Engineering Mechanics

Proceedings Vol. 21 (2015)

ENGINEERING MECHANICS 2015

Copyright © 2015 Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, v.v.i., Prague

 

D. Lehký, M. Šomodíková
pages 182 - 183, full text

The key step of the reliability and lifetime assessment of structures is the determination of reliability level, described by failure probability or reliability index. Some of the simulation or approximation techniques can be used for this purpose. In case of large structures analyzed using the nonlinear finite element method, it is necessary to develop more efficient procedures, reducing the number of evaluations of original limit state function to a minimum. Here, artificial neural network based response surface method in combination with small-sample simulation technique Latin Hypercube Sampling is utilized for the approximation of a limit state function. Thanks to ability of artificial neural network to generalize it is efficient to fit limit state function with a sufficiently small number of simulations.

Text and facts may be copied and used freely, but credit should be given to these Proceedings.

All papers were reviewed by members of the scientific committee.