Gaussian process surrogates for failure detection: a Bayesian experimental design approach
For engineers and scientists dealing with expensive computer models, this provides a more efficient method for failure detection and probability estimation.
This work develops a Bayesian experimental design approach for constructing Gaussian process surrogates to detect system failures and estimate failure probabilities, enabling efficient selection of multiple sampling points for parallel simulation. The method demonstrates improved accuracy and performance on academic and practical examples.
An important task of uncertainty quantification is to identify {the probability of} undesired events, in particular, system failures, caused by various sources of uncertainties. In this work we consider the construction of Gaussian {process} surrogates for failure detection and failure probability estimation. In particular, we consider the situation that the underlying computer models are extremely expensive, and in this setting, determining the sampling points in the state space is of essential importance. We formulate the problem as an optimal experimental design for Bayesian inferences of the limit state (i.e., the failure boundary) and propose an efficient numerical scheme to solve the resulting optimization problem. In particular, the proposed limit-state inference method is capable of determining multiple sampling points at a time, and thus it is well suited for problems where multiple computer simulations can be performed in parallel. The accuracy and performance of the proposed method is demonstrated by both academic and practical examples.