Use of Artificial Neural Networks and Response Surface Methodology for Evaluating the Reliability Index of Steel Wind Towers

The estimation of structural reliability is a process that requires large number computational hours when statistical data are not available since it necessary to perform amount analysis or numerical simulations estimate parameters related the reliability. A methodology proposed for estimating index, as well demand and capacity factors inherent structure, given fundamental vibration period height by using artificial neural networks (ANN) and, alternatively, response surface method (RSM). Both approaches applied steel wind turbine towers. For cases studied, ANN allow evaluating index both with greater accuracy than RSM.