Damage detection of truss bridge joints using Artificial Neural Networks

Recent developments in Artificial Neural Networks (ANNs) have opened up new possibilities in the domain of inverse problems. For inverse problems like structural identification of large structures (such as bridges) where in situ measured data are expected to be imprecise and often incomplete, ANNs may hold greater promise. This study presents a method for estimating the damage intensities of joints for truss bridge structures using a back-propagation based neural network. The technique that was employed to overcome the issues associated with many unknown parameters in a large structural system is the substructural identification. The natural frequencies and mode shapes were used as input parameters to the neural network for damage identification, particularly for the case with incomplete measurements of the mode shapes. Numerical example analyses on truss bridges are presented to demonstrate the accuracy and efficiency of the proposed method.