Structural reliability assessment with the aid of neural networks

Abstract eng:
A methodology based on Neural Networks is presented for the quick vulnerability assessment of steel moment resisting frames. The proposed method combines data extracted from ground motion records with a small number of nonlinear response history analyses to predict the seismic demand on a ten-storey moment-resisting steel frame. Having such a tool at our disposal we use Monte Carlo Simulation to calculate the conditional limit-state probabilities of the vulnerability curves. The framework we present allows taking into account uncertainties on both structural capacity and seismic demand with reduced computational cost. The use of Neural Networks is motivated by the approximate concepts inherent in the fragility assessment and the large number of time-consuming nonlinear response history analyses required for the accurate calculation of the probability of a limit-state being exceeded. The trained Neural Networks is used to obtain the level of seismic demand, which is expressed in terms of maximum interstorey drift.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket