Abstract:When the neural network is used to identify the damage of complex structure, a problem that the network size is too large to converge will be encountered. So, a two-step structural damage identification approach based on the curvature of mode shape difference and BP neural network is proposed in this paper. Firstly, the damage areas are localized approximately by the curvature of mode shape difference. Secondly, the damaged elements are determined accurately by the BP neural network in the selected areas. The effectiveness of the method is validated by the results of numerical simulation and experiment for a four-floor offshore platform.