A Hierarchical Neural Hybrid Method for Failure Probability Estimation
Open Access
- 15 August 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 112087-112096
Abstract
Failure probability evaluation for complex physical and engineering systems governed by partial differential equations (PDEs) are computationally intensive, especially when high-dimensional random parameters are involved. Since standard numerical schemes for solving these complex PDEs are expensive, traditional Monte Carlo methods which require repeatedly solving PDEs are infeasible. Alternative approaches which are typically the surrogate based methods suffer from the so-called “curse of dimensionality”, which limits their application to problems with high-dimensional parameters. For this purpose, we develop a novel hierarchical neural hybrid (HNH) method to efficiently compute failure probabilities of these challenging high-dimensional problems. Especially, multifidelity surrogates are constructed based on neural networks with different levels of layers, such that expensive highfidelity surrogates are adapted only when the parameters are in the suspicious domain. The efficiency of our new HNH method is theoretically analyzed and is demonstrated with numerical experiments. From numerical results, we show that to achieve an accuracy in estimating the rare failure probability (e.g., $10^{-5}$ ), the traditional Monte Carlo method needs to solve PDEs more than a million times, while our HNH only requires solving them a few thousand times.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (11601329, 11771289)
- Science Challenge Project (TZ2018001)
This publication has 28 references indexed in Scilit:
- Evaluation of failure probability via surrogate modelsJournal of Computational Physics, 2010
- Block-diagonal preconditioning for spectral stochastic finite-element systemsIMA Journal of Numerical Analysis, 2008
- Multiple design points in first and second-order reliabilityStructural Safety, 1998
- A benchmark study on importance sampling techniques in structural reliabilityStructural Safety, 1993
- A new look at the response surface approach for reliability analysisStructural Safety, 1993
- A fast and efficient response surface approach for structural reliability problemsStructural Safety, 1990
- Importance sampling in structural systemsStructural Safety, 1989
- New light on first- and second-order reliability methodsStructural Safety, 1987
- Learning representations by back-propagating errorsNature, 1986
- The Monte Carlo MethodJournal of the American Statistical Association, 1949