal design problems

Cited In (3)

• Novel learning functions design based on the probability of improvement criterion and normalization techniques
• Sequential experiment design for parameter estimation of nonlinear systems using a neural network approximator
• Parallel active learning reliability analysis: a multi-point look-ahead paradigm

This page was built for publication: Sequential Bayesian optimal experimental design for structural reliability analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2058731)