Deep neural network for structural damage detection in bridges

ABSTRACT A number of artificial intelligence techniques have been implemented to detect the damage in civil structure to partially replace onsite inspections. These artificial intelligence methods are primarily used to manipulate dataset and extract defect features, such as: the damage that occurs in concrete and steel bridges. This paper proposes a vibration-based method using a deep architecture of back propagation neural networks (BPNNs) for detecting damage. BPNNs can learn the features of images and signals. The robustness and adaptability of the proposed approach are tested on two different structures. The results show that the proposed method has better performances than existing ones.