Artificial Neural Networks and quadratic Response Surfaces for the functional failure analysis of a thermal-hydraulic passive system

George Apostolakis (1) , Nicola Pedroni (2) , Enrico Zio (2, 3, 4)

1 Department of Nuclear Science and Engineering
2 Dipartimento di Energia [Milano]
3 SSEC - Chaire Sciences des Systèmes et Défis Energétiques EDF/ECP/Supélec
4 LGI - Laboratoire Génie Industriel - EA 2606

In this paper, bootstrapped Artificial Neural Network (ANN) and quadratic Response Surface (RS) empirical regression models are used as fast-running surrogates of a thermal-hydraulic (T-H) system code to reduce the computational burden associated with the estimation of the functional failure probability of a T-H passive system. The ANN and quadratic RS models are built on few data representative of the input/output nonlinear relationships underlying the T-H code. Once built, these models are used for performing, in reasonable computational time, the numerous system response calculations required for failure probability estimation. A bootstrap of the regression models is implemented for quantifying, in terms of confidence intervals, the uncertainties associated with the estimates provided by ANNs and RSs. The alternative empirical models are compared on a case

Mots clés: epistemic uncertainties; passive system reliability; artificial neural network; quadratic response surface.

Mots clés: quadratic response surface

Domaines: Autre

George Apostolakis, Nicola Pedroni, Enrico Zio. Artificial Neural Networks and quadratic Response Surfaces for the functional failure analysis of a thermal-hydraulic passive system. Proceedings of the 10th International PSAM Conference, 2010, pp.1 - 12. ⟨hal-00721042⟩