Neural Networks for Damage Detection
Herbert M. Gomes, UFRGS, Porto Alegre, RS, Brasil
ABSTRACT Nowadays, the advances in Neural Networks have opened up windows for Structural Engineering. The adaptivity, robustness and capability in dealing with incomplete data make the Neural Networks the best alternative for this purpose. Associated to vibrational signature analysis, this technique has showed a robust behavior. The detection of structural damage and identification of the damaged element in a large complex structure is a hard task. It is well known that from variations in structural natural frequency measure in situ and a well-calibrated structural model, it is possible to detect as in position as in intensity damaged states. Some shortcoming in this procedure have been reported regarded the position evaluation of damaged states in symmetric structures. Some new advances have been made in this area by means of Neural Networks.
Publication Source: NDTISS '99 - International Symposium on NDT Contribution to the Infrastructure Safety Systems, Nov 1999 in Tores Brazil.
Publisher: Center of Tecnology, Federal University of Santa Maria (UFSM) - [Homepage]
© NDTnet
