Bayesian neural networks for damage identification of a cable-stayed bridge

In recent years there has been a growing interest on the application of soft computing methods for processing the large quantity of data coming from long term structural health monitoring. In particular, this work deals with the applicability of Bayesian neural networks for damage identification of a cable-stayed bridge. Bayesian neural networks come from the optimization of the neural networks model by using Bayesian inference at four hierarchical levels (Arangio and Beck, 2010). The selected structure is a real bridge proposed as benchmark problem by the Asian-Pacific Network of Centers for Research in Smart Structure Technology (ANCRiSST). They shared data coming from the long term monitoring of the bridge with the Structural Health Monitoring community in order to assess the current progress on damage detection and identification methods with a full scale example. The dataset includes vibration data before and after the bridge was damaged. The available data have been used to test a Bayesian neural networks-based damage detection strategy and this paper summarizes the preliminary results of the analyses. The proposed method is able to detect anomalies on the behavior of the structure, which can be related to the presence of damage. At the moment it is not possible to identify clearly the damaged zone but current studies are finalized at the optimization of the strategy in this sense.