A hybrid methodology for structural damage detection uniting FEM and 1D-CNNs: Demonstration on typical high-pile wharf

•This study establishes a method framework based on FEM and DL to realize the identification of overall structural damage of high-pile wharfs.•The datasets required for the training of the CNNs can be directly reconstructed by the displacement response of FE model.•The comprehensive performance of the CNNs based on the displacement response dataset in component form is significantly better.•The results show that the CNN are more sensitive to the longitudinal and lateral displacement responses at the beam-pile nodes.