Gaussian process surrogates for failure detection: A Bayesian experimental design approach
Research output: Contribution to journal › Article › peer-review
Colleges, School and Institutes
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.
|Journal||Journal of Computational Physics|
|Early online date||26 Feb 2016|
|Publication status||Published - 15 May 2016|
- Bayesian inference, experimental design, failure detection, Gaussian processes, Monte Carlo, response surfaces, uncertainty quantification